In the indie community, building a product that focuses on a single particular feature could probably be considered best practice until there’s no clear market validation.

So, why am I adding more and more features before even going full speed on distribution?

It’s because when I’m using HeyEcho myself, there are still things that annoy me. But they’re done now! 🥳 

Table of Contents

• From Shaky to Solid

• Annoyance-Driven Development

  • One more chance

  • Better flow consistency

  • “Did I publish this?”

• No Perf, No Crawl

• Conclusion

From Shaky to Solid

One thing that I fixed was really critical. 😅 

During a live blog post generation, the progress returned by the server using socketio was being broadcast to all connected clients. So if two different users were generating a blog post at the same time, they would both see each other's updates. This not only broke the UI very badly, but was a huge privacy issue. 🤦‍♂️ 

The blog post generation is also way more robust right now. Back-of-napkin calculation generating a blog post was probably failing once out of 6/7 with some random error. Now the whole workflow is way more robust. 💪 

Annoyance-Driven Development

In the last weeks, I discussed multiple times what my biggest problems were when using HeyEcho, and how they’ve been addressed by async generation and the Content Hub.

But I was still annoyed but a few things… 😬 

One more chance

As I mentioned before, the whole workflow is more stable. But before this week’s solidification (is this even a real word in this context? 😅), a failure would have meant going back to “Generate Blog Post” and going again through the wizard. Yeah, not that much of a big deal, but I’m super lazy… 😂 

So here it is, the retry button!

Better flow consistency

I guess that we all know that titles like “The top 3 ….“ or “Here are the main 5 …“ or “A simple 7-steps guide …“ are very good for SEO.

Indeed, HeyEcho was generating many blog posts with this listicle-type of title. The problem was that when reading the actual content, it was not really structured as you’d expect, that is, as a listicle.

This missed expectation was driving me nuts… 😡 

But that’s also fixed! 🥳 

“Did I publish this?”

Now that HeyEcho has the Content Hub and async generation, it’s super easy to queue multiple blog posts for generation.

At the same time, it’s now easier to get lost and ask yourself, “Did I publish this?“. Again, it’s not a deal breaker, but again, I’m lazy. 😂 

So I added a new action from the blog posts library!

Not only is it now possible to mark as published, but also to associate the public URL. 🚀

No Perf, No Crawl

Last week, I complained that after many blog posts were published on HeyEcho’s blog, only one was indexed! 😑

Well, this is very likely the biggest reason…

The frequency the Google bot crawls a website is also affected by its performance (alongside other key stuff). In general, the crawling budget for a site that is fairly new is not high AFAIK, and the fact that the performance is bad makes things even worse.

Improving the performance of the blog post is definitely on the checklist. ✅ 

Conclusion

Having tackled everything that was annoying me doesn’t mean that it’s ready for the actual launch. 😅 

But this time it’s no longer about features per se, but I plan to implement a proper onboarding flow.

After that, I’ll be focusing 100% on the distribution: landing page, assets preparation for the launch, review outbound messaging, etc. 🚀 

And ofc, as I mentioned before, I must improve the performance of the blog.

If you’re interested in following my journey, make sure to subscribe or follow me on Bluesky, X, and LinkedIn!