A plain-English dictionary of the terms founders, operators, investors, and builders throw around, from SAFEs and board consents to APIs, AI agents, RAG, and beyond.
An organization that supports very early startups with resources such as workspace, mentorship, business guidance, and founder community. Incubators are often less structured than accelerators and may help founders while they are still shaping the idea.
A program for early founders who are still validating an idea, testing customer demand, or preparing to apply to an accelerator.
A fixed-term startup program that helps companies grow faster through mentorship, education, investor introductions, and structured milestones. Many accelerators end with a demo day or investor-facing presentation.
Building a company using personal savings, customer revenue, consulting income, or internal resources instead of outside funding.
An early round raised from people in the founder's personal network, usually before professional investors are involved.
The earliest formal funding round, usually used to validate the idea, build an MVP, test demand, and form the founding team.
A financing round backed mainly by angel investors or angel groups, often before or alongside a pre-seed or seed round.
An early financing round used to support product development, hiring, customer acquisition, market validation, and early revenue growth.
Additional funding raised after an initial seed round, often to give the company more time to hit Series A milestones.
The first major institutional venture round, usually raised after a startup has meaningful traction, signs of product-market fit, or strong growth potential.
Short-term funding used to extend runway until the next major financing round or milestone.
A financing round completed at a higher valuation than the company's previous round.
A financing round completed at a lower valuation than the previous round. Down rounds can increase dilution and may trigger investor protections.
An individual who invests personal money into early-stage startups, often before institutional VC funds participate. Angels are often former founders, operators, executives, or high-net-worth individuals.
A network of angel investors that reviews startup opportunities together. Members may invest individually, collectively, or through a shared vehicle.
A pooled investment vehicle backed by angel investors that invests into startups as a fund.
A group of investors that participates in the same startup financing, often organized by a lead angel, fund, or platform.
Capital provided by professional investment firms to startups with high growth potential, usually in exchange for equity or preferred shares.
A professional investor who works at a VC firm and evaluates startups, negotiates investments, and supports portfolio companies.
The investor who helps set the terms of a round and often contributes a meaningful portion of the financing.
An investor who joins a round alongside the lead investor, usually on the same or similar terms.
An investor that may provide strategic value beyond capital, such as industry expertise, partnerships, distribution, or customer access.
A private investment group that manages capital for a wealthy individual or family. Some family offices invest directly in startups.
An investor who meets certain financial, professional, or regulatory qualifications that allow them to participate in certain private securities offerings.
A short presentation that explains the startup's problem, solution, market, product, traction, business model, team, and fundraising ask.
A brief summary of the business, usually one page, used to quickly introduce the company to investors.
A written overview of the company, including the opportunity, product, market, traction, team, and financing plan.
A spreadsheet that forecasts revenue, expenses, hiring, cash flow, and runway.
A secure folder where founders organize fundraising and diligence materials for investors.
The review process investors conduct before investing. This may include evaluating the team, market, product, financials, legal documents, customer traction, risks, and cap table.
A document outlining the main terms of a proposed investment before final legal agreements are completed.
The point when investment documents are signed and the financing becomes official. Funds are usually transferred at or shortly after closing.
An introduction to an investor made by someone the investor already knows or trusts.
An investor's stated agreement to invest, usually becoming official only once documents are signed.
A non-binding indication that an investor is interested in participating in a round.
The portion of a round reserved for a specific investor.
A round where investor interest is greater than the amount the company planned to raise.
A term that restricts the company from seeking competing offers for a certain period after signing a term sheet.
An investment offer that expires quickly unless accepted.
Raising money by selling ownership in the company.
Raising money through borrowed capital that is expected to be repaid, usually with interest.
A financing round where the company and investors agree on a specific valuation and price per share.
A debt instrument that can convert into equity in a future financing round, often with a valuation cap, discount, maturity date, and interest rate.
An early-stage financing agreement that gives an investor the right to receive equity in a future financing round.
The maximum valuation used to calculate how a SAFE or convertible note converts into equity.
A percentage reduction applied when a SAFE or convertible note converts into equity during a future priced round.
A provision that may allow an investor to receive better terms if the company later gives more favorable terms to another investor in a similar financing.
The date when debt is due to be repaid if it has not converted or been extended.
The rate at which debt accrues interest over time.
A class of stock usually issued to investors in priced rounds. Preferred stock often includes rights not held by common shareholders, such as liquidation preference, conversion rights, protective provisions, or board rights.
The basic class of company stock typically held by founders, employees, and advisors. Common stock usually sits behind preferred stock in payout priority.
A right to purchase company securities at a specific price during a defined period.
A record of the company's ownership structure, including founders, employees, investors, option pools, SAFEs, convertible notes, and shares.
An estimate of what a company is worth. In fundraising, valuation is usually negotiated between founders and investors.
The company's value before the new investment is added.
The company's value after the new investment is included.
An independent appraisal of the fair market value of a private company's common stock, commonly used to set the minimum strike price for employee stock options.
A reserve of shares set aside for future employee, advisor, or consultant equity grants.
Ownership calculated as if all options, warrants, SAFEs, convertible notes, and other equity rights have been exercised or converted.
The reduction in an existing shareholder's ownership percentage when the company issues new shares.
The price investors pay for each share in a priced equity round.
The total number of shares currently issued by the company. Depending on the context, people may discuss shares outstanding on an issued, outstanding, or fully diluted basis.
The ownership stake held by the company's founders.
An investor's right to participate in future financing rounds to maintain their ownership percentage.
A right allowing existing investors to buy new shares before outside investors, usually to help maintain ownership.
A right that gives preferred shareholders priority in receiving proceeds if the company is sold, liquidated, or wound down.
A provision that helps protect investors if the company later raises money at a lower valuation.
A more moderate form of anti-dilution protection that considers both the lower share price and the number of new shares issued.
A stronger anti-dilution mechanism that adjusts an investor's conversion price to match the lowest price at which new shares are issued.
A right that gives the company or existing investors the first opportunity to buy shares before they are sold to an outside buyer.
Rights that allow minority shareholders to participate in a sale if a major shareholder sells their shares to a third party.
Rights that allow certain shareholders to require other shareholders to support a company sale if approval thresholds are met.
Investor approval rights over major company actions, such as selling the company, issuing new shares, changing the charter, or taking on significant debt.
Rights that allow investors to receive company updates, financial statements, budgets, or other reporting.
A formal position on the company's board of directors, often negotiated by significant investors.
A person allowed to attend board meetings and receive board materials without having a formal vote.
Rights that allow a person to buy company stock at a set price within a defined period. Stock options are commonly used to compensate employees and advisors.
The price an option holder must pay to buy shares under a stock option.
The process by which founders, employees, or advisors earn equity over time or after meeting milestones.
A period before any equity vests, commonly one year for employee options.
Shares granted subject to restrictions, such as vesting or company repurchase rights.
A promise to deliver shares or equivalent value in the future once vesting or other conditions are met. RSUs are more common at later-stage startups than at the earliest stages.
Equity or options granted to an advisor in exchange for ongoing strategic help, introductions, or expertise.
A group of outside experts who provide advice, introductions, industry knowledge, or credibility without formal board authority.
The amount of time a company can operate before it runs out of cash, assuming current spending and revenue levels.
The rate at which a company spends cash, usually measured monthly.
Total cash spent by the company each month before accounting for revenue.
Cash spent each month after subtracting revenue or cash inflows.
The annualized value of recurring subscription revenue.
The amount of predictable recurring revenue generated each month.
The cost to acquire a new customer, often including sales and marketing expenses.
The estimated revenue or gross profit a customer generates over their relationship with the company.
The rate at which customers or revenue are lost over a period of time.
The percentage of customers or revenue a company keeps over time.
Evidence that a startup is gaining market interest, such as users, revenue, pilots, waitlists, engagement, or partnerships.
A state where the product clearly satisfies a strong market need and customers actively use, buy, or recommend it.
A basic version of a product built to test the core value proposition with users.
A limited test of a product with a customer or group of customers before a broader launch.
A document showing a customer, partner, or investor's intent to move forward, usually subject to further negotiation or conditions.
An event that allows shareholders to convert ownership into cash, such as an acquisition, IPO, tender offer, or secondary sale.
Transactions where companies merge, buy, sell, or combine businesses, assets, teams, or control.
When one company buys another company.
An acquisition primarily motivated by hiring the startup's team rather than acquiring its product or revenue.
The first sale of a private company's shares to public market investors. This is usually only relevant once a company reaches significant scale.
A transaction where existing shareholders sell shares to a buyer before a full company exit.
A company-approved process where shareholders are invited to sell some of their shares at a specified price.
The legal process of forming a corporation as a separate legal entity from its founders.
The core formation document filed with the state to create a corporation and define basic corporate structure.
The formation document typically used to create a limited liability company.
A corporation taxed separately from its owners and commonly used by venture-backed startups.
A corporation that elects pass-through tax treatment but has ownership restrictions that often make it less suitable for venture-backed startups.
A flexible business entity that can offer limited liability and pass-through taxation.
A corporation formed under Delaware law, commonly chosen by startups because of Delaware’s developed corporate legal framework.
The process of registering a company to do business in a state other than the state where it was formed.
A person or service designated to receive service of process and official legal or state documents for a company.
A federal tax identification number issued by the IRS for a business.
A company’s foundational governing document, often referring to the certificate of incorporation and related charter provisions.
