How to Research an Audience You've Never Served

When Tallo set out to reach an audience it had never served – people right at the start of their careers, rather than the users it already understood – the hard part wasn't building anything. The hard part was that nobody on the team had any instinct for these people. Every read on what they wanted was a guess wearing the costume of a fact.

Researching an audience you already serve and researching one you've never served are different jobs that happen to share a name. For the audience you know, you have years of accumulated pattern to check a new idea against. For an audience you don't, you have nothing to catch a wrong assumption before it becomes a roadmap. And the whole task, the entire risk, comes down to one thing: telling genuine market demand apart from polite interest.

Why a new audience is so easy to misread

The default failure isn't laziness. It's that the signals you get from people you don't understand are systematically flattering. One analysis of B2B validation found that 78% of technical founders conflate product interest with market demand – they hear "that sounds useful" and file it as "they'll buy." Those are not the same signal, and the gap between them is where the money goes.

It gets worse the more senior and polite the room. Decision-makers are trained to be encouraging, especially to someone earnestly solving a problem they'll agree is real. That politeness inflates your confidence while telling you almost nothing about whether anyone will actually change what they do. The same analysis notes that implementation cost often runs 300 to 500% of the subscription price – so a warm "yes, I'd use that" from someone who has no idea what adopting it would actually take is worse than useless. It feels like validation and predicts nothing.

This is the mechanism behind the single most common way companies fail. CB Insights puts "no market need" at the top of its post-mortem list, at 42% – teams that built something people were happy to say nice things about and unwilling to actually pay for. For an audience you already know, your instincts catch some of that. For one you don't, nothing does, unless you go looking for it on purpose.

What "on purpose" looks like

The good news is that the fix is cheap, and it's the opposite of the instinct to go build something and see who shows up. Five to eight real conversations with the exact people you're trying to serve will tell you more than a month of building in the dark – but only if you run them to disprove your idea, not to collect encouragement for it.

That means a few deliberate moves. Talk to the person who actually feels the problem and can act on it, not whoever is easiest to reach and most willing to be nice. Ask about what they do today and what it costs them, not whether they like your solution – behavior is evidence, enthusiasm isn't. And decide, before you start, what a real signal would look like and what would prove you wrong, so you can't quietly reinterpret every polite nod as a green light. For an audience you've never served, "validated" doesn't mean people said they'd use it. It means you found the specific, repeated evidence that this problem is painful enough for these particular people to change what they're doing – and you went looking for the evidence that it wasn't, and didn't find it.

None of that requires building the thing first. That's the point. When you have no muscle memory for an audience, the cheapest and most honest research you can do is the conversation you have before a line of code exists – the one where you're actively trying to talk yourself out of the idea, and the market won't let you.