A review of the blog in 2021
I took writing slightly more seriously in 2021.
Targets for the blog in 2021
In the 2020 review I set a few targets for 2021:
- ✅ Publish 50 posts with an average of 600 words
- Published 78 posts
- I wrote a total of 42868 words (counted using
- For an average of 549 words
- ✅ Follow a better structure
- I do thing the posts are better structured. I found Julian Shapiro’s handbook quite useful for this.
- ✅ Have at least 5 dominant categories rather than 2
- ✅ Have at least 1 new post be in the top 3 pages of the year
- ❎ Have at least 2 posts in every month of the year (made possible by the fact that I published in January posts that have been ready for a long time)
- I failed in January because I made these targets after January.
- I failed in August and September because I took a break, partially because I published 16 in July, and partially because I was working on The Marketplace Guide. From a productivity point-of-view, it is better that I published 16 in one month rather than 12 in three months. But, from a quality perspective, it probably isn’t.
- ✅ Keep track of, and regularly publish, posts that are complete
- I did this. There are at most two or three posts that are in progress.
I published 78 posts, totaling approximately 42868 words. The average blog post length is 549 words.
Overall, I’m satisfied with the average post length. Perhaps a bit longer would be better, but I didn’t try to artificially inflate the word-count. Instead, I tried to expand on my thinking while at the same time trying to be more succinct.
I have a nice distribution of topics now, from the arcane and technical to product, ethics and, business. I’m satisfied with the distribution.
I was quite surprised because I had several posts that got really quite popular.
Reddit’s disrespectful design was the runaway most popular post that hit the front page of Hacker News. It was, in fact, so popular that it was even in “webshit weekly”. It’s definitely not the post I’m most proud of, however.
Interestingly, Google Analytics says it had ~42k views. But the logs count it somewhere around 200k. That indicates that approximately 75% of the audience had Adblock. This is consistent with other posts talking about traffic generated from Hacker News.
Generating more interesting image previews using imagemagick is in second place with ~7k views. I’m much prouder of this post, however.
Technical blogging in the era of Stack Overflow is another post I’m quite happy with. It had ~5k readers. It also got reposted to several other aggregators.
Turns out Elastic aren’t the bad guys is an interesting one. It didn’t really do well on Hacker News. However, it got picked up by one of the Elastic guys who posted it on their LinkedIn. As a result, it had more than a thousand readers. I was quite happy to see that.
Don’t scan your face to buy glasses is probably the most important post I wrote, even though it didn’t get a lot of traction. It had ~800 readers.
The Prune Date was a really nice ego boost. It’s a post from 2019 that I posted to Hacker News. It didn’t do too well but, it was then put in the second chance pool, and I got an email about it. It was very encouraging.
Besides the blog, I also wrote TheMarketplace.Guide.
It doesn’t have a lot of traffic, yet. I need to spend more time on marketing it.
It consists of 70-80 posts with a total of 26k words. So, that’s an entire body of work that I did.
Even though it didn’t get a lot of exposure yet, I am satisfied that I managed to complete a much larger body of work. It is encouragement for another series or guide that I am thinking about.
It also helped me further improve my writing because I had to make several passes through all the lessons to ensure they all followed the same structure. That way of working helped me see that I was improving between passes.
I also got over a hump posting my project somewhere for others by posting it on Product Hunt. It didn’t do great. But two great things from doing it: 1. I had to get over myself to put it out there, and 2. I had to make some marketing graphics that ended up looking quite good.
And finally, one person actually bought the PDF version. I rode that high for about a week.
What I’ve learnt
I like writing. It helps to unload when you have an idea.
It takes time to produce quality.
It’s unnecessary to explore all the aspects of an idea the first time. It’s fine to indicate what others might want to write about, or to leave something for another post.
Practice does make perfect.
Building an audience is so important but difficult.
- Publish at least 50 posts
- Reach at least 100k readers (would that mean 300-400k measured from the logs?)
- Build an audience (either recurring readers or Twitter.)
- Publish something on at least two other sites
- Focus more on interesting and positive posts instead of whiny ones