Internal rules that govern how a corporation operates, including meetings, officers, and director procedures.
The governing contract for an LLC that defines member rights, management, economics, and procedures.
The collection of formation documents, approvals, ownership records, and legal paperwork a company must maintain.
The official record book containing key corporate documents, board approvals, stock records, and meeting minutes.
A written approval by the board of directors for a corporate action.
A written approval by stockholders for actions requiring owner approval.
A written approval signed by all required decision-makers instead of holding a formal meeting.
Written records of what happened at a board meeting, including decisions and approvals.
A formal written record of a decision approved by the board or stockholders.
The maximum number of shares a corporation is legally permitted to issue under its charter.
Shares that the company has actually issued to stockholders.
Shares previously issued and later reacquired by the company.
The nominal value assigned to shares in a corporation’s charter.
A document or electronic record representing ownership of shares.
Shares recorded electronically rather than represented by paper certificates.
A contract through which founders purchase their initial company shares.
A stock purchase agreement where shares are subject to restrictions such as vesting or company repurchase rights.
A company’s contractual right to buy back unvested or restricted shares under specified conditions.
A founder equity structure where the founder receives shares upfront but the company can repurchase unvested shares if the founder leaves.
The legal and operational process of handling a founder leaving the company.
A contract documenting the terms under which a founder exits or reduces involvement with the company.
An agreement or documented understanding about how founders divide initial ownership.
A board- and stockholder-approved plan that governs equity awards to employees, advisors, and consultants.
A contract granting a person the right to purchase shares under specified terms.
A type of employee stock option that may receive favorable tax treatment if legal requirements are met.
A stock option that does not qualify for incentive stock option tax treatment.
The ability to exercise stock options before they are vested, usually creating restricted shares.
A U.S. tax election for restricted stock or early-exercised stock that generally must be filed with the IRS within 30 days of the grant or transfer date.
The period during which an option holder can exercise vested options after leaving the company or before expiration.
The document summarizing key terms of an option award, such as grant date, number of shares, and exercise price.
Formal board or authorized committee approval of an equity award.
A U.S. securities-law exemption that private companies often use for compensatory equity grants to employees, advisors, and consultants.
A notice filing submitted to the SEC after the first sale in certain private securities offerings, commonly within 15 days of that first sale.
A set of SEC rules providing exemptions from public registration for certain private securities offerings.
A Regulation D exemption that allows private fundraising without general solicitation; companies can raise an unlimited amount and may sell to unlimited accredited investors plus up to 35 sophisticated non-accredited investors if additional requirements are met.
A Regulation D exemption that permits general solicitation if all purchasers are verified accredited investors.
A state securities notice or filing required in connection with certain securities offerings.
A non-public sale of securities conducted under an exemption from registration.
A contract in which an investor agrees to purchase securities from a company.
A form investors complete to confirm eligibility, sophistication, or accredited investor status.
A legal basis allowing a company to offer or sell securities without registering them publicly.
Public advertising or broad promotion of a securities offering.
A securities-law rule that can prevent reliance on certain exemptions if covered people have disqualifying events.
A securities-law exemption allowing eligible companies to raise capital from the public through regulated crowdfunding platforms.
A legal or contractual limit on selling, assigning, or transferring company securities.
The official record of a company’s stock issuances, transfers, cancellations, and owners.
A service provider that maintains shareholder records and processes transfers for a company.
A contract requiring the company to protect directors or officers against certain claims and expenses.
Legal duties directors owe to the company and its stockholders, often including duties of care and loyalty.
A fiduciary duty requiring directors to make informed decisions with appropriate attention and process.
A fiduciary duty requiring directors to act in the company’s interest rather than for improper personal benefit.
A situation where a person’s personal interests may interfere with their obligations to the company.
A transaction involving the company and a founder, executive, director, investor, or other connected party.
A fiduciary-duty principle that can restrict insiders from taking business opportunities that belong to the company.
A director who has a personal or financial interest in a transaction being considered by the board.
The minimum number of directors or stockholders required to hold a valid meeting or take action.
A person appointed to manage company functions, such as CEO, president, secretary, or treasurer.
A certificate from a company officer confirming corporate facts, approvals, or records.
A state-issued certificate showing that a company exists on the state’s records and has met required state filing and fee obligations for good standing.
A recurring state tax or fee required to maintain a business entity.
A recurring state filing that updates basic company information.
The legal process of formally closing a business entity.
The process of settling debts, distributing remaining assets, and closing operations after dissolution.
Legal rights in creations such as inventions, software, designs, names, logos, content, and confidential know-how.
A contract transferring ownership of intellectual property from a person or entity to the company.
An agreement requiring employees or contractors to assign work-related inventions and protect confidential information.
Non-public information that a company protects from unauthorized disclosure.
Valuable confidential business information that is protected because it is kept secret.
A government-granted right that can prevent others from making, using, or selling an invention for a limited time.
A temporary patent filing that can establish an early filing date while giving the applicant time to file a full application.
A word, phrase, symbol, or design that identifies the source of goods or services.
A trademark used to identify services rather than goods.
Legal protection for original works of authorship, such as software code, writing, images, and designs.
A legal doctrine under which certain works created by employees or specially commissioned contractors may be owned by the company if requirements are met.
A license that allows software to be used, modified, or distributed under specified conditions.
An open-source license that may require derivative works to be distributed under the same license terms.
An open-source license with fewer restrictions on reuse, modification, and distribution.
An agreement defining rights granted by contributors to a project or company.
The contractual or legal allocation of rights to use, control, access, or benefit from data.
A contract governing how a service provider processes personal data for a customer, especially where privacy laws require controller-processor terms.
A public notice explaining how a company collects, uses, shares, and protects personal information.
The online contract setting rules for using a product, website, or platform.
Rules describing prohibited uses of a product or platform.
A notice explaining how a website uses cookies and similar tracking technologies.
The process of collecting, storing, and honoring user consent choices for data use.
Information that identifies or can reasonably be linked to an individual.
Personal information that receives heightened protection, such as precise location, health data, or financial identifiers.
The party that determines the purposes and means of processing personal data.
The party that processes personal data on behalf of a controller.
A third party engaged by a processor to help process personal data.
A security incident involving unauthorized access, disclosure, or loss of protected data.
The legal or contractual process of notifying affected parties or regulators after a data breach.
The European Union’s General Data Protection Regulation governing personal data protection and privacy rights.
California privacy laws that give California consumers rights over certain personal information.
A U.S. healthcare privacy and security law that can apply to protected health information.
An independent attestation report based on AICPA Trust Services Criteria, commonly requested by enterprise customers to assess controls around security, availability, confidentiality, processing integrity, or privacy.
A written policy describing how a company protects systems, data, access, and security practices.
Security measures that limit who can access systems, data, or functions.
The process of evaluating and monitoring third-party vendors for legal, security, privacy, and operational risks.
A primary commercial contract governing the overall business relationship between a service provider and customer.
A document describing specific services, deliverables, timelines, and fees under a broader services agreement.
A document specifying commercial terms such as products, subscription length, users, and pricing.
A contract requiring parties to keep certain shared information confidential.
A confidentiality agreement where both parties agree to protect each other’s confidential information.
A confidentiality agreement where only one party is primarily disclosing confidential information.
A contractual obligation to compensate another party for specified losses or claims.
A contract clause that limits the amount or types of damages one party may owe.
A promise that certain facts, services, or products meet stated standards.
A statement limiting or excluding certain promises, liabilities, or responsibilities.
Statements of fact made in a contract that another party relies on.
A promise in a contract to do or not do something.
Failure to perform a contractual obligation.
A contractually defined period to fix a breach before stronger remedies apply.
Ending a contract because the other party breached or failed to meet specified obligations.
Ending a contract without needing to prove breach, if the contract allows it.
A contract provision that renews the agreement automatically unless a party cancels on time.
A contract commitment about service performance, uptime, support response time, or remedies.
A contract clause promising a customer terms or pricing at least as favorable as those given to certain other customers.
A contract term restricting one or both parties from working with competitors or alternatives.
A contract clause preventing a party from bypassing another party to deal directly with introduced contacts.
The jurisdiction’s law that controls how a contract is interpreted.
The geographic location or court where disputes may be heard.
A court’s or authority’s power to hear a dispute or enforce rules.
A private dispute-resolution process where an arbitrator decides the outcome instead of a court.
A non-binding dispute-resolution process where a neutral mediator helps parties try to settle.
A court order requiring a party to do or stop doing something.
A contractually agreed amount of damages for a specific breach, if enforceable.
A clause addressing extraordinary events beyond a party’s control that prevent performance.
A contract clause controlling whether rights or obligations may be transferred to another party.
A clause that gives rights or triggers obligations if ownership or control of a company changes.
A contract restricting a person from competing with a business, subject to major legal limits and significant state-by-state variation.
A contract restricting solicitation of employees, customers, or business relationships.
An employment relationship that either employer or employee may generally end at any time, subject to legal limits.
A letter describing basic employment terms such as role, salary, start date, and equity eligibility.
A document describing workplace policies, expectations, benefits, and procedures.
A contract defining services, payment, IP ownership, confidentiality, and contractor status.
The legal determination of whether a person is an employee or independent contractor.
A contract for advisory, technical, business, or professional services.
A contract describing an advisor’s role, confidentiality duties, compensation, and equity terms.
