I went viral for the first time

woohoo

I maintain a secondary blog for professional purposes called Bytes by Ying, and recently I had the privilege of going viral on Hacker News. It was an exhilerating experience, and welcome validation that I have learned some useful things over the years and can communicate them effectively.

I use Google Analytics (although I have heard that Piwik/Matomo may be better for privacy-focused techies), and I've learned quite a bit about my users:

  • The vast majority of users come from the United States: The U.S. remains the clear leader in consumption of (my) English technical content, with the U.K., India, and Germany coming in as distant seconds. It's hard to internationalize content (though the burden is made significantly lighter with Hugo which supports internationalization out of the box), and thus difficult to communicate to large parts of the world. This realization was hammered home during the last couple of days.

  • Most people do not click around (or browse the entire article): Most people clicked on the article, browsed for maybe thirty seconds, and then left the website altogether. My bounce rate (rate at which users only visited one page) was quite high (~90%), with the lowest bounce rates and highest session times coming from India. The session times were so low that I thought many "users" were just bots scraping my website. Eventually though, I recognized the low likelihood of the number of users that are bots, given the mixture of new and returning users, and steadily decreasing bounce rate.

    I do wonder how sticky users are. I got 81 upvotes, and less than fifty users that clicked through to other pages, out of more than five thousand clicks. I got five signups for my mailing list. It will take time and careful tending in order to cultivate a strong following.

  • Having a great mobile experience is non-negotiable: Thank God the theme I used for my website was mobile-first or mobile-friendly. I think the number of users divided between mobile and desktop was about even (maybe varying by 0.1%). For a guy who usually does things on his desktop, having a great mobile-first experience is a new and good lesson to learn early.

I was worried about whether virality would create price spikes in my AWS costs. However, after double checking what the costs of AWS Cloudfront were (about ten cents per gigabyte), and realizing that my compiled files were only several kilobytes, I stopped worrying. It would take literally hundreds of thousands to millions of hits for my website to generate significant AWS costs, and at that point I could probably afford to blog full-time or something.

So yeah. Brave new world.