Poll: what tech do you use to build personal sites in 2021?

đź”” This article was originally posted on my site, MihaiBojin.com. đź””

Finding topics to write about

Like many others starting their writing journey, I struggled to imagine what to write about.

For years, I didn't think I had anything to teach.

Until I came across the Indie Hackers podcast and heard a thing repeated over and over again...

What’s something that’s easy for me to do but hard for others?

What's a skill or asset I have that’s very difficult for people to reverse engineer?

The breakthrough here is that you can #LearnInPublic and then teach others what you learned, through writing!

When I started this site, I knew nothing of Gatsby. But, as I learned how to use this technology to build my personal site, I realized I could teach others about what is needed to create a publishing platform from scratch.

But I didn't want to invest loads of time in a comprehensive tutorial without doing market research to assess the need for such a thing.

Promoting the poll

So I created this poll on Twitter!

For the first few days, I got under ten responses. My audience is so low (no surprise) that it was barely seen. Since I wanted to get a more accurate picture, I decided to promote it to increase responses.

Through its UI, Twitter encourages you to spend a minimum of 50 EUR per day. This felt steep, but I was determined to get actionable data, so I thought I'd try it.

It turns out that going through that flow first asks you to create a Twitter Ads Manager account. And there, I had better control over the daily spend - awesome!

Twitter ads: stats

I created a campaign, set a maximum budget, and then created an Ad Group with a single tweet: my poll. I let it run with automatic bidding and targeted the following interests:

  • Enterprise Software
  • Business Software
  • Graphics software
  • Computer programming
  • Web design

With these settings, Twitter told me I could reach an audience of 12.1M – 14.8M people.

I set a daily budget of 10 EUR and set it loose around 4 pm. After a few hours, I had spent 7 EUR and garnered 240 votes. This was already enough to be reliable enough statistically, but then I got greedy.

I reconfigured the campaign to a 20 EUR per day budget, and in a few hours more, I quickly saw most of my daily budget exhausted. At that time, I reconfigured the campaign to end at noon the next day.

Lesson learned: you can get far with Twitter promos without breaking the bank! In retrospect, I shouldn't have doubled my daily budget.

When I woke up the following day, I had already spent 38 EUR. After ~19 hours at 11 am the next day, I stopped the campaign at around 643 votes (though it's up at 658 since then; it looks like it reached a few more people since I stopped the ads.)

Other stats:

  • 339 likes (341 total)
  • 15 visitors to the site from Twitter
  • 4 new followers on Twitter

Poll interpretation

This is hard since there's no way to tell if I'm affected by confirmation bias.

Only 21% of respondents selected they use SSGs like Gatsby - these are people who may or may not already know what I learned - their familiarity with this stack makes them a viable audience, or at least receptive to such content.

28% use services like Medium and Substack. A few reasons I can think of for using publishing platforms are:

  • lack of time and focusing on producing content rather than maintaining a website (a very legitimate reason, in my opinion)
  • being less knowledgeable in frontend web development
  • generally not feeling like going through the hassle of building and maintaining a site from scratch
  • cost (publishing to Medium is free compared to hosting a WordPress or Ghost website)

17% do none of these things , but my poll didn't get many comments, so it's hard to figure out what this means. I'll ignore this category for the time being.

Leaving the most popular choice by far - 32% preferring frameworks like Next or Nuxt.

I can imagine part of the motivation behind this last choice when compared to SSGs:

  • prior familiarity with the stack (e.g., React)
  • a need for more dynamic features
  • lack of familiarity with SSGs like GatsbyJS (high initial barrier of entry)
  • dislike of Gatsby, Hugo, or other SSGs

Conclusion

To sum this up, I think there is potential for teaching others "how to Gatsby." I will continue to share my learnings in the personal site series.

If I missed anything here or want to share your opinion, I started this Twitter thread to collect feedback!

If you liked this article and want to read more like it, please subscribe to my newsletter; I send one out every few weeks!

15