4 min read

How we picked one startup idea out of 125.

Systematically evaluating our startup ideas helped us pick one to proceed with confidence.

After launching my last product (a Twitter bookmarking browser extension),  I was ready to experiment with my next idea.

The goal is to quickly validate what works and what doesn't, so the approach is to build and launch small ideas as quickly as I can.

Ideas that require big up-front R&D time (e.g. a new jet engine) or ideas that require the creation of a new market (guided tours on unicycles?) will have to wait.

125 to 17

So where do we start? Well, that old "Ideas" notebook of course.

The problem is, I had 125 "ideas" in it. Even if I launched one a week, that'll take over a year to validate all of it.

So the first step is to filter most of these out. We decided to remove:

  1. Ideas that we just didn't feel like working on. Mostly ideas that we don't think we have a deep knowledge about or ideas that didn't feel exciting enough.
  2. Ideas that didn't have a clear problem statement. A clear targeted segment and problem statement makes gives your business a good start. Otherwise you're more likely to build a solution in search of a problem.

17 to 9

That took the number of ideas down from 125 to 17! Phew.

Now we can do a more detailed evaluation of the 17. For this, I'm eating my own dogfood, a workbook for evaluating ideas in a structured and repeatable way.

Google sheet template for evaluating startup ideas

17 is still quite a number to get through with this much detailed work, but we wanted to see how manageable this is. It did take a few days but we got quicker and better at it, the more we did it.

As we were doing this, we also started streamlining how we carry out market research on each idea. We ended up with a simple market research template that we can use to quickly determine is the market for an idea gives good signals for proceeding.

I'm going to add the market research template to the idea evaluation workbook too. (If you have bought this workbook, you'll get the update for free).

Basic market research template

We ended up with 9 ideas that we thought were promising. How did we get to 9?

We did it manually by sorting the idea scores and then discarding ones that didn't have the criteria which we felt were important to use, like the ability to launch quickly with it.

9 to 3

To narrow down the list further, we had to dig deeper into these 9 ideas. For this, we went back to some basic online customer research.

We mostly looked at Reddit, LinkedIn, Facebook Groups, Slack groups, Twitter, and any other social media relevant to the audience we were researching. In most cases, we'd find them on Reddit.

What we wanted to do was look for more evidence of the problem we're trying to solve. Evidence, in this case, would be people posting about the problem and the post would have a reasonable amount of engagement on it with others saying "me too!" or people proposing solutions.

Another thing we looked for were people posting "hacks" to get around a problem, or describing a "system" to do something which typically takes many steps, where we think we can automate.

And this gave us the final 3 ideas.

3 to 1

By this point, we have 3 ideas we want to work on. But we can only do it one at a time.

The first idea started out as an idea for a simple tool to send out recurring invoices. This was a need that I had personally. We have since refined this idea and made it more focused. It is an invoicing solution for Notion (the productivity tool).

Notion Invoice. One of the three ideas we'll be building and launching this year.

It will help freelancers and small businesses generate beautiful PDF invoices which syncs with projects they already have in Notion.

Next steps

Will this idea succeed? Who knows. But going through the ideas and evaluating them systematically gave us the confidence to go ahead with it.

The only way to find out is to get it out there and validate all our assumptions about the idea.

Here's what we're doing next to validate them:

  1. Experiment with various distributions channels and see which one gives us the most traction.
  2. Validate our assumption of the customer problem with the messaging in those channels.
  3. Validate the solution with a product demo on our landing page, and measuring conversion.

The method we're using is called the Bullseye method, from the book titled "Traction: How Any Startup Can Achieve Explosive Customer Growth" by Weinberg and Mares. It's a great book for getting you thinking systematically about testing and growing your distribution channels. A couple of the recommended channels do feel slightly dated already and there are some new channels that are missing from the book (e.g. targeting newsletters), but overall it's a great guide on a vital activity in a startup.

Using this method we have already picked 4 channels to test: SEM, social ads, SEO and content marketing. Of course, I'll be sharing our learnings on this blog/newsletter (do subscribe if you don't want to miss them).

You can also follow my progress as I build in public on Twitter: https://twitter.com/farez

Here's the idea evaluation workbook we used: https://startstrongbook.com