What Is Market Validation? A Practical Guide for B2B Offers

A lot of new offers fail for a dull reason. Not enough people wanted them at that price.

The data backs this up. When CB Insights looked at 431 venture-backed companies that have shut down since 2023, it could find a reason for 385 of them. Poor product-market fit was cited in 43% of those. Running out of money topped the list at 70%, but as the report puts it, that's "almost always the final cause of death, not the root problem."

Market validation is how you find out before the money's spent. I wrote this for two kinds of reader. Founders and commercial leads at established B2B companies who are launching, repricing or repackaging an offer. And founders at the idea stage who want to know if anyone will pay. The method is the same. The stakes look a bit different.

What Market Validation Is

Market validation is a set of tests that checks three claims, in order. First, that a specific group of buyers has a problem worth solving. Second, that your offer solves it better than whatever they do today. Third, that they'll pay the price you need to charge. Each claim can fail on its own, and a strong answer to the first tells you nothing about the other two.

Plenty of things feel like validation and aren't. A survey where people say they like the idea. A big top-down market figure from an analyst report. Warm reactions from friends, existing fans or your own sales team. They're pleasant, and they rarely predict sales.

Product people have a tidy way to frame this. Marty Cagan of SVPG calls it value risk: "whether customers will buy it or users will choose to use it." Market validation is the work of shrinking that one risk before you spend on the others.

Market Validation vs Idea, Offer and Product Validation

The four terms get used as if they mean the same thing. They ask different questions, and mixing them up is how teams end up answering the wrong one.

If you're at the idea stage, start with our startup idea validation framework, then come back here. If you already sell something and you're adding a new service line, a premium tier or a price rise, you need offer validation. That's market validation narrowed to one offer, one segment and one price, and it's the work our Offer Validation service is built for.

Why Market Validation Matters

Skipping validation is expensive because the cost lands late. You find out after the build, the launch campaign and a quarter of sales effort.

Harvard Business School's Tom Eisenmann spent years studying failed start-ups. In his 2021 HBR article he names a pattern he calls "false starts", where founders rush to launch and skip a step: "Many overlook a crucial step in the lean start-up process: researching customer needs before testing products." The fix he describes is "validating concepts with real customers in real-world settings."

For an established company the maths is different but the shape is the same. You won't run out of runway over one bad offer. You'll lose a quarter of pipeline, some credibility with the sales team, and the option to try something better in that slot. We go through the failure figures in more depth in what the 43% really means.

Steve Blank makes the case in two minutes. His point is that the answers you need aren't in your office.

Play Video (Loads YouTube)

How to Do Market Validation: 6 Steps

Market validation works best as a short sequence with a decision at the end. Here's the one I'd use for a B2B offer. Each step can kill the offer, and that's the point.

1. Write the hypothesis

Write down who has the problem, what you'll sell them, at what price, and what result would prove you wrong. Do it before you collect any evidence. "Operations directors at UK logistics firms with 50 to 249 staff will pay £X a month to stop fleet-scheduling errors, and if fewer than three of the first twenty agree to a paid pilot, we stop." That's testable. "SMEs need better scheduling" isn't.

The last clause matters most. If you don't set the bar now, you'll move it later.

2. Confirm the problem

Talk to people in the target segment about how they handle the problem today. What did they do last time it happened? What did it cost? What have they already tried and dropped? Ask about the past. What someone did last time is a far better guide than what they think they'll do next time.

Expect to be surprised. One founder building a tool for people who run paid ads described the moment the interviews turned their thinking around:

That's what good problem interviews do. They don't confirm your plan. They tell you which problem the buyer would actually pay to make go away. If you want a practical guide to running these conversations, YC's Gustaf Alströmer covers who to talk to, what to ask and which questions to avoid in this Startup School talk.

Play Video (Loads YouTube)

3. Size the segment from the bottom up

Count the companies that actually match your target, then estimate what they spend on the problem today. Top-down figures ("the global market is worth $80bn") tell you nothing about whether you can reach or sell to anyone in it.

In the UK, the official numbers are a good place to start. The government's business population estimates for 2025 count 5.7 million private sector businesses. But 4.3 million of those (75%) don't employ anyone besides the owners. If you sell to companies with real teams and budgets, your pool is far smaller than "UK SMEs" suggests.

Add the three bands together and you get about 267,000 businesses with ten or more staff, out of 5.7 million. That's the real starting pool for most B2B offers. From there, narrow by sector and region. Companies House advanced search lets you filter active companies by SIC code (nature of business), registered office location and incorporation date. It won't tell you headcount, so pair it with the size bands above or with LinkedIn. You end up with a count you can defend, and often a starter list of names to call.

4. Map the alternatives

List what buyers use now. That includes a competitor, a spreadsheet, an in-house person, an AI tool, and doing nothing at all. Note what each one costs them. "Doing nothing" is the competitor most new offers lose to, and it's free.