An agreement requiring employees to assign work-related inventions to the company.
Permission and disclosures required before conducting certain employment background checks.
A policy prohibiting harassment and describing reporting and investigation procedures.
Compliance with laws governing pay, overtime, breaks, wage statements, and working time.
An employee category not entitled to overtime under certain wage laws if legal requirements are met.
An employee category generally entitled to overtime and wage-hour protections.
Taxes employers must withhold or pay in connection with employee wages.
Compliance with rules for verifying work authorization and sponsoring employees when needed.
The U.S. process for confirming an employee’s identity and work authorization.
Tax rules affecting stock, options, RSUs, and other equity-based compensation.
Qualified Small Business Stock; a U.S. tax concept that may allow eligible stockholders to exclude certain capital gains if strict requirements are met.
The U.S. tax code section governing the Qualified Small Business Stock exclusion and its requirements.
A U.S. tax concept that may allow certain losses on small business stock to be treated more favorably if requirements are met.
A formal choice made under tax rules that can affect how income, equity, or transactions are taxed.
Rules governing pricing for transactions between related entities, especially across jurisdictions.
A connection with a state that can require a company to collect and remit sales tax.
A state or local permit allowing qualifying businesses to buy goods for resale without paying sales tax upfront.
A permit or registration required to legally operate certain businesses in a jurisdiction.
A license required to provide regulated professional services.
Following laws, regulations, standards, and rules that apply to a company’s business.
Processes used to verify customer identity and assess financial-crime risk.
Laws and controls intended to prevent money laundering and financial crime.
A state license that may be required for businesses transmitting money or stored value.
Rules restricting the transfer of certain products, technology, software, or data to particular countries, users, or uses.
Checking customers, vendors, or transactions against government sanctions lists.
Laws protecting customers from unfair, deceptive, or abusive business practices.
A statement in marketing that may need substantiation to avoid being misleading.
A disclosure required when influencers, reviewers, or endorsers have a material connection to a brand.
Designing digital products and services to meet legal accessibility obligations and standards.
Privacy rules governing collection and use of data from children.
A mechanism used to ask or verify a user’s age for legal or compliance reasons.
Rules and procedures for reviewing, removing, or restricting user-generated content.
Legal risk related to operating a platform where users, sellers, or third parties interact.
Content created or uploaded by users rather than the company.
A copyright process for requesting removal of allegedly infringing online content.
A legal protection available when specific requirements are met.
A contract transferring certain business risks to an insurer in exchange for premiums.
Insurance covering certain third-party bodily injury, property damage, and related claims.
Insurance covering certain claims related to professional mistakes or service failures.
Insurance covering certain losses or liabilities from cyber incidents.
Insurance protecting directors and officers against certain claims related to company management.
Insurance covering certain employment-related claims.
Insurance sometimes used in acquisition transactions to cover breaches of certain deal representations.
A contract describing the scope, fees, and terms of professional services by a lawyer, accountant, or advisor.
A formal statement from counsel about specified legal matters.
A legal protection for confidential communications between attorney and client made for legal advice.
A protection for materials prepared in anticipation of litigation.
A notice requiring preservation of relevant documents and data when litigation is reasonably anticipated.
A letter asserting a claim or requesting action before litigation.
A letter demanding that a person or company stop allegedly unlawful or harmful activity.
The legal document that starts a lawsuit and states the plaintiff’s claims.
A legal notice informing a defendant that a lawsuit has been filed and a response is required.
A contract resolving a dispute between parties.
A contract provision where a party gives up specified legal claims.
The intentional relinquishment of a known right.
A contract setting terms for an employee’s departure, often including pay, releases, and ongoing obligations.
A contract clause restricting negative statements about another party.
The transfer of contractual rights or legal claims from one party to another.
A contract process where a new party replaces an original party with consent of the relevant parties.
Risk that a buyer or successor entity may inherit certain liabilities of another business.
A contract for buying selected assets of a business rather than the company itself.
A contract for buying ownership interests in an LLC.
A document in a transaction that lists exceptions, details, and disclosures tied to contract representations.
A contract where a third party holds money, documents, or assets until conditions are met.
A deal term where some payment depends on future performance or milestones.
Requirements that must be satisfied before a transaction can close.
A significant negative change that may affect a company or transaction under a contract.
A contract clause stating that only the parties can enforce the agreement unless otherwise specified.
A clause stating that the written contract is the complete agreement between the parties.
A clause providing that if one part of a contract is invalid, the rest can still remain effective.
A contract clause allowing parties to sign separate copies that together form one agreement.
A digital method of signing a document that can be legally valid if requirements are met.
A formal process where a notary verifies identity and witnesses a signature.
The total money a company earns from selling products or services before subtracting expenses.
Revenue that repeats over time, such as subscriptions, retainers, or long-term contracts.
Revenue from a single transaction that does not automatically repeat.
The amount of predictable recurring revenue a company earns each month from subscriptions or recurring contracts.
The annualized value of recurring revenue. ARR is commonly used by SaaS and subscription companies.
The increase in ARR after accounting for new revenue, expansion revenue, contraction, and churn.
Additional recurring revenue from existing customers through upgrades, more seats, add-ons, or increased usage.
Revenue lost when existing customers downgrade, reduce seats, or lower usage, excluding full churn.
Average revenue generated per user or customer over a specific period.
Average revenue generated per customer account. This is often more useful than ARPU for B2B companies where one account may include many users.
The average annual value of a customer contract. ACV is commonly used in B2B sales.
The total value of a contract over its full term, including all committed revenue.
The average amount spent per order or transaction.
The total value of transactions processed through a marketplace before subtracting fees, refunds, discounts, or costs.
The percentage of each transaction that a marketplace or platform keeps as revenue.
Revenue that increases or decreases based on customer usage.
The rate at which a metric increases over time, often revenue, users, customers, or usage.
The percentage change from one month to the next.
The percentage change compared with the same period in the prior year.
Evidence that customers or users want what the company is building. Traction can include revenue, usage, waitlists, pilots, retention, partnerships, or customer growth.
A customer who has actually paid for the product, not just signed up, joined a waitlist, or started a free trial.
The single metric that best reflects the core value your product delivers to customers.
A metric used to measure progress toward an important business goal.
A metric that may predict future performance.
A metric that reflects results after they happen.
A measurable achievement that helps prove progress and reduce risk.
The average cost to acquire a new customer, usually including sales and marketing spend.
CAC calculated across all acquisition channels combined.
CAC calculated only from paid acquisition channels, such as ads, sponsorships, or paid partnerships.
The cost to drive a specific acquisition action, such as a signup, lead, or purchase.
The amount paid for each click in a paid marketing campaign.
The cost per 1,000 ad impressions.
The percentage of people who click after seeing an ad, email, or link.
Revenue generated for every dollar spent on advertising.
The percentage of people who complete a desired action, such as signing up, booking a demo, starting a trial, or purchasing.
The percentage of trial users who become paying customers.
The percentage of waitlist users who become active or paying customers.
Customers gained without directly paying for each acquisition, often through SEO, referrals, word of mouth, content, or community.
A path used to acquire customers, such as SEO, paid ads, partnerships, outbound sales, content, events, affiliates, or referrals.
A rapid experimentation approach to finding scalable ways to acquire, activate, and retain users, often using creative, low-cost tactics.
A growth mechanism where existing users invite new users, creating a repeating acquisition cycle.
A product or marketing loop where users naturally bring in more users as part of using or sharing the product.
The average number of new users each existing user brings in.
Another way to describe viral growth, usually based on how many invitations users send and how many invited users convert.
The growth rate of qualified leads month over month.
The steps a customer takes from first discovering the company to becoming a paying or retained customer.
The earliest stage of customer interest, often website visitors, impressions, leads, or signups.
The moment a user experiences the product's core value for the first time.
The percentage of new users who reach the activation milestone.
How quickly a new user reaches the product's core benefit.
The process that helps new users understand and start using the product successfully.
The number of unique users who use the product in a day.
The number of unique users who use the product in a week.
The number of unique users who use the product in a month.
A measure of how frequently monthly users return daily.
A measure of how many weekly users are active daily.
How often and how deeply users interact with a product.
The average amount of time users spend in a product during a session.
How often users return to the product during a given period.
Tracking groups of users or customers who started in the same time period to understand behavior over time.
A group of users or customers tracked over time to measure whether they keep using or paying for the product.
How frequently users return to the product, often measured using ratios such as DAU/MAU.
The percentage of customers or users who continue using or paying for the product over time.
The percentage of customers lost during a period.
The percentage of revenue lost from existing customers during a period.
Customer churn measured by number of customers rather than revenue.
The percentage of customers retained over a period, measured by customer count instead of revenue.
The percentage of revenue retained from existing customers before expansion revenue.
Revenue retained from existing customers after churn, downgrades, upgrades, and expansion.
Another term often used similarly to net revenue retention, especially in SaaS and subscription businesses.
Additional revenue from existing customers through upgrades, add-ons, more seats, or higher usage.
A score used to estimate how likely a customer is to renew, expand, or churn.
A customer satisfaction metric based on how likely users are to recommend the product.
A metric that asks customers how satisfied they are with a product, service, or interaction.
The percentage of customers who make more than one purchase.
A person or company that may become a customer.
A lead that marketing considers likely to become a customer based on behavior or fit.
