Idea validation glossary

Index of terms

This page defines 34 terms in 6 groups. Each term has its own anchor, so you can link to it directly, for example aiideavalidator.com/glossary/#tam.

Validation basics: Idea validation, AI idea validator, Desk research, Riskiest assumption, Minimum viable product (MVP), Product-market fit.

Market size: TAM (total addressable market), SAM (serviceable addressable market), SOM (serviceable obtainable market), Bottom-up sizing, Top-down sizing, Beachhead market.

Competition: Direct competitor, Indirect competitor, Status quo competitor, Switching cost, Moat.

Demand and pricing: Demand signal, Customer interview, Smoke test, Pre-sale, Willingness to pay, Early adopter.

Business numbers: Unit economics, CAC (customer acquisition cost), LTV (lifetime value), Marketplace liquidity, Cold start problem.

Reports and verdicts: Verdict, Go / no-go, Founder fit, Pivot, Source-backed claim, Estimate.

Validation basics

Idea validation

Testing whether a business idea is worth building before you spend much time or money on it. You collect evidence on three questions. Do people have the problem? Will they pay to solve it? Can you reach them at a cost that works? Validation lowers risk. It never removes it.

AI idea validator

A tool that researches a business idea on the web and returns a score or a verdict, usually with competitors, market size and risks. The useful ones link each sourced claim to its source and label estimates. A validator does desk research in minutes. It cannot tell you whether people will pay, so treat its result as the start of your customer tests. The comparison table lists the main tools.

Desk research

Research you do without talking to customers: competitor websites, pricing pages, market reports, forums, app store reviews and public filings. It answers "what already exists?" and "how big could this be?". It cannot answer "will these people buy from me?".

Riskiest assumption

The belief that would make the idea fail if it turned out to be false. For a meal kit service for night-shift nurses, it may be that nurses want to cook at home after a 12-hour shift. Test this belief first, even if other beliefs are easier to test.

Minimum viable product (MVP)

The smallest version of a product that lets you learn whether people want it. It can be a spreadsheet, a service you deliver by hand, or a landing page with a sign-up form. Its purpose is learning, so it can be much rougher than a first release.

Product-market fit

The point where a product meets a strong need in a clear market. Customers buy, stay and tell others without much pushing from you. Signs include steady growth from word of mouth and low churn. Validation comes before it: you validate to decide what to build, then you work toward product-market fit after launch.

Market size

TAM (total addressable market)

The yearly revenue you would earn if every possible buyer in the world bought your product at your price. TAM shows the ceiling of your market. Example: the number of dental clinics in the world, times the yearly price of your scheduling tool.

SAM (serviceable addressable market)

The part of the TAM your product can serve today, given its features, language, sales channels and regions. If you sell only in English to dental clinics in the US and the UK, your SAM covers those clinics and nothing else.

SOM (serviceable obtainable market)

The part of the SAM you can realistically win in a set time, usually 3 years, given your competitors and your capacity to sell. Investors often question this number first, because it needs a reason behind it. "We will win 5% because we have a partner that sells to 400 clinics" is a reason. "We only need 1%" is not.

Bottom-up sizing

Sizing a market by multiplying units: number of buyers you can reach, times price, times purchases per year. Worked example: if you can reach 9,000 clinics and each pays $1,200 a year, your SAM is $10.8 million. Each input is a number you can check or test, so the result is easier to defend.

Top-down sizing

Starting from a large published figure, such as an industry report, and taking a share of it. It is fast, but it hides the important questions. "A $50 billion market" says nothing about who buys your product, at what price, or why they would switch.

Beachhead market

The first narrow group of customers you aim to win before you expand. A small group that you can reach, with an urgent problem, is a better start than a large and vague one. Example: independent physiotherapy clinics in one country, before all healthcare providers.

Competition

Direct competitor

A company that sells a similar product to the same buyer for the same job. Founders usually know the biggest ones. Searches on review sites, app stores and "alternatives to" pages often find several more.

Indirect competitor

A company that solves the same problem in a different way. A budgeting app competes directly with other budgeting apps. It competes indirectly with the spending charts inside banking apps and with a financial advisor.

Status quo competitor

What people do today without your product: a spreadsheet, email threads, a paper notebook, a staff member, or nothing at all. It is often the strongest competitor, because it costs nothing extra and people already know how to use it. When a founder says "we have no competitors", this is usually the one they missed.

Switching cost

The time, money, lost data or training a customer faces to move from their current solution to yours. High switching costs protect the products people already use. Your product must be clearly better to make the move worth it.