5. Test willingness to pay

Ask for a commitment that costs the buyer something: a paid pilot, a deposit, a letter of intent or a signed order. Stated interest doesn't count. The research backs this up. Vicki Morwitz, then at NYU Stern, puts it plainly in her 2014 review: "Purchase intentions are correlated and predict future sales, but do so imperfectly."

Jason Cohen, who founded Smart Bear, put the practical version bluntly back in 2010: "When ten people say they'll give you money, that's the only validation that counts." His essay on customer validation is old, and the bar still holds. Our guide to testing an offer or price change covers the ways to run this test on a live offer without upsetting existing customers.

6. Decide against your criteria

Go back to the bar you set in step 1 and compare. The honest answers are go, go with conditions, change course, stop, or wait. If the evidence is mixed, say which part is weak and what test would settle it. Don't round a maybe up to a yes because the team has already started. Stopping is a legitimate result, and I explain why in why many ideas should be stopped before they're built.

What Counts as Evidence, and What Doesn't

Strong evidence is something a buyer did or gave up. Weak evidence is something a buyer said, or something you assumed. The table is the quick version.

That last row deserves a word. It's now very easy to paste an idea into a chat assistant and get a confident, well-written market analysis back. It reads like validation. It isn't, because the tool hasn't spoken to a single buyer, and chat assistants tend to agree with whoever's asking. I wrote about that problem in why ChatGPT says your idea is brilliant.

Rob Fitzpatrick, who wrote The Mom Test, answered a related question in June 2026: can AI run your customer interviews for you? His short answer in the video description is "run them, no. Rehearse them, potentially. Help around the process, definitely." It's seven minutes and worth it.

Play Video (Loads YouTube)

If you grade evidence on a scale, keep it simple and write the grade next to the claim. We use A to E, and the scale is published on our methodology page. A paid pilot is near the top. "Our sales team thinks so" is near the bottom.

Validating a New Offer Inside an Established Company

If you already have customers, you have a head start most founders would envy. You can test a new offer with people who already trust you, and you have records of what they've asked for.

Start with what customers already tell you without being asked. Jason Cohen made the point about his first company in 2025:

Support tickets, lost-deal notes, renewal conversations and the questions prospects ask on sales calls are all evidence about real behaviour. Read them before you write a single survey question.

There's a catch, though. Existing customers are a biased sample. They already like you, and they're not the new segment you might be trying to reach. And new buyers are harder to get on a call than they used to be. Gartner reports that 75% of B2B buyers prefer a rep-free sales experience. So pair conversations with tests that don't need a meeting: a priced landing page, a pilot offer to a defined list, or a paid add-on shown to a slice of renewals.

The trap I'd watch for most in an established company is treating the sales team's enthusiasm as demand. Sales teams are optimistic by trade, and they're often the ones who asked for the new offer. Their view is a hypothesis. Buyer commitments are the test.

Market Validation in a Pitch Deck

Investors look for commitments, not enthusiasm. The strongest validation slide shows paying customers or pilots, signed letters of intent, retention if you have it, a bottom-up market size with sources, and two or three short quotes from interviews. Every figure needs a named, dated source, because investors check.

One more founder story shows why sources matter as much as the numbers. A founder building a B2B workflow automation product posted in June 2026 that they'd spent 14 months shipping features before looking properly at the market around them:

An investor will spot that gap in minutes. Our guide to how investors verify startup claims shows what they test first, and it's usually the market slide.

Where ThriveFinity Fits

We sell Offer Validation as a structured, evidence-graded read on one offer. It suits established B2B companies deciding whether to launch, reprice or repackage, and founders whose idea is concrete enough to describe as an offer.

• Offer Signal (free, under an hour): describe the offer in plain English and get a directional Go, Go With Conditions, Change Course, Stop or Wait answer with your single biggest risk named. It's automated and unsigned. Get a free Offer Signal.
• Build Verdict (£149, within 24 hours): a named verifier grades the offer on 12 evidence lenses from A to E, runs a stress test, sets stop criteria and signs the verdict. It carries our 30-Day Honest Verdict Guarantee.

What neither of these does is talk to your buyers for you. We can tell you where the evidence is weak and which test to run next. Only your buyers can tell you if they'll pay. If you need a wider plan for the move, rather than a verdict on one offer, our Decision Brief is the better fit. And if you're still exploring which problem to solve, you're not ready for either. Do the interviews first.

If you're comparing options, our comparison of idea and offer validation tools sets out prices, methods and guarantees side by side, including where other tools are the better choice.

Pranav Unni

Founder · ThriveFinity Connect on LinkedIn →

Pranav Unni is the founder and lead verifier of ThriveFinity. He reads and signs every paid deliverable personally, and writes about go-to-market decisions for established B2B companies.