A lead that sales considers ready for direct outreach or an active sales process.
A qualified sales prospect with a real chance of becoming a customer.
The total set of active sales opportunities being worked by the sales team.
Sales opportunities that have been vetted and are more likely to close.
The amount of sales pipeline compared with the sales target.
The time it takes to convert a prospect into a paying customer.
The percentage of sales opportunities that become customers.
Another term for the percentage of prospects or opportunities that become paying customers.
A measure of how quickly revenue moves through the pipeline, often influenced by deal size, win rate, number of opportunities, and sales cycle length.
A sales target assigned to a salesperson or team.
The value of signed customer contracts, often before all revenue is recognized.
A measure of how effectively sales and marketing spend turns into new revenue.
A SaaS sales efficiency metric that compares new recurring revenue growth to prior-period sales and marketing spend.
Early sales driven directly by the founder before hiring a full sales team.
The revenue and costs associated with one customer, transaction, or unit of business.
The estimated revenue or gross profit a customer generates over the full customer relationship.
A comparison of customer lifetime value to customer acquisition cost.
Revenue remaining after subtracting the direct cost of delivering the product or service.
Revenue remaining after variable costs associated with selling or delivering the product.
The time it takes to recover the cost of acquiring a customer from gross profit.
How long it takes to recover an upfront cost. In startups, this often refers to recovering customer acquisition cost.
The point where revenue equals costs, so the business is no longer losing money.
A capital efficiency metric comparing net burn to net new ARR.
A SaaS benchmark where revenue growth rate plus profit margin equals 40% or more. It is usually more relevant for later-stage SaaS companies than very early startups.
The rate at which a company spends cash, usually measured monthly.
Total cash expenses before accounting for revenue.
Cash spent after subtracting revenue or cash inflows.
The amount of time a company can operate before running out of cash.
The movement of money into and out of the business.
Ongoing costs required to run the business, such as salaries, software, rent, and marketing.
Direct costs required to deliver the product or service.
A startup is default alive if it can reach profitability before running out of cash without needing another fundraise.
A startup is default dead if it will run out of cash before becoming profitable unless it raises more money or changes course.
A plan for how the company expects to spend and receive money over a period.
A projection of future business performance based on assumptions about revenue, expenses, hiring, and growth.
The difference between planned performance and actual performance.
A basic version of a product built to test the core value proposition with users.
A state where the product clearly solves an important problem for a defined market and customers actively use, pay for, or recommend it.
A description of the type of customer most likely to buy, retain, and get strong value from the product.
A profile of a specific user or buyer within the target customer base.
The total revenue opportunity if the company captured the entire relevant market.
The portion of the total market the company can realistically serve with its current or near-term product.
The portion of the market the company can realistically capture over a defined period.
The strategy for reaching customers, selling the product, and generating revenue.
A growth strategy where the product itself drives acquisition, activation, conversion, and expansion.
A growth strategy where a sales team drives customer acquisition and revenue.
Evidence that customers are actively asking for, adopting, or buying the product.
A durable advantage that makes the business harder to copy or compete against.
A dynamic where the product becomes more valuable as more people or participants use it.
The difficulty or cost customers face when moving from one product to another.
The percentage of customers who buy an add-on, upgrade, or companion product.
How easily buyers and sellers can find and complete successful matches in a marketplace.
The sellers, providers, inventory, or resources available in a marketplace.
The buyers, users, or customers seeking a product or service in a marketplace.
The percentage of demand that is successfully fulfilled.
The percentage of users or transactions successfully matched in a marketplace.
The percentage of transactions that are refunded.
The percentage of products customers return.
How quickly inventory is sold and replaced over a period.
For AI products, the cost of running the model each time a user request is processed.
The cost of cloud infrastructure, GPUs, servers, or processing used to deliver the product.
Whether users continue using the product over time, especially in usage-based or AI products.
Gross margin calculated at the customer level to understand whether certain customers are more profitable than others.
How a startup plans to spend the capital it raises.
The total capital a startup is seeking in a financing round.
The percentage of the company sold to investors in a round.
Raising enough capital to reach the next meaningful proof point.
A recurring message sent to investors or advisors summarizing progress, metrics, asks, and challenges.
A secure folder where founders organize fundraising and diligence materials.
The explanation that connects startup metrics to the company's strategy, progress, and fundraising story.
The specific proof points a startup needs to reach before raising the next round.
Investors want to understand who the founders are, what they have done before, and why their experience matters for this company.
Investors want to understand whether the founders are fully committed, how long they have been working on the company, and whether this is a serious venture or still exploratory.
Investors may ask when the company began because traction is more meaningful when measured against time.
Founder-market fit is the connection between the founding team and the problem they are solving. Investors want to know why this team has a special right to win.
Domain expertise means the team deeply understands the industry, customer, workflow, or pain point they are solving for. This can be especially important in healthcare, fintech, legal, enterprise software, insurance, logistics, defense, and regulated markets.
Investors look at whether the team has the skills needed to build, sell, and operate the business.
Investors look for whether the founding team has a balanced mix of skills across product, engineering, sales, operations, and domain knowledge.
For technology startups, investors often want to know whether the team has in-house technical ability or depends heavily on contractors.
Investors also want to know who is responsible for customers, sales, partnerships, and fundraising.
Investors want clarity on who owns product, technology, sales, operations, fundraising, and major company decisions.
Investors may ask how equity is divided among founders and whether the split reflects contribution, role, risk, and commitment.
Founder vesting helps protect the company if a founder leaves early. Investors often prefer that founder equity vests over time.
Investors want to know how the company plans to grow the team after funding.
Investors want to know which roles are missing and whether the fundraise will help fill those gaps.
Investors want to know what painful, urgent, or expensive problem the company is solving.
Investors want to know whether the problem is painful enough that customers will take action now.
Customer pain is the specific cost, frustration, inefficiency, risk, or missed opportunity that motivates customers to seek a solution.
Investors want a precise view of who the startup serves, not a vague answer like "everyone."
In many startups, the person using the product is not the person approving the purchase. Investors want to understand both.
Investors want to know where the money comes from inside the customer's organization.
Investors look for evidence that founders have spoken with enough customers to understand the problem deeply.
Customer validation goes beyond interviews and shows that customers are willing to act, pay, pilot, sign, or switch.
Investors want to understand how customers solve the problem today, even if they are not using a direct competitor.
Investors want to know what happens if the customer does not solve the problem.
Investors want to know whether the opportunity is large enough to support venture-scale returns.
TAM is the total addressable market, SAM is the portion you can realistically serve, and SOM is the portion you can realistically capture in a defined period.
Investors often prefer market sizing based on real customer counts and pricing rather than broad industry numbers.
A beachhead market is the first narrow segment the startup targets before expanding into a broader opportunity.
A wedge is the initial product, feature, use case, or customer segment that helps the company enter the market.
Investors want to know why now is the right moment for the company to exist.
Investors want to understand whether the company is entering an existing category, redefining a category, or creating a new one.
Investors want to know whether the company can become large enough to matter to a venture fund.
Investors want to understand whether the startup can expand beyond its first customer segment or product.
Investors want a simple explanation of how the product solves the customer problem.
Investors want to know what has actually been built versus what is still planned.
Investors often want to see the product in action, even if it is early.
Investors want to know what has been built and what has been validated.
Investors want to understand whether the technology creates a real advantage or whether the product is easy to replicate.
Technical risk is the chance that the product may be difficult, expensive, or impossible to build at the required quality or scale.
Investors want to know what the company plans to build next and how those features connect to customer demand or revenue.
Scalability is the ability of the product, infrastructure, and operations to handle more customers without breaking or becoming too expensive.
Some startups become stronger as they collect proprietary data, improve models, or learn from customer usage.
For AI or automation companies, investors want to understand whether the product uses AI in a meaningful and defensible way.
Investors want to know how difficult it is for customers to start using the product.
Investors want to know if the product depends heavily on third-party APIs, platforms, data providers, models, or infrastructure.
Traction is evidence that customers or users want the product. At the early stage, this may include revenue, pilots, waitlists, usage, retention, signed LOIs, or customer interviews.
Investors want evidence that the company is making measurable progress. Relevant traction depends on the business model and stage.
Revenue shows that customers are willing to pay, even if the company is still early.
For consumer, marketplace, or product-led companies, investors may care about active users, engagement, and usage frequency.
Retention shows whether customers or users keep coming back after trying the product.
Engagement shows how deeply customers or users interact with the product.
Investors want to see whether the company is growing and whether growth is accelerating.
Investors want to know whether potential customers are serious, qualified, and likely to close.
An LOI shows a customer or partner's intent to move forward, but it may not be binding. Investors will want to know how strong the commitment really is.
A pilot is a limited test with a customer before broader rollout. Investors care whether pilots are paid, active, and likely to convert.
Customer concentration means too much revenue depends on one or a few customers. This can create risk.
Investors want to know whether metrics are meaningful or inflated.
Cohort performance shows whether groups of customers or users retain and expand over time.
Investors may ask to speak with customers to verify value, urgency, and satisfaction.
Investors want to understand how the company makes money and whether the model can scale.
Pricing shows how the company captures value from customers.
Investors want to know whether revenue is recurring, usage-based, transactional, services-based, marketplace-based, or a mix.
Gross margin shows how much revenue remains after direct costs of delivering the product or service.
CAC shows how much it costs to acquire a customer.
