Traction Isn’t Validation: Why Your Clients Can’t Tell You Everything

This week a founder came to us to show off their new tool.

It was polished. You could tell AI had done a lot of the building, and it had done it well: a clean interface, a long feature list and a demo that ran without a hitch. The founder had customers too, and some traction in the market. By most of the measures a startup is taught to chase, they were doing well.

So we asked a simple question: how many credible organisations have tested the platform to check it does everything it says it does?

There was a pause, and then a few examples.

That isn’t a criticism of this founder. It’s the most common answer we hear, and it points to a bigger problem.

Shiny tools, thin evidence

A wave of good technology and real creativity is coming into the market. AI coding tools mean a capable founder can now build in a weekend what used to take a team six months. That’s a good thing. More people can build, and more ideas get a chance.

But building something and proving it are not the same job.

Many of the tools arriving now have traction. They have early customers, a growing network and some word of mouth. What they rarely have is evidence that the product holds up under pressure:

• when it is used every day, not just in a demo
• when the person using it isn’t the person who bought it
• when a regulator, a lawyer or a procurement team looks closely
• when the early enthusiasm wears off after three months

Traction tells you people are willing to try something. It doesn’t tell you they’ll keep using it, that it saves them time, or that it is safe to roll out across an organisation. A few customers who like you are not the same as practitioners across your industry who have tested you.

Why validation goes missing

The honest reason is cost. Proper research and development is a luxury. Large companies have labs, test teams, compliance departments and pilot budgets. Universities have research groups and ethics boards. A founder with a laptop and a small round of funding has none of these.

So founders do the sensible thing with what they have. They build, sell to whoever will buy, and treat sales as proof. Investors often reward that, because traction is easy to see and easy to compare. The stress-testing gets pushed back to “after the next round”, and sometimes it never happens.

The cost of skipping it shows up later. A product that looked strong in the pitch deck meets a real workflow, a real regulation or a real user with no patience, and it breaks. By then the money has been spent.

What validation actually looks like

Validation doesn’t have to mean a lab. It means putting the product in front of the people who will use it, regulate it and pay for it, and listening to what they say, before you scale.

For a health tool, that’s the nurse on a night shift, not just the clinical director who signed the trial. For a legal tool, it’s the paralegal and the compliance officer, not just the partner who liked the demo. For a smart city tool, it’s the residents and the council officers who will live with it.

There’s a second blind spot, and it changes what validation means. It isn’t only about your clients, or about another organisation signing off a pilot. A client can tell you whether the product is useful to them. An organisation can tell you whether it fits their setup. Neither can tell you how it fits your whole industry. A tool can delight its clients and still fail a compliance review, clash with the way a sector buys, or fall apart in a different city.

Validation means hearing from practitioners in different roles across your industry and domain, each answering a different question:

• Sector: how does it fit the way this sector actually works, and the way it buys?
• Regulation: which rules apply, and does it meet them?
• Location: does it work here, for this community, and not just where it was first tried?
• Academia, where it helps: what does the research already say, and does the evidence hold up?

No founder understands every one of those levels, and nobody expects you to. In a startup you tackle what’s in front of you: the next customer, the next release, the next round. That’s normal. It’s also why the views you don’t have are the ones most likely to catch you out.

And it means measuring the things that predict real adoption, not the things that feel good:

• Are people still using it after 90 days? Not signed up: still using it.
• Would they recommend it to a colleague? Unprompted, not because you asked for a testimonial.
• Does it save measurable time? Hours a week, measured rather than estimated.
• Is someone paying for it under a signed contract? Not a free pilot, and not a favour.

Some ideas won’t pass. That’s fine. Finding out early, at a workshop table rather than after launch, is the cheapest result a founder can get. A clear “not yet” or “not this” saves time and money that would otherwise go into building the wrong thing.

Five questions to ask before you scale

If you are building something now, these are worth answering honestly:

1. Who, outside your own team and your friendliest customer, has tested this in their real work?
2. Have you heard from the people who use it every day, or only from the people who bought it?
3. What would a regulator, a lawyer or a procurement lead say if they looked closely?
4. What happens to usage after the first three months?
5. If an investor asked for evidence rather than traction, what would you show them?

If those are hard to answer, you’re not behind. You’re in the same position as most founders, and the answers are within reach.

Making validation accessible

That second blind spot is the reason we started the UKBT Institute. If proper R&D is a luxury, the answer isn’t to skip it. The answer is to make a version of it that founders can actually reach.

The Institute’s Townhouse model brings different practitioners to the same table: daily users, engineers, academics, legal and regulatory experts and budget holders, from the sector and the place where the product will be used. Each stage ends in a clear decision to go ahead, change course or stop. It is designed to be accessible and affordable, so the founders who most need evidence aren’t the ones least able to pay for it.

The founder we met this week has a good product and real momentum. What they need now isn’t more features or more customers. It’s evidence.

Great technology is arriving every week. The next step is proving it works.

Building something and want to test it with the people who will use it? Get in touch with the UKBT Institute.