Picture this: you're at an AI event in London, surrounded by brilliant minds and cutting-edge technology. The energy is electric, the possibilities seem endless. But as you chat with these tech wizards, a sobering reality starts to emerge.

You see, a great builder does not a great business make.

Time and time again, I've seen fantastic teams with mind-blowing AI tools, but there's one tiny problem - no sales. Nada. Zilch. 没有钱 (˚ ˃̣̣̥⌓˂̣̣̥ ).

And it's not just at events. Hop onto any AI directory or Product Hunt, and you'll find yourself scrolling through a graveyard of dead products. Tools that someone poured their heart, soul, and (probably!) a fair bit of cash into, only to have them languish in obscurity.

Hell, maybe you've been there yourself. I know I have. It's a tough pill to swallow, but in today's world, "build it and they will come" is about as outdated as a flip phone. And less likely to have an ironic comeback…

If this is all hitting a bit too close to home and you’re annoyed with me right now…that should tell you something! This will be an important Part for you.

The hard truth is this: with AI making tool creation easier than ever, we're drowning in a sea of products. And it’ll get worse.

But don't worry, today we're looking at how to avoid this fate. We're going to look at into validation - the step that separates the AI apps that thrive from those that dive. It rhymes so it must be true.

Let's get started:

The Validation Imperative

Before we dive in, let's get one thing straight: validation isn't just a nice-to-have step in your AI app journey. It's the difference between building something people want and building something that collects digital dust.

It’s a topic covered in lots of startup books and courses that everyone agrees is a good idea. And then they plunge straight into the fun building part instead.

Think of validation as your reality check. It's your chance to test your assumptions, refine your idea, and make sure you're solving a problem people actually care about - all before you invest significant time and resources into development.

You might spend MORE time validating than actually building your first version. And that’s fine. That’s not wasted time or procrastination. That’s ensuring what you build is worthwhile.

Competition

Here's a counterintuitive truth: if you find similar non-AI tools to your idea, that's actually a good thing. Why? Because it confirms there's a market out there. People are already looking for solutions to this problem.

Market risk is waaaay worse than competition risk. We can outmanoeuvre competition. We can’t outmanoeuvre a market not existing.

Your job isn't to create a market from scratch (that's a whole different ballgame and you need DEEP pockets). Your job is to bring AI superpowers to an existing market. Think of it this way: you're not trying to convince people they need a car. You're offering them a faster, more efficient car when they're already in the market for one.

Thankfully this is now trivial using AI.

An example. But you have to promise not to laugh at me.

The other day I was looking to buy some plastic boxes for storage. I went to the (aptly named) PlasticBoxShop.co.uk website:

Here’s the thing.

I want a plastic box that is a certain size to fit in a certain space.

Sounds trivial right?

NOPE. Absolute nightmare of a task to do via any of the shops that sell boxes. I’m faffing around with multiple sliders. Eventually I had to do maths to convert dimensions in centimetres into volume in litres.

All to find a bloody box that fits nicely under a kitchen cabinet.

OK what about this then:

• collect up all the boxes and their specifications on this site (and other sites if you’re feeling funky)

• chuck them into an AI as data

• build a tool that suggests the best boxes based on your dimensions

EASY. Like…I almost scraped and built this just to find my kitchen boxes I kid you not.

That’s an AI solution to an existing problem that has previously been solved using no AI.

Fix that and you can:

• make a tool that affiliates to different box shops for a cut

• sell the tool to an existing box shop for a few thousand

• set up your own box drop-shipping store

Lots of options. Just by taking an existing problem and applying a bit of AI to solve it.

So, do your homework. Look for existing solutions that could be instantly improved with a pinch of AI. If people are already using the crappy non-AI solution they’ll bite your hand off for a better experience.

Also: if anyone wants to run with the plastic box tool above be my guest. I’ll tell you for nowt that I want that problem fixed! 😁

The Power of Customer Conversations

Finding existing solutions is passive validation.

Now, here's where the rubber meets the road: active validation.

You need to talk to real, live humans. I know, I know - as tech enthusiasts, sometimes we'd rather talk to our AI models than actual people. But trust me, this step is crucial. Sorry!

Aim to book 10-20 calls with potential customers. These aren't sales calls - they're learning opportunities. Your goal is to understand their problems, their current solutions, and what they're looking for.

Here are some key questions to ask:

1. What's the biggest challenge you face when it comes to [problem your app solves]?

2. How are you currently solving this problem?

3. What do you like about your current solution? What do you wish was better?

4. If you had a magic wand, how would you solve this problem?

5. How much time/money do you spend on this problem currently?

Remember, your job here is to listen more than you talk. You're gathering intelligence, not pitching a product.

Listen to their problems and (just as importantly!) the language they use to describe their problems.

Now, most of you won’t actually do this. I know this. But it’s one of the biggest differentiators of who will be successful. Sorry to be brutal but pulling your finger out here will exponentially increase success.

Leveraging Your Audience

If you've been following previous Playbooks you've already started building an audience. Now's the time to leverage it.

Create posts asking for feedback on your idea. Run polls. Ask the "Is this a stupid idea?" question (you'd be surprised how honest people are when you frame it this way + your ego isn’t threatened when everyone agrees it’s dumb!).

The goal here is to get as much feedback as possible. And remember, silence is also feedback. If you put your idea out there and hear crickets, that's valuable information - they DGAF! 😭

Validation Checklist

To help you navigate this process, I've created a validation checklist. Use this AI prompt to generate a customised checklist for your specific idea:

You are an AI specialising in market validation for AI applications. Based on the previous AI app idea, create a comprehensive validation checklist. The checklist should progress from easy to implement tasks to more complex ones, acting as a red, orange, green light system for proceeding with development.

If no app idea was given previously prompt the user.

Please include:
1. At least 10 validation tasks, ranging from market research to customer interviews
2. Specific metrics or goals for each task (e.g., "Interview at least 15 potential users")
3. A scoring system to determine whether to proceed, pivot, or abandon the idea
4. Suggestions for pivoting if certain checkpoints aren't met

Run this prompt with your refined idea from Part 2. The resulting checklist will guide you through the validation process, helping you make data-driven decisions about your AI app's viability.

But you actually need to do the work here.

Remember, every "no" you hear during validation is saving you countless hours and dollars down the line. Embrace the nos and keep asking until you are hearing “tell me more”.