profile

Jack Bridger

Validating DevTools Ideas

publishedabout 2 months ago
5 min read

In the last edition, I promised you Developer Marketing lessons from my trip to Japan - you'll find this at the end.

We also have:

  • An update on a DevTool launch and some thoughts on validation
  • Takeaways from four Scaling DevTools episodes, including Ramiro from Supabase!
  • Summaries of two great talks I watched this month from swyx and Phil Leggetter.

Validating my DevTool idea

I am working on validating a DevTool idea for a problem I encountered in my last job as a mobile developer. The problem is this - we spent a huge amount of development time making minor changes as directed by the marketing team.

We (myself & Nick) feel confident about this but we are working on validating it.

We decided the first step was to launch on Reddit (thanks Matija for the idea). Here's what we did:

  1. We focused on r/reactnative. We're both React Native devs (& Nick is a really good RN dev) and so we started to engage in the subreddit and answer questions. One of Nick's responses got over 20 upvotes.
  2. We looked at previous product launches in the subreddit and found that videos + 'looking for beta testers' had a history of doing well
  3. We made sure to follow all guidelines - no linking to landing pages or self promotion etc.

The results

  • 26k impressions & 100+ upvotes (4th of the week)
  • 3 direct messages asking to be beta testers
  • Several supportive messages
  • Several useful clarifications
  • Discovered a very similar product we didn't know about
  • Two questions on our open source strategy

Lessons/next steps

It's exciting to see that there is strong interest among individual developers. But it doesn't really validate the idea yet.

We watched this great video from Rob Walling on validation and now we're focusing on finding one or two companies who are so passionate about this problem that they're willing to come on the journey with us.


Community helps you cross the chasm

The idea of crossing the chasm is a big deal and will be a focus for my learning in the coming months.

Some developers will adopt technology because it does a job. But many others - the mainstream market - care about things like:

  • What are the job opportunities with this technology?
  • How is the third-party ecosystem (e.g. libraries)?
  • Do my friends use it and what do they say about it?

Crossing this chasm at an early stage is a huge challenge. Community building builds social proof and momentum so that the mainstream market feels like it's not such a risky technology to adopt.

This came from my notes on an amazing talk by swyx.


Which activities will help you reach YOUR goal?

To grow your DevTool, the list of different approaches is practically infinite. I like the AAARRRP framework from Phil Leggetter because it helps you to narrow down your efforts in a systematic way.

Phil suggests focusing on some (but not all) of seven possible goals e.g. Awareness, Activation and Product.

Then look at all possible activities and see which of them will help you reach your goal

Then you can look for activities that meet multiple goals and give each activity a score.

You can find my notes on this here.

P.s. thanks to Carla for recommending this talk


Thomas & Julius from Storyblok


Ramiro from Supabase


Kyle from Gravity


Ed from OpenCage


Two Developer marketing lessons from visiting Japan

1) Naturalness

In Japanese gardens, a lot of effort is expended to create a feeling of naturalness - hiding the hand of the designer.

This also works well in developer marketing. The best content has a feeling of whimsicalness.

Rather than obviously emerging from some ahrefs keywords analysis, good content feels like it arose naturally from a pure interest in the subject matter.

2) Specialisation

Most Japanese restaurants do one type of food and they do it well. You'd be very hard-pressed to find a place that does sushi and ramen for example - it would probably be treated with suspicion: "how can they do both well?".

As DevTools - especially at the beginning - we should have conviction in saying no to the things which aren't helping our core audience to solve the problem we set out to solve.


Finally, if you enjoyed this newsletter. Would you consider forwarding it on to a fellow DevTools aficionado?