LTV estimates how much value a customer generates over the relationship.
CAC payback measures how long it takes to recover the cost of acquiring a customer.
Unit economics show whether each customer, transaction, or account can become profitable.
Investors want to know whether existing customers can spend more over time.
If a company has services revenue, investors want to know whether that revenue supports or distracts from scalable product revenue.
GTM strategy explains how the company reaches, converts, and retains customers.
Sales motion describes how the product is sold, such as founder-led, self-serve, product-led, outbound, channel-led, or enterprise sales.
Sales cycle is the time it takes to turn a prospect into a paying customer.
At the early stage, founders often sell directly before hiring a sales team. Investors want to know what founders are learning from that process.
Distribution strategy explains how the company will reach customers at scale.
Channel risk occurs when the company depends too heavily on one acquisition channel, platform, partner, or algorithm.
Partnerships can support distribution, credibility, integrations, or sales. Investors want to know whether they are real and valuable.
Investors want to know whether the company can repeat its sales and growth process, not just close one-off deals.
Scalable acquisition means the company can acquire more customers without costs rising too quickly or relying only on founder hustle.
Sales pipeline is the set of potential customer opportunities in progress.
Investors want to understand how efficiently the company converts prospects into customers.
Investors want to understand when the company should move from founder-led selling to dedicated sales hires.
Investors want to know who else is solving the problem and how customers compare options.
Direct competitors solve the same problem for a similar customer with a similar product.
Indirect competitors solve the same customer pain in a different way.
Investors want to know why a large existing company cannot easily copy, bundle, or underprice the startup's product.
Competitive advantage is what helps the company win against alternatives.
A moat is a durable advantage that makes the business harder to copy or compete against over time.
Investors want to know how the company becomes harder to compete with as it grows.
Switching costs are the effort, money, risk, or disruption a customer faces when moving away from a product.
Network effects happen when a product becomes more valuable as more users, customers, or participants join.
Investors may ask whether the team can move faster than competitors.
Investors want to know whether the startup fits their stage, sector, geography, check size, and ownership strategy.
Investors want to know whether someone is leading the round or whether the company is still searching for a lead.
Investors want to understand the urgency and process around the fundraise.
Investors want to know how much capital has already been raised and how much the company is seeking now.
This explains how the company will spend the money.
Runway shows how long the company can operate before needing more capital.
Investors want to know what the company will accomplish before raising again.
Investors review the ownership structure to understand founder ownership, prior investors, SAFEs, notes, option pool, and dilution.
Investors may ask about valuation expectations to understand whether the round is realistic.
Investors may ask how much capital the company truly needs to reach the next proof point.
Investors need to confirm the company is properly formed and in good standing.
Investors want to confirm the company owns or has the right to use the technology, brand, content, or data it depends on.
Investors review major customer, supplier, partner, distribution, and vendor contracts to understand obligations and risks.
Investors want to know whether the company faces legal, regulatory, privacy, security, or industry-specific risks.
Investors want to understand how the company protects customer and user data.
Investors want to know whether there are lawsuits, threatened claims, founder disputes, employee disputes, IP disputes, or customer conflicts.
Investors want to know whether employees and contractors have proper agreements, including confidentiality and IP assignment.
Investors want founders to be honest about what could go wrong.
Investors want to know how founders plan to reduce the biggest risks.
Location may matter for program eligibility, investor focus, local ecosystem access, customer proximity, or hiring strategy.
For accelerators, fellowships, or venture programs, investors may ask whether the company fits the program's goals, stage, and sector focus.
Investors may want to understand whether the company aligns with a specific thesis, sector focus, or strategic priority.
A concise presentation covering the problem, solution, market, product, traction, business model, team, and fundraising ask.
A forecast showing revenue, expenses, hiring, cash flow, burn, and runway.
A spreadsheet or platform export showing company ownership.
Customers or users investors can speak with to validate demand and satisfaction.
Usage, retention, activation, engagement, conversion, and revenue metrics.
Corporate formation documents, IP assignments, employment agreements, financing documents, and major contracts.
Documents showing customer commitments, pilots, paid trials, LOIs, or signed contracts.
A secure folder where diligence materials are organized.
A form of private investment focused on high-growth startups, usually in exchange for equity or equity-linked securities.
An investment vehicle that pools capital from LPs and is managed by GPs who invest in startups.
The fund manager responsible for raising the fund, making investment decisions, supporting portfolio companies, and managing fund performance.
An investor in a venture fund who provides capital but typically does not make day-to-day investment decisions.
A newer fund manager, often raising Fund I, Fund II, or Fund III.
A major LP whose early commitment helps a fund manager raise the rest of the fund.
The total amount LPs agree to provide to a fund over time.
A request for LPs to send a portion of their committed capital so the fund can make investments or pay expenses.
An annual fee paid to the fund manager to cover salaries, sourcing, diligence, operations, and fund administration.
The share of fund profits paid to the GP after LPs receive the returns required under the fund agreement.
The expected duration of a fund, often around 10 years with possible extensions.
The period when a fund actively makes new investments.
The later part of a fund's life when the manager focuses on exits, distributions, and portfolio value realization.
The year a fund begins investing, often used to compare performance against funds from the same market cycle.
Capital that has been committed to a fund but not yet invested.
Capital set aside for future investments in existing portfolio companies.
The practice of reinvesting certain proceeds or returned capital into new or follow-on investments, if allowed by the fund documents.
A separate agreement giving a specific LP additional rights or terms.
An investment vehicle that invests in other funds rather than directly into startups.
A standalone investment vehicle created for a specific investment or transaction.
The group within a VC firm that reviews and approves investment decisions.
A written investment recommendation summarizing the company, market, team, traction, risks, deal terms, and expected return potential.
A senior investor who usually sources deals, leads diligence, wins allocations, supports portfolio companies, and participates in investment decisions.
A mid-to-senior investment role, often between associate and partner, with responsibility for sourcing, diligence, and sometimes leading deals.
An investment team member who often sources companies, screens pitch decks, builds market maps, prepares memos, and supports diligence.
A junior investment role focused on research, sourcing, market analysis, financial analysis, and CRM management.
A part-time or affiliated investor who helps source deals, advise companies, or support a fund's sector expertise.
A specialist who helps portfolio companies with functional areas such as hiring, sales, finance, marketing, product, or go-to-market.
A VC team focused on non-investment support such as talent, marketing, community, business development, events, or founder services.
A person outside or adjacent to the fund who helps identify promising startups, often in exchange for compensation or upside.
The stream of investment opportunities a VC firm sees.
Investment opportunities that a firm sees through differentiated access rather than broad market circulation.
Deals that come to the firm through pitch submissions, referrals, demo days, or founder outreach.
Proactive efforts by investors to identify and contact promising companies.
An introduction through a trusted mutual connection.
A structured view of companies in a sector, often grouped by category, customer type, business model, or technology layer.
An investment approach based on a defined view of where a market is going.
Investing outside a strict thesis when a compelling deal appears.
Estimating market opportunity using broad market data and industry-level assumptions.
Estimating market opportunity from customer counts, pricing, usage, and realistic adoption assumptions.
The total revenue opportunity available if a company captured the entire relevant market.
The portion of the TAM the company can realistically serve with its current or near-term product.
The portion of the market the company can realistically capture over a defined period.
An exit large enough to materially return capital to a venture fund.
The venture capital pattern where a small number of investments often drive most of a fund's returns.
Comparing a company's metrics against peers, historical data, or expected stage standards.
The earliest institutional or semi-institutional funding stage, often used to build an MVP, validate demand, and assemble the team.
An early round used to build product, acquire initial customers, and prove early traction.
A round typically used to scale a product that has early product-market fit or strong evidence of repeatable demand.
A growth-oriented round used to scale sales, expand markets, and strengthen operations.
Short-term financing used to help a company reach the next milestone or financing event.
Additional capital raised after a seed round, often before Series A readiness.
A financing completed at a higher valuation than the prior round.
A financing completed at a lower valuation than the prior round.
A financing completed at roughly the same valuation as the prior round.
A round funded mainly by existing investors.
An additional investment by an existing investor in a later round.
An investment where capital is exchanged for ownership in the company.
A financing where the company and investors agree on a valuation and price per share.
Debt that can convert into equity in a later financing, often with a valuation cap, discount, maturity date, and interest rate.
A financing agreement that gives investors the right to receive equity in a future financing.
A class of stock commonly issued to VC investors, usually carrying rights not held by common stockholders.
The basic class of stock usually held by founders, employees, and advisors.
Preferred stock that can convert into common stock, often at the investor's election or automatically in an IPO.
A right to purchase securities at a specified price within a certain period.
Debt financing provided to venture-backed companies, often used alongside equity to extend runway without immediately increasing dilution.
Capital that does not require selling equity, such as grants, certain loans, or revenue-based financing.
The estimated worth of a company, often negotiated between investors and founders.
The company's valuation before the new investment.
The company's valuation after the new investment is included.
A record of company ownership, including founders, employees, investors, options, SAFEs, notes, and other securities.
Ownership calculated as if all options, warrants, convertible instruments, and other rights have converted or been exercised.
Shares reserved for employee, advisor, or consultant equity grants.
Negotiating whether the option pool increase is included in the pre-money valuation or added after the investment.
The reduction in ownership percentage when new shares are issued.
The percentage of a company a fund aims to own after investing.
