Image by Tima Miroshnichenko

You have a startup idea. It’s brilliant. You Googled it extensively, and made sure it hasn’t been done yet. “Excellent!”, you think. Now the question is… Should you quit your job to pursue it? If this dilemma is playing out in your mind, keep reading. By the end of it you might not have a definitive answer, but you should have, at the very least, an idea of what other key questions you should be trying to answer instead.

---

Most startups fail because people start with the wrong motivation. “No one has done this before, therefore I will succeed“. Most advice then tells you to go and check the market size, and if it’s big enough, go for it. But then how do you know there is a market if your idea (or an idea similar to yours) is not making money yet?

In reality, a startup idea is worth pursuing when you can answer two questions with clear evidence:

1. Are the people you think will buy it already trying to solve this problem?

2. Will they give you something valuable (time, money, or data) to have it solved properly?

If the answer to both is a clear yes, then keep going.

If you genuinely don’t know, then you have work to do–and no, don’t drop everything to do it yet.

And, if the genuine answer is that nobody is trying to solve it today, then no market size number or hope can save you. Instead, take the lack of similar products in the market as a sign that this is not a company worth building.

And what about the market size checking? Well, I would put that data in the bottom of the list of things you should check, after you have gathered evidence of yes for questions 1 and 2 above.

So instead of wasting your time quitting your job and building a pitch deck that is guaranteed to be ignored by investors, here are the tests you should be running instead.

Question 1: Why hasn’t anyone done this already?

Start with this one, and be honest about the answer.

Most first time founders will treat “there is nothing like this out there” as proof of an opportunity, but, usually it is proof of one of the following:

• You might have spotted something real and unique others have ignored, yes. Or…
• Someone already tried it and it didn’t work. Which is most likely the case.

Most failed companies won’t show up in a Google search, so an empty market will look like the land of opportunities, when it’s actually more often than not a desert people have already abandoned.

Both answers are meaningful data, but you only get to find out which one is true by going out and asking real people. Try your best to look for the graveyard: search for products that did this five, ten, fifteen years ago, ask other people in the industry, read the online forums where your future customers are likely to be on and complain. If you find remains, work out why they are there before assuming you will do better.

Question 2: Is this a real problem, or a mild inconvenience?

If there’s one key life lesson that my time spent in UX research taught me, it’s this: the vast majority of people will prefer to put up with a bad process they are familiar with, rather than learn a good one from scratch.

Making people change behaviour is one of the hardest thing to accomplish in the product industry. Internalise this now if you are really planning on going down the startup route.

The strongest evidence for this is proof that your future customers are already spending something on this. Money is the clearest signal, but time and elaborate workarounds count too. If someone is trying to maintain a complex spreadsheet, or paying a contractor to do it, or even building their own makeshift solution, they have essentially told you the problem is real and worth solving. Someone who says “oh yes, that would be nice” is being agreeable and polite, and is not evidence.

So don’t go asking them IF this is a problem for them, go ask them WHAT they are doing to solve it. Big difference.

If not enough people are trying to solve it now, in any way, be very careful. It usually means the pain is not strong enough to push them to change behaviour, and that brings us to the test most founders ignore, because they are most afraid of hearing a “no”.

Question 3: Does your idea need people to fundamentally change their behaviour for it to work?

This is the “clever idea” assassin.

An idea can be original and genuinely unexplored, but still fail because it demands that people do something fundamentally differently from what they already do to accomplish a similar goal.

If your product only works if your customer adopts a new routine, lets go of something they know, or persuades others to take action, then no marketing stunt will solve it later.

So save yourself the future heartbreak and ask the question now:

How much of my customer’s day has to change for my thing to work?

The smaller the chunk, the better your odds. The ideas that work fall somewhere inside a behaviour pattern that already exists, making it either cheaper, faster or less painful.

Question 4: Who, specifically, has this problem?

If your answer is “everyone”, you don’t have a customer.

At early stage, the ideas most likely to succeed are narrow to the point of feeling that there is no real market for them.

Not “small businesses” but “independent physiotherapists in the UK who bill insurers”.

Not “busy people” but “new parents on a second child who already used a nanny for the first”.

Have a specific slice of people, with a specific version of the problem, whom you can find and talk to.

This sounds backwards, especially when you put it side-by-side with the common (misguided) advice to look at the TAM (total addressable market). But it matters because of two things.

• It is the only way to actually validate anything. Simply because you cannot interview “everyone”. And:
• It is the thing investors will ask you about first, long before they ask about market size. Depth is that matters most at early stage, breadth is for later. Market size is what you talk about once you can show how winning this small group opens the door to a bigger one.

