The Entrepreneur Insider network is an online community where the most thoughtful and influential people in America’s startup scene contribute answers to timely questions about entrepreneurship and careers. Today’s answer to the question “How do you come up with a new startup idea?” is written by Allison Berliner, founder and CEO of Cataluv.

On day one of my “Intro to Entrepreneurship” class at Wharton, I learned an interesting lesson: Different types of entrepreneurs come up with startup ideas in different ways. Some people know they want to be entrepreneurs and are looking for ideas—they are problem solvers hunting for a problem. Other people become entrepreneurs out of necessity—they have a deep problem and no one else is solving it.

Over the last five years, I’ve mostly thought of myself as a problem-solving entrepreneur. I studied the various tools and frameworks that can help identify gaps in markets and then validate business solutions. (FYI: A number of these are listed on the website of Professor Ethan Mollick, who teaches entrepreneurship at Wharton, and was in fact my instructor.)

For the first and second products I launched, I followed a fairly rigid process: 1) brainstorm with business partners; 2) create a rubric with which to assess ideas; 3) rank ideas with said rubric; 4) rapidly prototype and test top ideas to see which ones, if any, can get customer traction.

See also: What Every Entrepreneur Should Learn From the Webvan Failure

Based on what I’d learned in the classroom and from other successful leaders, each of these steps was fairly rigorous. Brainstorming with partners was a several-day or weeks-long event, during which we would each contribute anonymously to a shared idea document, only coming together to discuss once each partner had taken time to independently get inspired from different sources. For the rubric, I channeled the lectures I’d heard from investors: We ranked each idea for market size, feasibility, and competitive landscape. And when it came time to test and prototype, we used an MVP approach to throw up landing pages for each idea to see where we could get web traffic and conversion to action (email signup, for example).

Both times I followed this process, I landed on software products that targeted small businesses. Both of these products generated initial revenue and seemed to have potential. But in both cases, growth stalled and the products hit a dead end.

 

Forced to confront my failures, I re-evaluated my entrepreneurial identity and my idea-generation process. Maybe the problem wasn’t my process, but rather that I was forcing myself into an entrepreneurial category that didn’t fit. For my third act, I added a new category to my rubric: personal relevance. I decided that I’d only work on an idea if I could relate deeply and personally to my target customers and their needs.

Cataluv, my third and current business, grew out of this approach. Instead of launching a product targeted at small businesses, I launched a product targeted at consumers in my own demographic.

From the outset, I could see the difference in every aspect of the business. On the product, I had a much better intuition for what customers wanted, which fueled higher growth. On sales and marketing, I was genuinely more enthusiastic, which fueled partnerships and press. And internally, I had more stamina and passion, which made me a better and more motivating leader.

In the end, I think that the best startups come from problem-solving entrepreneurs who wait to pick an idea until they find something relevant to their own lives. I’d still recommend putting together the rubric and rapid testing, but the rubric should be heavily weighted for personal relevance and the testing should be designed to prove that this highly personal problem is felt by lots of people beyond yourself.