The valuation or price at which an investor enters an investment.
The price paid for each share in a priced financing.
An independent appraisal of a private company's common stock value, often used for setting option strike prices.
A private company valued at $1 billion or more.
A document outlining the key terms of a proposed investment before definitive legal documents are completed.
A right giving preferred shareholders priority in receiving proceeds from a sale or liquidation.
Preferred stock that receives its liquidation preference and then also participates in remaining proceeds with common shareholders.
Preferred stock where the investor typically chooses between taking the liquidation preference or converting to common stock.
A provision that adjusts investor economics if the company later issues shares at a lower price.
An anti-dilution formula that considers both the lower price and the amount of new stock issued.
A more investor-favorable anti-dilution mechanism that resets the investor's conversion price to the lowest new issue price.
Rights allowing investors to participate in future rounds to maintain ownership.
Rights allowing an investor to buy more than its current ownership share in a future round.
A right allowing the company or investors to buy shares before they are sold to a third party.
Rights allowing investors to participate in a sale by another shareholder, often a founder.
Rights that can require shareholders to support a company sale if specified approvals are met.
Investor approval rights over major actions such as selling the company, issuing new shares, taking on debt, or changing the charter.
Rights to receive financial statements, budgets, updates, and other company information.
A formal voting position on the company's board of directors.
A person allowed to attend board meetings and receive materials without a formal vote.
The minimum number or percentage of participants required for a valid board or shareholder decision.
The investigation investors conduct before investing, covering team, market, product, customers, finances, legal matters, and risks.
Reviewing the market, customers, competitors, pricing, and go-to-market strategy.
Assessing the product, architecture, engineering quality, defensibility, and technical risks.
Reviewing incorporation documents, IP ownership, contracts, compliance, employment matters, and prior financing documents.
Reviewing revenue, expenses, cash balance, projections, burn, runway, and accounting quality.
A conversation with people who have worked with the founder, company, product, or market.
A conversation with current or potential customers to validate product value and demand.
A concern that could materially affect the investment decision.
The risk that a company depends too heavily on one individual.
Revenue dependence on a small number of customers.
A durable advantage that helps a company defend its market position.
The amount of time a company can operate before running out of cash at its current burn rate.
The rate at which a company spends cash, usually monthly.
Total monthly cash expenses before revenue.
Monthly cash outflow after subtracting cash inflows such as revenue.
A measure of capital efficiency, often calculated as net burn divided by net new ARR.
Annualized recurring subscription revenue.
Recurring subscription revenue generated each month.
The rate at which revenue increases over time.
Revenue remaining after direct costs of delivering the product or service.
The cost to acquire a customer, usually including sales and marketing spend.
The estimated economic value of a customer over the customer relationship.
A comparison of customer lifetime value to acquisition cost.
The time required to recover customer acquisition cost from gross profit.
The rate at which customers or revenue are lost.
Revenue retained from existing customers after expansion, contraction, and churn.
Revenue retained from existing customers before expansion revenue.
The percentage of users who complete a key action that signals product value.
A group of users or customers tracked over time to measure ongoing usage or revenue.
A metric used to measure business performance.
Evidence that the company is gaining real market pull.
A state where customers strongly need, use, and often recommend the product.
The strategy for how a fund allocates capital across stages, sectors, check sizes, ownership targets, and reserves.
The amount a fund invests in a company.
The first investment a fund makes in a company.
Capital reserved to invest in later rounds of existing portfolio companies.
The percentage of investments that return less than the capital invested.
The percentage of investments that become meaningful winners.
An investment outcome large enough to return capital equal to the entire fund size.
Total value divided by invested capital.
A time-sensitive return metric that estimates annualized performance.
Cash returned to LPs divided by capital paid into the fund.
Distributed value plus remaining unrealized value, divided by paid-in capital.
Remaining unrealized value divided by paid-in capital.
The estimated value of a fund's remaining assets after liabilities.
The pattern where a fund may show negative returns early due to fees and immature investments, then improve as portfolio companies appreciate or exit.
An increase in the carrying value of an investment.
A decrease in the carrying value of an investment.
Reducing an investment's value to zero or near zero.
A governing body responsible for major company oversight and strategic decisions.
Rights allowing an investor representative to attend board meetings without voting.
Hands-on help provided by investors or platform teams.
An investor who contributes meaningful support beyond capital.
A company in which a fund has invested.
An investor approach that seeks to support founders with fair terms, useful help, and long-term alignment.
Introductions to customers, hires, partners, advisors, or future investors.
Help with recruiting executives, employees, advisors, or board members.
Helping a portfolio company prepare for and raise its next round.
A liquidity event that allows investors to realize value, often through acquisition, IPO, secondary sale, or tender offer.
A view of how investors may eventually realize returns.
When one company buys another.
Transactions involving the purchase, sale, or combination of companies.
A sale of a company to an operating business, often a strategic acquirer.
The first public sale of a private company's shares.
A sale of existing shares by current shareholders rather than a new issuance by the company.
A company-approved process where shareholders can sell shares at a specified price.
An acquisition primarily focused on hiring the team rather than buying the business.
A valuation multiple used to estimate or evaluate an exit price.
Cash or stock returned from the fund to LPs.
A distribution of securities rather than cash.
An investor meeting SEC financial, professional, or regulatory qualifications for participating in many private securities offerings.
Publicly advertising or broadly marketing a private securities offering, subject to securities-law rules and exemptions.
A confidentiality agreement restricting how shared information may be used or disclosed.
A legal or ethical obligation to act in the best interests of another party, depending on role and context.
A situation where competing interests could affect judgment or obligations.
A clause allowing an investor to receive better terms granted to later investors in similar securities.
A step-by-step method for solving a problem or completing a task.
Application Programming Interface; a way for software systems to communicate with each other.
A value passed into a function when the function is called.
An ordered collection of values stored together.
The server-side part of an application that handles logic, databases, authentication, and APIs.
A base-2 number system using only 0s and 1s.
A data type with only two possible values: true or false.
A separate version of code used to work on changes without affecting the main codebase.
An error or flaw that causes software to behave incorrectly.
The process of preparing code so it can run or be released.
Temporary storage used to make data access faster.
To run or invoke a function.
In object-oriented programming, a blueprint for creating objects.
The software or device that requests data or services from a server.
A model where clients request resources or services from centralized servers.
The on-demand delivery of computing resources over the internet, often using pay-as-you-go pricing.
The complete collection of source code for a project or product.
A text-based interface for typing commands to interact with a computer.
A program that translates source code into machine-readable code before execution.
Code that runs different instructions depending on whether a condition is true or false.
A packaged application environment that includes code, dependencies, and configuration so software runs consistently.
Create, Read, Update, Delete; the four basic operations for managing stored data.
Cascading Style Sheets; the language used to style web pages.
A way of organizing and storing data for efficient use.
A system for storing, organizing, retrieving, and managing data.
The process of finding and fixing errors in code.
External code, software, or services that a project relies on.
A design pattern where a component receives its dependencies from the outside instead of creating them itself.
The process of releasing code to an environment where it can run.
Practices that combine software development and IT operations to improve delivery and reliability.
A platform for building, shipping, and running containerized applications.
A specific URL or location where an API can be accessed.
The process of converting data into a secure format that unauthorized users cannot easily read.
The setting where software runs, such as local, development, staging, or production.
An error or unusual condition that interrupts normal program execution.
A structured set of tools, libraries, and conventions for building applications.
The user-facing part of an application, including interface, layout, and interactions.
A reusable block of code that performs a specific task.
A distributed version control system used to track code changes.
A platform for hosting Git repositories and collaborating on code.
HyperText Markup Language; the standard language used to structure web pages.
HyperText Transfer Protocol; an application-layer protocol used for communication on the web.
Integrated Development Environment; an application for writing, running, and debugging code.
An object-oriented programming concept where one class receives properties or behavior from another class.
A specific object created from a class.
A contract that defines required behavior or methods for a component or class.
A program that runs code directly, often line by line, instead of compiling it first.
Repetition of a process, or a gradual improvement cycle.
A programming language widely used for web development and also used outside browsers through environments like Node.js.
JavaScript Object Notation; a lightweight data format commonly used by APIs.
Reusable code that developers can import into projects.
Code that repeats until a condition is met or a sequence is complete.
A field of computing where systems learn patterns from data to make predictions or decisions.
An architecture where an application is split into small, independent services.
Software that sits between systems or request/response steps to process data.
A self-contained unit of code that can be reused or imported elsewhere.
Model-View-Controller; a design pattern that separates data, user interface, and control logic.
A bundle of data and behavior, commonly used in object-oriented programming.
Object-Oriented Programming; a programming style based on objects, classes, and methods.
Software whose source code is publicly available for use, review, modification, or distribution.
A bundled piece of reusable software, often installed through a package manager.
A named input defined by a function.
Improving software speed, efficiency, memory use, or responsiveness.
The hardware, operating system, or software environment where applications run.
The live environment used by real users.
An early rough version built to test an idea.
A general-purpose programming language known for readability and broad use.
A request for data, often from a database or search system.
Improving the structure of code without changing what it does.
A storage location for a project's code and version history.
An architectural style for designing stateless, resource-based APIs.
The period when a program is executing, or the environment required to execute it.
Software Development Kit; tools, libraries, and documentation for building on a platform.
A system that provides data, services, or resources to clients.