Question 5: What do you know that other people don’t?

Ideas are cheap. Insights are not.

The idea itself is rarely what makes a company successful, because:

• Most startup ideas change shape once the company starts experimenting with real customers. And:
• If it is any good and it makes money, it will be copied eventually anyway. So forget about the whole “I need to protect it before I launch it”. No one wants to copy an idea that doesn’t make money.

What is genuinely hard to copy is your unique understanding of why this problem exists and persists, which is a result of you spending time inside it and your unique experiences with it, and cannot be copied by looking at a langing page. Maybe you lived the problem for years… Or maybe you have talked to two hundred people who have. Maybe you have seen the inside of the industry in a way outsiders couldn’t have.

If you can’t name one single thing that makes your understanding of this problem unique enough that you are the only one that can generate this specific value for your customers, then the idea is not yet yours anyway.

How to go out and get these answers

None of these methods need a product, a pitch deck, or investment.

Find ten people who match your incredibly narrow and specific subset of customer, and ask them to talk about how they experience the problem and how they try to fix it today. Not if they would pay for your idea. People lie, remember that. Instead, focus on asking what they do now, what it costs them (monetarily and emotionally), and what they have already tried to do to solve it.

Then, after you’ve spoken to at least 10 people, compare their answers to your idea. How similar is it to their workarounds? Does it solve their emotional pain? Is it worth the monetary cost you’re imagining they will pay?

Then, if you want to test reaction to ideas, run another round of tests, but now with a few storyboards, depicting:

• A 6-ish frame story from problem, to using your solution (including its price), all the way to having the problem solved.

Then show it to at least another 10 potential customers and get their honest reactions to it. Don’t worry, no one will steal your idea.

This is a great way to get a feel on how your customers would react to your product, but it’s still not evidence that you have a market. The only real evidence is if they either open their wallets, or at least signal that they would with a behaviour, such as by clicking a “buy now” button.

What if the answer is “no”?

Sometimes you run these tests and the idea doesn’t survive them. Believe me, if you’re genuinely trying to validate whether or not to start a business, this is a good result, and you will not be afraid of finding it out.

If it’s of any consolation, most successful companies did not start as the founder’s first idea. YouTube was a dating website. Twitter was a podcasting company. Slack was the internal chat tool of a video game. What each founder kept was the insight and the customer, and what they let go of was attachment to the initial idea. Founders who are willing to kill a weak idea early keep will their savings, energy, and sanity.

The founders who fail hardest are the ones who fall in love with the idea and use it to validate themselves, biasing their “research” phase towards yeses because they can’t bear to face reality.

If you would rather not run these tests alone…

Deciding an idea is worth pursuing is the first of a hundred micro decisions, and it’s the one founders most often get wrong on their own, because it’s really hard to see your own idea clearly from the inside.

That is a lot of what I do at Lumni. I run a founder-side advisory practice in the UK, working with early-stage founders on strategy, story and positioning at the very early stage, or ahead of a raise or an exit.

But the bulk of the work usually starts here: working out who is your customer, what you know about the problem, what you have only assumed, and whether your idea deserves the time you think it does.

If you are at that point, and you would rather have that conversation now rather than after your savings are gone, get in touch.

---

FAQ

How do I know if my startup idea is good?

A good idea solves a problem that a specific group of people already tries to solve in other ways, and some of them will commit money, time or access to have it solved properly. If nobody is trying to solve it today, be sceptical, and go and talk to your future customers before spending any meaningful money building it.

Should I check the market size first?

No. Market size is a report someone else wrote, and it tells you nothing about whether your idea will work or make money. Check first whether the problem is real, who specifically has it, and whether solving it needs people to change their behaviour. Market size matters later, once you can show the path from a small group to a larger one.

What if nobody has done my idea before?

Treat it as a warning, not an opportunity. Either you have found something genuinely new, or someone has already tried and failed (most likely). Search for the graveyard, ask people in the industry, and find out which is true before you invest.

How many people should I talk to before deciding?

Enough to hear the same things repeatedly, which is usually ten to twenty people who closely match your specific customer. More conversations with vaguely relevant people are worth less than fewer conversations with exactly the right ones.

Do I need an NDA to test my idea?

Usually not. You can explore the problem and how people handle it today without revealing the intricacies of your solution. Use an NDA for anything technical or novel that you might later want to patent, but do not let fear of theft stop you talking to people.