How to increase developers metrics 5x times: WeLoveNoCode story
My name is Alex, and I am a product manager at WeLoveNoCode — the biggest marketplace of no-coders in the world. I helped rebuild the web application from the ground up and improve customer and developer funnel conversions which drove our metrics 5x times! This article will show you how we achieved these results and increased quality without losing the data.
Here’s what you will learn from my article:
1. How to find core problems.
2. Strategy to increase conversion from lead to completed registration.
3. Strategy to make people complete their portfolio & tests.
4. What else you should do.
How to find core problems
I work at no-code marketplace. We went through a hyper-growth stage, going from 0 to about 18k no-coders in just a year. And as you may know, hypergrowth usually does not come without issues; in our case, we struggled with the supply quality.
We faced the most problems in our developer funnel. At the very top, our sign-up form abandonment rate was 66%. Just imagine, 4 out of every 10 developers left without completing the registration.
Then, on our platform, we also have certifications that allow developers to prove their knowledge of certain no-code tools and make this tool show up on their profile for all the clients to see. Usually, developers only complete a couple of these tests, if any at all.
Last but not least important issue was the quality of the portfolio. They often have been either outdated or inaccurate and, as a result, perform a poor job of showcasing the developer’s skills.
In our effort to optimize the funnel, we started exploring how other companies are solving similar problems. Some of you may have heard about TopTal — it’s a platform for hiring freelance talents that primarily focuses on traditional developers, as well as managers and designers. Their main motto is that only 3% are top talent.
The idea behind this rule is to funnel people through a multi-step screening process that only top-notch and most motivated candidates are able to complete. The method includes testing their communication skills, professional skills and completing take-home projects that are pretty comprehensive and may take 1–3 weeks to complete but present candidates with real-world scenarios. And last but not least, a proven track record of excellence, backed by client reviews.
I thought it was really cool and decided to explore creating something similar but fine-tuned for us. The idea sat exceptionally well with us as we had a lot of no-coders, but we didn’t really know anything about their professional background; we even had cases of our clients deciding to cancel on us because of the quality of services no-code developer offered them. So it seemed like we needed to focus on forming a group of top-performing no-coders that we would trust and recommend to clients and make the rest strive to become part of that group (while growing our MRR!).
By that time, we already had a screening process and a simple scoring system in place, but developers didn’t know that — we didn’t do an outstanding job of communicating that. We decided to build everything on top of the existing system. This was the resulting funnel we came up with:
- Developer signs up.
- Developer picks the tools they are good at and self-rate their skills.
- Developer passes necessary tests — English language test and basic design.
- Developer passes additional tests for tools they picked earlier.
- Developer uploads their portfolio.
After that, our matching manager reviews the developer’s portfolio and their verbal communication skills in a 1-on-1 call in English. The general idea was that the more complete your profile is, the more tests you pass, the better your score would be.
But how can we improve the quality of tests and portfolios so that developers are more willing to fill out the data?
Strategy to increase conversion from lead to completed registration.
We started conducting problem interviews to find out what was wrong with the sign-up form. This is what we got out of them:
- The form was long and confusing.
- The form was not responsive and looked terrible on mobile devices.
- The form was not easily reachable — all sign-up callouts led to a separate landing page, and the form was only accessible from there.
We had our work cut out for us. These are the hypotheses we formulated on how to improve the sign-up conversion rates:
- If we decrease the number of steps and questions in a developer’s registration form from 4 to 2 steps, the number of registered developers will increase.
- If we remove the developer sign-up landing page and guide people straight to the sign-up form, the number of registered developers will increase.
- If we add a responsive mobile version and fine-tune the UX, the number of registered developers will increase.
Pro tip — this is the template we’re using to present hypotheses, feel free to use it in your explorations:
What we are testing: If we decrease the number of steps and questions in a developer’s registration form from 4 to 2 steps, the number of registered developers will increase.
Why: around 66% of developers drop off on 2 and 3 steps of the sign-up form.
What we want to achieve: +20% conversion rate.
How to measure it: Conversion rate (lead -> registered developer).
Who will lead this: Alexandra Zaitseva
Implementing all of the above resulted in multiple wins — we now have a simple yet dashing design, and we were able to increase the sign-up form’s conversion rate from 34% to 51%.
Strategy to make people complete their portfolio & tests.
OK, so the sign-up process is working great. What’s next? Creating a funnel and selecting the best of the best of no-coders. We’re still a startup, so we don’t have the resources to book a call with every developer or give them a take-home assignment to test their skills, so TopTal’s approach does not work for us. We’ll have to look for another way.
Initially, we added a pop-up message that asked developers to complete the tests. This helped increase the share of developers with at least 1 passed test up to 50%. But it still was not much because the number of tests on our platform grew, but nobody wanted to use them. Moreover, new developers started taking just the English test. We needed a way to make them take the other tests as well in order to be able to track and measure developer skills and quality.
We decided to look into gamifying the developer onboarding experience. Tweaking the dashboard that we had on the developer homepage was the most hassle-free solution: we added a progress bar and several required tasks. Basically, they had to complete the tasks to get access to the platform’s full functionality.
We also focused on improving the general UX of the platform, things such as cleaning up and hiding unused links in the menu. We added a progress bar that tracked the developer’s progress in completing the tests on the tests page. All in the effort to make them more prominent. But the general cleanup did not finish there — we also fully redesigned the progress steps in the developer’s dashboard to better explain the game’s rules and put yet another progress bar in the sidebar, indicating the profile completion.
The next step was to wait — and it was totally worth it. Our conversion skyrocketed. Indeed we expected to achieve good results, but the numbers we received were even better. Conversion from registered developer to developer that passed 1 test went up to 74%, and conversion from 1 test to passed 2+ tests went close to 70,1%.
Now that the sign-up process was working well and we had a working developer quality metric, it was time to divert our attention to a developer’s track record — their portfolio. At the time, it was more of a formality. Almost everyone’s portfolios were severely out-of-date, and this part of the app just did not receive enough love, neither from developers nor from us. We decided to stick to the well-working solution and gamify the experience here — it turns out an incomplete progress bar can motivate many people!
The results were terrific: 67,2% of developers uploaded at least 1 item to their portfolios, and 34,62% uploaded 3 or more items.
What else you should do.
Here is what I recommend you do. Use problem interviews to get a bird’s eye view of your product. Brainstorm with your team. Experiment with gamification to better engage users. Make user experience more clear — but don’t overdo it. Don’t be afraid of changes; even stable processes can be improved if approached thoughtfully.
As for WeLoveNoCode, we are still searching for ways to improve and increase our metrics. But thanks to the changes from this article we’ve been able to add 30 developers to our top-tier developers cohort in the last two weeks.
