Analysing article failure
Dear Reader,
This is more of a technical dive on blogging & reflecting. So it's not for everybody. But for those of you who are curious, join me on this analysis.
I was sitting and wondering, why do some posts catch fire while others barely spark? Let's dive into such a case, now that I've had the experience. Two blog posts: one, a hit; the other, a miss.
After a week long of pondering and trying to make sense of it, I decided I was determined to learn from this failure. (I've mentioned this in one of my newsletters).
This dives into that, the where did it go wrong and what I've learnt.
Be warned though, I don't have a success story to back any of this up, this is pure failure and reflection. So if the future changes, that will be another post for another time.
'Success is not built on success. It's built on failure. It's built on frustration. Sometimes it’s built on catastrophe.' – Sumner Redstone
Similiar posts, yet different:
This is where the complexity starts, two posts:
The first contender, a post on Mixtral, Phi2 & Dolphin, all three oozing with innovation, drawing in those with curious minds, perhaps wanting to know more. I'd say it was well-received, a tiny glimpse of success.
The second post A look at Google Gemini 1.5, published in it's private preview phase, a post written to be more polished, more examples, more thoughts. More & more yet it found itself receiving zero views, yes you read that right zero. The difference was so disproportioned that it left me with only valuable insights I could study. While a failure, I managed to find something of value within.
Let's do a quick analysis of what is different:
- Article read time (Major difference) – 14 mins vs 24 mins.
- An article involving Three LLMs (released around the same time) vs one LLM.
- Published not to far apart (Perhaps the Mistral article was actually badly received all along?)
Audience & Post Autopsy:
Even though both posts were LLM based and have similar structure, ask and answer similar questions in their fields, turns out that the audience was not there.
With the Gemini article displaying a 24 minute read time, maybe that was just too long for the casual reader and was ignored. Perhaps it failed to connect, possibly because it missed the mark on relatable content, or perhaps more simply a desire. Or perhaps you dear reader, might have had a taste of what the post layout is and perhaps that did not resonate with you to come back.
Conversely, Mixtral's article resonated more with the tinkerers of the "tech enthusiasts". People who are more interested in getting something to work than using out the box solution which may or may not solve their problem. This audience, driven by a thirst for knowledge and hands-on, saw more value in Mixtral. The allure was not just in the operational prowess of 'Open weight' LLMs but in the broader narrative of innovation, resilience, and the boundless lessons (And frustrations) from from DIY. Well let's not forget the obvious, with little traffic it's hard to fully uncover the why, so take it with a grain of salt. Even in hindsight here (at the time of writing) I am still not seeing 20-20.
Keyword Fumbling:
We're not playing buzzword bingo here. Real engagement comes from real interest that resonate with real people. I am not sitting here optimizing keywords but due having a growth mindset I naturally had to look into why it was happening, so let's look at it from a SEO perspective for once. SEO is far more nuanced than mechanical stuffing of phrases designed to appease Google. If you are going to purposefully add or chase keywords, adding too much or too little can alter your "formula". In this case sorry but "Google Gemini" is just ridiculously difficult to rank for. So the SEO savvy among you would not even bother with it as the ROI is bad. But in my case, well I write for own interests and speak my mind. But it's important to bear in mind that keywords become more than tools for visibility, they can clearly make or break traffic to a page.
Combining SEO and Common Sense:
It's not all black and white, there is harmony between the realms of "development" and SEO, common ground where both disciplines converge in their quest to connect content with curiosity.
For developers or those in the world of web development, the concept of "linking"—creating hyperlinks that connect one piece of content to another—is as natural as breathing. In the SEO world this technique is called "Internal Linking", and seen as a strategy. Other obvious things is adding <meta>
HTML elements and using the <title
> element. I recognized this is an area I failed in and has contributed to that post's failure. It not only guides search engines but it guides users as well. So I am not just doing a dis-service to myself but to the reader as well. By creating a simple information bridge I don't have to engage in the dark arts of SEO or cloaking the content in a veil of optimization tricks. Rather, it's about drawing from the basic principles that devs, designers, UX, marketers and SEO experts hold dear: creating clear, intuitive pathways that enhance the user's journey. A shared mission, yet took a failure for me to recognize that I missed something so simple. Optimizing for the human experience is as important. In this case, there is no conflict of interest they all work to making the users experience better.
Embracing & Learning from the Void:
With how easy it is to "succeed" and "fail", the true art lies not in never failing, but in learning and rising each time a failure occurs. Yes the story does not end in just recognising you failed, but requires the resolve to recognise & learn, chin up and try again.
Each overlooked hyperlink, each unclicked title, speaks volumes—not just of missed connections, but of opportunities for recalibration and re-engagement. In this case, getting no data was useful, but having some data would have been way more helpful in trying to analyse where it went wrong.
All of this has ultimately led me down the road of introspection and improvement. Which is why I felt the urge to publish this post. Yes it breaks the rules that nobody asked for this, but to not speak my mind would be a shame. In this case even though it might have been a good topic and an improvement over the last just because it has zero views does not mean there was nothing to learn from it. I took it upon myself to make it better than the Mixtral article and fully embraced it was going to be chunky. So I'd say if you're content is relevant and helps someone, do it because you can.
This is just one of those classic examples where setbacks are not the end but seeing failure as a stepping stone to greater success and understanding.
To not publish my findings and learnings would be just as regrettable, so dear reader I hope you managed to take something away from this.
A set back here, a minor success there, it just reminds me that life is not a straight line.