The best way to validate your startup idea

When validating your startup idea, ask your potential customers to pay you for it as early as you can—ideally before building your product.

875 words • 5 min read

One question that I’m frequently asked by new founders is this: How do I know if I’m working on a real problem? How do I know if my idea is viable?

I’ve written about tactics for finding your next startup idea, but once you have the idea, how do you verify that the idea is promising? The most straightforward way to test if it has legs is to actually try to sell it, as soon as possible. (Yes, even before the product exists.)

Don’t fall into the founder failure mode of spending a ton of time building v1 before talking to a single customer, or polishing your investor deck before you’ve actually gotten any validation. The most important thing for you to do when evaluating your idea is this: get out of the building, talk to three potential customers, and ask them, “If I had a product that could do XYZ, would you pay $X/month for it? Would you give me your credit card right now?”

(This advice is biased towards B2B businesses, but the lessons apply more broadly than you’d expect.)

How can I sell what I don’t even have?

You might be thinking, “But I’m very far from having a thing to sell right now. And this person is doing me a favor by talking to me—will it turn them off if I’m too ‘salesy’ with them?” (Aside: you can ask for the sale without being salesy.)

So you’re instead going to say something like, “Hey, I’m working on XYZ, what do you think about the idea?” Unfortunately, framing the question that way will lead you astray. The people you’re talking to in the early days want you to win. You’re a fledgling startup. You’re the underdog. The reason they’re talking to you at all is because they want to help you. And because they want to help you, they’re not going to crap all over your hopes and dreams.

So, even if they hate the idea, they’ll say, “Well, that sounds kind of interesting. Sure, I guess I could imagine that someone might want that.” 

There are two problems here: (a) this is what they might say if they were interested, and it’s also what they’d say if they weren’t interested and were being polite, and (b) “someone” doesn’t buy software—real, specific individuals do.

So you’ve learned nothing from this exchange.

To actually learn something, you have to make the question very tangible and very focused on this specific person, sometimes aggressively so.

Instead, approach the conversation as follows. Start by framing the question around their personal pain points: “What are your three biggest pain points right now? Great, I’m actually building a tool that solves your #1 problem by doing XYZ. Does that sound like something you’d use? Would you pay me $X/month for it?”

If they say “yes”, you follow up with, “Okay, if it existed today, would you give me your credit card right now?”

Any time you bump into a “no”, drill down on the why. If the answer is, “No, I wouldn’t pay $X right now,” that’s totally OK. Why? Maybe they’d rather spend that budget on a more pressing issue. (In that case, you might’ve found an even better business idea.) Maybe they don’t have the budget for it. Maybe their boss is the actual buyer. Regardless of the precise answer, you’re learning something tangible and concrete about what you need to build and how you’ll sell it when it’s ready.

How we validated Pilot’s idea

When my cofounders Jeff, Jessica and I started Pilot, our first day did not involve a single line of code. Instead, I reached out to a few founders in our network to ask, “Hey, if we could do your startup’s accounting for $100 a month, would you buy it from us? Yes? Ok, we’re actually doing it, would you buy it from us today? Great, I’ll send you an order form after this call.” (It turns out we needed to charge way more than $100/month to do a really good job, but that’s another story for another post.)

After we’d signed a few clients up, Jeff and I did the books by hand, while Jessica peered over our shoulders and wrote the initial lines of code. That’s how we validated that we were working on the right idea—by getting paying customers as early as we could.

As a business, Pilot was particularly well-suited to this “get the sale before you build the product” approach, but for another business, the call could have easily ended with, “Great, we’ll add you to our beta list and we’ll let you know when we have something ready for you.”

The right idea is the one with paying customers

If you want to validate your idea, find three people who represent your target customer, and ask them for the sale. Don’t start building the product. Don’t make an investor deck. Talk to your potential customers immediately and ask them: Will you give me $X today if I could solve Problem Y for you?

If the answer is no, it doesn’t matter how good the product is—it won’t have legs. And if the answer is yes, congratulations, you might actually be onto something.