Human-written instructions that make up a software program.
Structured Query Language; a language for managing and querying relational databases.
Either a last-in, first-out data structure or the set of technologies used in a project.
A type system where variable types are checked before runtime.
A type system where variable types are checked during runtime.
The grammar rules of a programming language.
A text-based application for entering commands.
The process of checking whether software works as expected.
Test-Driven Development; writing a failing test first, then writing code to pass it, then refactoring.
A unit of execution within a process.
A category of data, such as string, number, boolean, or object.
A test that checks a small, isolated unit of code.
A named container for storing a value.
A system for tracking changes to files over time.
Creating virtual versions of computing resources, such as servers or operating systems.
A software system that allows applications to communicate over a network.
Extensible Markup Language; a markup language for storing and transporting structured data.
A human-readable data format commonly used for configuration files.
Continuous Integration and Continuous Delivery or Deployment; automated processes for testing, building, and releasing code.
An iterative approach to software development focused on flexibility, collaboration, and frequent delivery.
An Agile framework organized around roles, events, artifacts, and short iterations called sprints.
A fixed time period during which a team works toward a set goal.
A workflow method that visualizes tasks and limits work in progress.
A prioritized list of future work, bugs, features, and improvements.
A short description of a feature from the user's perspective.
Conditions that must be met for work to be considered complete.
A short recurring team meeting to discuss progress, plans, and blockers.
A meeting where a team reflects on recent work and identifies improvements.
A person or group affected by, involved in, or invested in a project.
A high-level plan showing future product or engineering priorities.
The future cost created by shortcuts, outdated code, or suboptimal design decisions.
The process of reviewing another developer's code before it is merged.
A proposed set of code changes submitted for review, discussion, checks, and merging.
To combine code changes from one branch into another.
A situation where Git cannot automatically combine competing code changes.
A saved snapshot of code changes in version control.
Reverting software to a previous version after a problem.
An urgent fix, usually for a production issue.
A new bug that breaks something that previously worked.
A quick basic test to check whether the most important functionality works.
An unusual or extreme situation that may reveal bugs.
The normal successful flow through a product or system.
Explaining a problem out loud to clarify your thinking and find the issue.
Looks good to me; a common approval phrase in code reviews.
Work in progress; used for unfinished tasks, drafts, or pull requests.
Slang meaning something is ready to release.
Something preventing progress.
A small, usually non-blocking comment or suggestion in a code review.
Spending too much time debating minor details.
Doing a chain of side tasks before being able to complete the original task.
Urgent, chaotic work caused by an unexpected issue.
Responsible for responding to production incidents or alerts during a scheduled period.
Receiving an urgent alert about a system issue.
An incident review focused on learning and prevention rather than blaming individuals.
The risk created when too few people understand a system or process.
Important information known informally by people but not properly documented.
A brand-new project with little or no legacy code.
Work on an existing project or system with legacy constraints.
Older code that may be difficult to maintain but still supports important functionality.
Code with tangled, confusing structure that is hard to follow.
An object or class that has too many responsibilities.
A feature or design that makes it easy for users or developers to make mistakes.
The set of technologies used to build and run a product.
Slang for an extremely productive engineer, though the term is often debated or used jokingly.
Startup culture phrase meaning speed is prioritized, sometimes at the expense of reliability.
Improving performance before there is evidence that optimization is needed.
The field of building systems that perform tasks normally associated with human intelligence, such as reasoning, perception, language, planning, and decision-making.
A subset of AI where systems learn patterns from data instead of being explicitly programmed for every rule.
A machine learning approach that uses multi-layer neural networks to learn complex patterns from large datasets.
A computing model made of interconnected units that transform inputs into outputs through learned weights.
A large model trained on broad data that can be adapted to many downstream tasks; not every foundation model is a frontier model.
A highly capable general-purpose AI model at or near the leading edge of current AI capabilities.
AI that creates new content such as text, images, audio, video, code, or synthetic data.
A large AI model trained to predict and generate language, often capable of writing, summarizing, coding, and reasoning.
A smaller language model designed to run with lower cost, lower latency, or on-device constraints.
An AI model that can process or generate more than one type of input, such as text, images, audio, video, or code.
A neural network architecture based on attention mechanisms that powers many modern language and multimodal models, especially because it can model relationships across long sequences efficiently.
A technique that helps a model focus on the most relevant parts of the input when producing an output.
A form of attention where tokens in the same sequence compare against each other to capture relationships.
A unit of text or data processed by a model, often a word, part of a word, character, or symbol.
The process of splitting input text or data into tokens that a model can process.
A numerical representation of data that captures semantic meaning or relationships.
A database designed to store and search embeddings using similarity metrics.
Search that retrieves results based on meaning rather than exact keyword matches.
A measure used to compare how similar two vectors are based on the angle between them.
The maximum amount of input and output text a model can consider at one time.
A model designed to handle very large inputs, such as long documents, transcripts, or codebases.
The input or instruction given to an AI model.
The practice of designing prompts to get more useful, reliable, or structured model outputs.
A high-priority instruction that guides a model’s behavior, role, constraints, or style.
A prompt that defines how another prompt should be interpreted, improved, or executed.
A prompting method that encourages step-by-step reasoning, though many systems now keep detailed reasoning hidden.
A private or intermediate reasoning space used by an AI system to work through a problem.
Providing a model with a few examples of the desired input-output pattern.
Asking a model to perform a task without providing examples.
Training a model on instruction-response examples so it better follows user requests.
Further training a model on a specific dataset to adapt it to a task, domain, or style.
Fine-tuning using labeled examples that show desired outputs for given inputs.
A training method that uses human preferences to help models produce more helpful or aligned outputs.
A method that uses AI-generated feedback instead of, or alongside, human feedback to improve a model.
A preference-training method that optimizes a model directly from chosen-versus-rejected outputs, often without explicitly training a separate reward model.
A model trained to score outputs according to human or AI preferences.
The effort to make AI systems reliably follow intended human goals, values, instructions, and constraints.
The field focused on reducing risks from AI systems, including misuse, accidents, bias, and loss of control.
Practices for developing and deploying AI in ways that are fair, transparent, accountable, secure, and ethical.
Policies, processes, controls, and oversight used to manage AI risks and responsible deployment.
Documentation that explains a model’s intended use, limitations, training context, and evaluation results.
Documentation that describes a dataset’s source, contents, collection process, limitations, and risks.
Tests used to measure model performance, reliability, safety, or behavior on specific tasks.
A standardized test or dataset used to compare model capabilities.
A public or internal ranking of models based on benchmark results.
Massive Multitask Language Understanding, a benchmark that tests knowledge across many academic and professional subjects.
A coding benchmark that evaluates whether models can generate correct solutions to programming problems.
A benchmark that tests whether AI systems can solve real software engineering issues from GitHub repositories.
Stress-testing an AI system by trying to find failures, vulnerabilities, or harmful behaviors.
A prompt or method designed to bypass an AI system’s safety rules or restrictions.
A technical or policy control that constrains an AI system’s behavior.
A confident but false or unsupported output generated by an AI model.
Connecting model outputs to reliable external sources, data, tools, or context.
A method where a model retrieves relevant external information, such as documents or database records, before generating an answer so the response is grounded in non-parametric knowledge.
An emerging RAG pattern where an AI agent plans retrieval steps, chooses tools or sources, evaluates retrieved evidence, and decides when it has enough context to answer.
A RAG approach that uses graph structures, entity relationships, knowledge graphs, or graph-derived summaries to improve retrieval, summarization, and reasoning over connected information.
An advanced RAG variant that evaluates retrieval quality and revises, retries, or filters retrieval when the initial results are weak or irrelevant.
Splitting documents into smaller pieces so they can be embedded, retrieved, or processed by a model.
Reordering retrieved results using a stronger model or scoring method to improve relevance.
Search that combines keyword-based retrieval with semantic or vector-based retrieval.
A structured network of entities and relationships used to represent knowledge.
A formal structure that defines concepts, categories, and relationships in a domain.
An AI system that can plan, use tools, make decisions, and take actions toward a goal.
A model’s ability to call external tools, APIs, databases, calculators, or software functions.
A structured way for a model to request that a specific function or API be executed.
The process by which an AI system breaks a goal into steps or actions.
Breaking a complex task into smaller, manageable subtasks.
An AI agent that can operate with limited human intervention over multiple steps.
A system where multiple AI agents collaborate, debate, specialize, or coordinate.
Coordinating tools, workflows, memory, and agents to complete complex tasks.
Stored information that helps an AI system preserve context, preferences, facts, or previous interactions across steps or sessions, subject to product and privacy controls.
Memory of specific events, interactions, or past experiences.
Memory of general facts, preferences, or concepts rather than specific episodes.
An open protocol for connecting LLM applications and AI assistants to external tools, data sources, business systems, and development environments in a standardized way.
The process of running a trained model to produce outputs from inputs.
The process of adjusting a model’s parameters using data so it learns patterns.
Initial large-scale training on broad data before task-specific adaptation.
Additional training after pretraining, often to improve instruction-following, safety, or preferences.
A learned numerical value inside a model that affects how it processes inputs and generates outputs.
A training or configuration setting chosen by developers rather than learned directly by the model.
A generation setting that controls randomness; lower values usually make outputs more deterministic, while higher values make outputs more varied.
A generation method that samples from the smallest set of likely tokens whose combined probability exceeds a threshold.
