[HN Gopher] Reverse engineering my #1 Hacker News article
       ___________________________________________________________________
        
       Reverse engineering my #1 Hacker News article
        
       Author : wirtzdan
       Score  : 83 points
       Date   : 2025-01-20 14:53 UTC (8 hours ago)
        
 (HTM) web link (danielwirtz.com)
 (TXT) w3m dump (danielwirtz.com)
        
       | uhoh-itsmaciek wrote:
       | Nice analysis. The point about titles is interesting: given how
       | content spreads on the Internet, it's really an invitation to
       | potential readers. If you take that too far, it can devolve into
       | tacky link-bait, but it's worth thinking about how to summarize
       | your post into a handful of compelling words.
        
         | wirtzdan wrote:
         | Exactly!
         | 
         | As mentioned in another comment, I think it's sad if one writes
         | an amazing article but it stays unrecognized or gets little
         | traction simply because the title was a bad choice.
         | 
         | But at the same time, we have lived long enough with clickbait
         | to know it works (unfortunately), and this shouldn't become a
         | place where people are lured into low-quality posts.
        
       | valtlfelipe wrote:
       | And you made it again! :D Nice writeup. Very interesting
       | analysis.
        
         | wirtzdan wrote:
         | I thought it would be nice to immediatly do a write-up.
         | 
         | Otherwise I push it into the future and it never happens.
         | 
         | Happy that I found some time this week.
        
       | stevage wrote:
       | I'm torn between "nice writeup" and "ugh, a guide to making viral
       | content on HN?"
        
         | wirtzdan wrote:
         | Yes, fair point!
         | 
         | I tried to reflect on some lessons from this, because I learned
         | from the experience that small things (e.g. the title) can make
         | large differences.
         | 
         | Which means that often good ideas or nice articles might get
         | unrecognized, because they lack a good "packaging".
         | 
         | Making sure those things are on point helps.
         | 
         | But that should be it.
        
         | rscho wrote:
         | One just needs to write "LLM" to get the same effect ;-)
        
           | dylan604 wrote:
           | at this point in time, but the rule is too specific.
           | 
           | "One just needs to write $currentTrendingFad to get the same
           | effect"
           | 
           | Not too long ago it could have been crypto/blockchain, or
           | rewriting $app in $lang
           | 
           | That's the thing of viral is it is just trend following, so
           | recognizing the trend helps the changes of going viral
        
         | cafeinux wrote:
         | Being viral, for content, is not always bad: it can make an
         | otherwise boring or tedious subject engaging and enjoyable (and
         | on the other side, a fun subject might appear boring if badly
         | written), which I think is what this article is about :-).
        
           | numinix wrote:
           | The title "I've acquired a new superpower" might have the
           | most virality, but it's not a great title for users. Chasing
           | viral content doesn't _have_ to be bad, but you do often get
           | this imbalance between clicks vs UX.
        
       | Liquix wrote:
       | something not mentioned is the importance of timing. for example
       | a link posted on saturday 11:00PM pacific will not receive nearly
       | as much attention as the same link posted on a friday at 7:00AM
       | pacific.
        
         | qianli_cs wrote:
         | I've seen several blog posts trying to analyze HN data on the
         | best time to post. However, the results are all over the place.
         | For example, the below ones have different recommendations
         | (weekend vs weekday).
         | 
         | - https://blog.rmotr.com/the-best-time-to-post-on-hacker-
         | news-...
         | 
         | - https://medium.com/@mi.schaefer/what-is-the-best-time-to-
         | pos...
         | 
         | But the precondition is that you're submitting high quality
         | content.
        
         | wirtzdan wrote:
         | Yes, I think so too.
         | 
         | At least for this post, I made sure to post it in the
         | afternoon. To have an overall between Eurpean and US time. (I'm
         | based in near Amsterdam)
        
       | SethMLarson wrote:
       | Love it, having an "aha-moment" is always a great indicator that
       | it's something worth sharing! I encourage others to share their
       | ahas so we can all learn together.
        
         | wirtzdan wrote:
         | 100%!
        
       | block_dagger wrote:
       | The "try-it-yourself element" was what got me. Very close to the
       | top of the article was a simple exercise that the reader could
       | perform themselves, and when it "clicked," it was magical. I
       | upvoted it for this reason.
       | 
       | EDIT: Typo
        
       | fallinditch wrote:
       | Ooops you did it again! Good analysis.
       | 
       | It makes me think: we may be living in a golden age when human
       | analysis and thought still produce the most interesting content.
       | Soon AI-assisted content generation processes will take the next
       | evolutionary step: doing proactive analysis and concept
       | development with the human being more like the back seat driver
       | and director.
       | 
       | AI powered writing assistants, music generators, life coaches,
       | etc, will get so good that authentic solo human creation may
       | become a rarity.
       | 
       | (Tangentially, if you ask Perplexity for tips on writing blog
       | posts that are likely to go viral on HN you'll get the above
       | post, only 6 hours old, as a reference)
        
         | wirtzdan wrote:
         | Don't you think there could be a world where the link to
         | something human might triumph what an AI can generate.
         | 
         | For example: Music is not only about how good it sounds. But
         | also a lot about who the singer is. Their history. What they
         | stand for and what story is behind a album or song.
         | 
         | With AI, I don't see that we will ever reach that dimension.
         | Expect maybe when AI becomes sentient?
        
       | jarenmf wrote:
       | the technique itself is very interesting and not very well known
       | to many people. also the writing was clear and to the point
        
       | vunderba wrote:
       | The fact that this article (which follows none of its own advice)
       | is steadily making its way to the top of hacker news sort of
       | invalidates all the actual recommendations in the article itself.
        
       | AndrewStephens wrote:
       | My highest rated HN link was a sarcastic troll post about the
       | MacBook Touch Bar in 2016. Nothing I have written since has even
       | gotten close to the same level of attention. Someone reposted a
       | more recent article from my blog over the weekend that gathered 4
       | upvotes and a few hundred views. It's nice to be noticed but I
       | don't live for attention.
       | 
       | I think that unless you want to form a cabal to get something on
       | the front page, there is a great deal of luck in what takes off
       | on HN (and other similar sites). Unless a post quickly gets those
       | all-important few dozen upvotes it will languish on the second or
       | third page wastelands.
        
       | Dilettante_ wrote:
       | Not having a link to the original article in the write-up seems
       | like a bit of an obvious oversight, in a piece about reader
       | engagement no less ;)
        
       ___________________________________________________________________
       (page generated 2025-01-20 23:00 UTC)