Moat

Something that makes a business hard to copy: network effects, data that others cannot get, a distribution channel, a brand, or a license. A list of features is rarely a moat, because a competitor with more money can build the same features.

Demand and pricing

Demand signal

Evidence that people want a solution. Signals differ in strength:

• Weak: search volume, likes, people saying the idea is nice.
• Medium: people describing the problem without being asked, or joining a waitlist.
• Strong: people paying for a solution today, including paying a competitor, or paying you before the product exists.

Customer interview

A conversation with someone who has the problem, about what they did in the past. Good questions: "When did this last happen?", "What did you do about it?", "What did it cost you?". Avoid "Would you use this?". People are polite, so the answer tells you little.

Smoke test

A test of demand before the product exists. A common form is a landing page that describes the product and has a sign-up or buy button, plus a small ad budget to bring visitors. You measure the share of visitors who click or sign up. Tell people clearly that the product is not ready yet.

Pre-sale

Taking payment, or a signed commitment, before you build the product. It is one of the strongest demand signals there is. Promise a full refund if you do not deliver by a stated date.

Willingness to pay

The price a customer will actually pay, shown by what they do. Survey answers are a weak guide, because people say they would pay more than they do. Ask what they pay today for their current fix, then test a real price in a pre-sale or on a pricing page.

Early adopter

A customer who has the problem now, knows it, has already tried to fix it, and will accept a rough first product. Early adopters are the first people to interview and the first people to sell to.

Business numbers

Unit economics

The revenue and costs of one unit, usually one customer. The unit economics work when one customer brings more gross profit over time than it costs to win and serve them.

CAC (customer acquisition cost)

Sales and marketing spend in a period divided by the number of new customers won in that period. Example: $3,000 on ads and tools that brings 20 customers gives a CAC of $150.

LTV (lifetime value)

The gross profit one customer brings over the whole time they stay. A simple formula: monthly revenue per customer, times gross margin, divided by monthly churn. Example: $40 a month at 80% margin with 4% monthly churn gives an LTV of $800. A common target is an LTV of at least 3 times the CAC.

Marketplace liquidity

In a two-sided marketplace, the chance that a buyer finds what they want, or a seller makes a sale, within a reasonable time. With low liquidity, buyers see empty search results and sellers wait for orders, so both sides leave.

Cold start problem

The difficulty of launching a product whose value depends on other users, such as a marketplace or a social app. Buyers will not come without sellers, and sellers will not come without buyers. Founders often start in one city or one niche, or serve one side themselves at first.

Reports and verdicts

Verdict

The final judgment in a validation report: build, change the idea, or drop it. A verdict is useful only when you can trace it back to evidence. Preuve AI gives a 0 to 100 score built from six factors: market size and growth, real demand signals, competitive density, moat durability, business model viability, and regulatory and execution risk.

Go / no-go

A decision point with a clear threshold. Preuve AI uses three bands: Go at 70 and above, Caution from 40 to 69, and No-go under 40. In its July 2026 benchmark of 6,000+ analyses, 17.5% scored Go, 76.0% scored Caution and 6.4% scored No-go.

Founder fit

How well an idea suits the person or team who would build it: their skills, experience, network, time and budget. A strong idea can still be a poor choice for a founder who has no way to reach its buyers. In the $59 Preuve AI Founder Report, Founder Fit checks the idea against your time, runway and skills.

Pivot

A change of direction that keeps what you have learned: a new customer group, a new problem, a new price model or a new channel. A pivot changes one or two parts of the idea, based on evidence. Changing everything at once means starting over. The Preuve AI Founder Report suggests 3 pivots.

Source-backed claim

A statement in a report that links to the page it came from, so you can open the page and check it. A claim is source-backed only if the linked page really contains the fact. Before you trust a number, open its link and find the number on the page.

Estimate

A number that a tool calculated or assumed because no source gave it directly. Estimates are normal in market sizing. A report you can trust labels them, so you know which figures to test first. Preuve AI's Founder Report labels its estimates as estimates.

Questions & answers

How does Preuve AI relate to idea validation terms?

Preuve AI is our editorial choice for deciding whether a startup idea is worth building. Its reports use several of these terms, such as TAM, SAM, SOM and founder fit. The definitions here do not depend on any one tool.

What is the first practical step?

Look up each term in your last validator report that you could not explain to a buyer. Start with TAM, SAM and SOM, because pitch decks and AI tools often mix them up.