A decoding method that keeps several likely output candidates at each step.
The time it takes for an AI system to respond after receiving input.
The amount of work an AI system can process in a given time.
Processing multiple inputs together to improve efficiency.
Reducing the numerical precision of model weights or activations to make models smaller or faster.
Training a smaller model to imitate a larger model’s behavior.
Removing less important model weights or connections to reduce size or computation.
A model architecture that routes inputs to selected expert subnetworks instead of activating the entire model for every token.
A model that uses sparse weights, sparse activations, or conditional computation so only part of the model is active for a given input.
A model that generally uses all or most of its parameters for each input.
Graphics Processing Unit; hardware widely used to accelerate AI training and inference.
Tensor Processing Unit; specialized hardware designed to accelerate machine learning workloads.
Floating-point operations per second, a measure of computational performance.
The computational resources needed to train or run AI models.
The process of annotating data with correct answers or categories for training or evaluation.
Artificially generated data used for training, testing, simulation, or privacy-preserving workflows.
Creating modified versions of training examples to improve model robustness.
When a model learns training data too closely and performs poorly on new data.
When a model is too simple or poorly trained to capture important patterns in the data.
A model’s ability to perform well on new data it has not seen before.
Techniques used to reduce overfitting and improve generalization.
An optimization method that updates model parameters to reduce error.
The algorithm used to calculate how model parameters should change based on errors.
A mathematical measure of how wrong a model’s prediction is.
One full pass through the training dataset.
A hyperparameter that controls how large each training update is.
Predicting which category an input belongs to.
Predicting a continuous numerical value.
Grouping similar data points without predefined labels.
Training a model using examples that include correct labels or answers.
Training or analyzing data without explicit labels.
A training method where labels are created from the data itself.
A learning approach where an agent improves by receiving rewards or penalties for actions.
Training a model or agent to copy expert behavior.
An internal model that predicts how an environment changes in response to actions.
A generative model that learns to create data by reversing a gradual noise-adding process.
A generative model with a generator and discriminator trained against each other.
A generative model that learns compressed latent representations and can generate new samples.
A compressed mathematical space where meaningful features of data are represented.
The AI field focused on interpreting images and videos.
The AI field focused on understanding, generating, and analyzing human language.
AI that converts spoken language into text.
AI that converts written text into spoken audio.
AI-based creation of new images from prompts, references, or other inputs.
AI-based creation or editing of moving visual content.
AI that imitates a person’s voice from audio samples.
Synthetic media that realistically imitates a person’s face, voice, or actions.
Adding visible or invisible detectable signals to AI-generated content to indicate origin, though watermarking is not always foolproof.
Information about where content came from, who or what created it, and how it was modified over time.
Methods that make AI decisions easier for humans to understand.
The degree to which humans can understand how a model works or why it produced an output.
Research that tries to reverse-engineer neural networks by understanding internal circuits and representations.
Systematic unfairness or skew in data, model behavior, or outputs.
The goal of reducing unjust or discriminatory outcomes in AI systems.
Techniques that train or use models while reducing exposure of sensitive data.
Training models across decentralized devices or servers without centralizing raw data.
A mathematical privacy technique that limits what can be inferred about any individual in a dataset.
Encryption that allows computation on encrypted data without decrypting it first.
An attack that tries to reconstruct training data or sensitive features from a model.
An attack where malicious data is inserted into training data to manipulate model behavior.
An attack that uses malicious instructions in user input or retrieved content to override intended model behavior.
A prompt injection attack hidden in external content that an AI system reads, such as a webpage or document.
An input intentionally designed to fool a model.
Degradation in model performance when real-world data changes over time.
A change in the input data distribution compared with the data used during training.
A change in the relationship between inputs and outputs over time.
Tracking AI system performance, reliability, costs, safety issues, and user outcomes after deployment.
The ability to understand system behavior using logs, metrics, traces, and evaluations.
A design where humans review, guide, approve, or correct AI system outputs.
A human-in-the-loop review step where a person approves, corrects, or rejects AI outputs before they are used.
An AI assistant that helps a person complete tasks while the human remains in control.
A conversational software system that interacts with users through text or voice.
A structured process that uses AI models, tools, data, and human review to complete work.
Designed around AI capabilities from the beginning rather than adding AI later.
A product that mainly packages or lightly customizes an existing AI model or API.
A system that chooses which AI model to use for a request based on cost, quality, speed, modality, risk, or task type.
A system that tries cheaper or faster models first and escalates to stronger models when needed.
Operational practices for deploying, evaluating, monitoring, securing, and maintaining LLM applications in production.
Operational practices for deploying, monitoring, and maintaining machine learning systems.
Emerging industry lingo for versioning, testing, evaluating, approving, and managing prompts in production AI systems.
Industry lingo for search optimization practices aimed at AI-powered search experiences, answer engines, AI summaries, and generative results; it is best viewed as an extension of SEO rather than a completely separate discipline.
Search-marketing lingo for optimizing content so it can be selected as a direct answer by search engines, assistants, or answer engines.
Emerging search-industry lingo for optimizing content for visibility in AI-generated search responses and generative answer engines; the term is not universally standardized.
Emerging industry lingo for optimizing content, data, brand presence, and technical signals so LLMs can retrieve, understand, and mention information accurately.
AI-generated summaries or answer blocks shown in Google Search results as part of Google’s AI search features.
A search result where the user gets the answer without clicking through to a website.
Improving how search engines and AI systems understand a brand, person, place, or concept as an entity.
Perceived expertise built by publishing comprehensive, high-quality content around a topic area.
SEO focused on meaning, entities, context, and relationships rather than only keywords.
Machine-readable markup that helps search engines understand page content.
A type of structured data vocabulary used to describe webpage entities and content.
Search and AI-content lingo for content that is clear, authoritative, specific, original, and well-supported enough that AI systems or search engines are more likely to reference it.
The unique value or new information a piece of content adds compared with existing results.
Tools or methods that estimate whether content was generated by AI, usually probabilistically and with possible false positives or false negatives.
AI-assisted content edited by humans for accuracy, clarity, originality, usefulness, and voice; it should not imply deception or guaranteed detector avoidance.
Experience, Expertise, Authoritativeness, and Trustworthiness; a Google Search quality framework where trust is especially important.
A search or AI retrieval pattern that expands one user query into multiple related subqueries to gather broader or more precise evidence.
A reference or mention of a source inside an AI-generated answer.
The collection of documents or data that a retrieval system searches.
Emerging AI product and engineering lingo for designing the information, tools, memory, retrieval, constraints, and state available to an AI system.
Connecting multiple prompts so the output of one step becomes the input to the next.
A component that converts model output into a required format such as JSON or a database record.
Model output constrained to a specific format, schema, or data structure.
A model or API mode intended to produce valid JSON, though it may not guarantee adherence to a strict schema unless combined with structured output features.
A research-inspired term for training or prompting models to decide when and how to use external tools.
A prompting pattern that combines reasoning and acting, often allowing models to alternate between thoughts and tool use.
A reasoning method that explores multiple possible solution paths before choosing an answer.
A method that samples multiple reasoning paths and chooses the most consistent answer.
A training approach that uses a written set of principles to guide model behavior, often including supervised and reinforcement-learning-style feedback phases.
Research on supervising AI systems when their actions, outputs, or reasoning are too complex or numerous for humans to directly review unaided.
A reusable system for running model evaluations across datasets, prompts, and metrics.
A trusted set of examples used to evaluate model quality over time.
An evaluation designed to catch behavior that got worse after a model or prompt change.
A classifier or rule system used to detect and block unsafe inputs or outputs.
A model used to classify content for policy violations such as hate, self-harm, or sexual content.
Evaluation using AI-generated test cases or labels.
Data showing which outputs people prefer, often used for model alignment or ranking.
Emerging lingo for the extra cost, latency, complexity, or product friction added by safety, alignment, evaluation, or governance measures.
A relationship describing how model performance changes with more data, compute, or parameters.
A capability that appears or becomes much stronger at a certain model scale or training setup, though the term is debated because some emergence can depend on measurement choices.
A training phenomenon where a model suddenly generalizes after long periods of memorization-like behavior.
A model’s ability to learn or adapt from examples provided in the prompt without updating its weights.
Extra computation used during inference to improve answer quality, such as search, reasoning, or sampling.
Product and research lingo for a model optimized to spend more computation on multi-step reasoning, planning, problem solving, or deliberation.
A model or system that checks whether another model’s answer is correct or acceptable.
A model used to evaluate, critique, or improve another model’s output.
A technique where a model reviews and improves its own output.
A training strategy where the system selects the most useful examples for humans to label.
Training models over time on new data while trying to avoid forgetting old knowledge.
When a model loses previously learned abilities after being trained on new data.
Parameter-Efficient Fine-Tuning; adapting models by training only a small subset of parameters or adapters.
Low-Rank Adaptation; a PEFT method that adds small trainable matrices to adapt a model efficiently.
A small trainable component added to a model to adapt it to a task.
Combining LoRA adapter weights into a base model for deployment.
Training a small set of prompt-like parameters while keeping the main model fixed.
The time required to fetch relevant information before generation.
Informal emerging lingo for degraded model performance when too much irrelevant, stale, conflicting, or low-quality context is included.
A tendency for models to use information at the beginning or end of long context better than information buried in the middle.