Stop Chasing Ideas: Start Validating Products

We’re a service agency, and we get ideas for new products weekly, if not daily. Pretty much all of them are great ideas, but only a few become successful products.

After working for about 14 years in the tech industry, I’ve stopped believing in ideas. Why? Because sticking to the idea blindly is the end of any product.

The Trap of “Great Ideas”

How do you move forward with your idea? With processes and validation. It’s really simple, not hard at all... but honestly, it can be boring.

We’re all wired to want to create and build. Designing a new product is fun. Building a product is fun. Testing it out is exciting. The whole delivery process gives you dopamine hits every step of the way. But when the product fails, the delivery doesn’t matter.

The issue at hand here? Most people start from the solution, not the problem. And the earlier you validate that problem and its relevance, the higher your chances of success.

A Lesson from Hospitality Industry

My cousin is an interesting (and quite untypical) business mentor. He’s not from tech, he’s in the hospitality industry, but he’s built and sold several successful ventures.

When he starts a new project, he doesn’t open a restaurant or a bar right away. He spends weeks visiting competitors, ordering food, observing who and when comes in, chatting with staff, talking to chefs about bestsellers, and watching delivery patterns. He looks like he’s “procrastinating”, enjoying life, sipping coffee and feasting on a delicious tart at midday, as it may seem to an unknowing eye. He’s actually validating that an idea he has can be successful.

Then he takes it further. Before spending a cent on renting a location or hiring staff he hires a marketing agency to build a brand, create a landing page, run ads, and test engagement. He checks whether people actually want what he’s thinking of offering. Only if he sees traction does he move forward to build the business model.

That’s product discovery in action: observe, test, validate, and only then build.

Translating This to Digital Products

When it comes to digital products, early-stage validation follows the same principles. Before a single line of code is written, you need to answer three things:

Is the problem real?

Are people actually facing this problem often enough, and more important is it painful enough, that they’re willing to pay for a solution?

Is the solution aligned with how they behave?

Does your proposed product fit naturally into their existing workflow, or does it demand a behavior change that won’t happen?

Can it be tested quickly and cheaply?

Can you create a minimum viable experiment to see whether people engage with your concept before you invest in development?

Frameworks and Methodologies for Early Validation

Early validation is the backbone of successful digital product development. It helps teams test assumptions before investing heavily in design and engineering. The goal is simple: reduce risk and increase learning.

I wont go deep as there are so many smart people out there that have written about frameworks and methodologies, instead I’ll just mention a few core ones. Frameworks and methodologies like Lean Startup, Design Thinking, and Value Proposition Design provide structured approaches to discovery, helping teams validate problem, solution fit before chasing product-market fit. Books like The Mom Test for unbiased user interviews, Tools like RICE and ICE scoring for idea prioritization, and JTBD (Jobs To Be Done) for understanding user motivations bring clarity to decision-making.

Using AI in Product Discovery

In practice, modern validation often blends AI-driven idea simulation (e.g., generating personas or fake landing pages for A/B tests) with classic MVP experiments. The key is to gather real data fast through prototypes, fake doors, or waitlists and let that data guide the next step.

Here are a few AI prompt examples you can use during discovery:

1. Market Validation Prompt

“List the top 10 online communities or forums where people discuss [problem]. Summarize the most common pain points and language they use.”

2. Customer Persona Prompt

“Based on the following product concept, create three distinct customer personas including motivations, frustrations, and decision triggers.”

3. Awesome business model validation prompts

1. Tarpit tester 

2. RedditA ChatGPT Prompt to validate any business idea

3. RedditThe Ultimate Prompt for Deciding on Your Business/Filtering Startup Ideas

4. RedditChatGPT Prompt of the Day: THE BRUTAL BUSINESS IDEA VALIDATOR

5. RedditHow I used Claude to validate my idea in 10 minutes (Now at $2.3k MRR)

AI can act like an on-demand strategist helping you simulate feedback, generate hypotheses, and structure your validation experiments. At the end of the day it can help to be a non-biased observer and punch holes in your ideas or business models.

Just be careful it’s still just AI, and you're the human that brings the idea to life.

Where Great Ideas Lie - in Open Sight

Sometimes great ideas lie at our doorstep. A lot of world leading accelerators, VS’s and companies are actually listing problems that they need to be solved. They are called requests for startups and I found out about them just recently - a bit of a mind-blowing situation. Why not simply go there and look for some inspiration for your next idea. 

Here is a list:

1. ycombinator.com/rfs

2. a16z.com/big-ideas-in-tech-2025

3. stripe.com/startup-industry-trends

4. spc.com/request-for-curiosity

5. conviction.com/startups

Honestly, it feels a bit like cheating when you go through them.

The Real Goal: Validation, Not Confirmation

At LambdaWorks, we’ve seen one pattern repeat over and over: the best founders are the ones who are least attached to their ideas. They’re obsessed with truth, not with being right.

Great product discovery is about falling out of love with your idea and falling in love with the problem.

When you do that, your “idea” stops being fragile inspiration and becomes an information-backed opportunity.