A case study in how small UX issues can create unintended user behaviors
Medium has a great platform that can be readily used as a case study in UI/UX design. But there is one area where they are falling short, and it may be a good lesson for anyone out there building their own platform or app.
At the very least, it demonstrates the importance of getting feedback from users on how they actually use your product versus how you think they will use it — a core concept in growth hacking.
Medium thrives on getting authors to continually produce original engaging content, and get readers hooked on making Medium their single go-to source for recreational content reading. They have done a great job at creating a positive feedback cycle for authors where they get financial rewards — paid by number of views if they are in the Medium partner program — and they get social recognition through claps and comments from readers. It creates a powerful feedback loop. The better authors do, the more they get financial and social rewards, the more they want to write high quality content.
But there is one small place where the feedback loop is broken. And by making a small change in this area, it’s possible that Medium could increase user engagement.
Authors generally want to engage with the people who read their posts — it’s part of that social recognition feedback loop. Especially when readers leave long thoughtful comments on their posts, it’s a good opportunity for authors to engage with them directly. It’s rewarding to the author, and it rewards the commenter by showing that their time writing out the response was put to good use.
If a reader took the time to type out a comment, I feel like it’s only fair for me to take the time to write out a response. Articles on Medium are supposed to be a conversation after all.
Unfortunately, this is where we run into a small UI issue that may be stopping authors from engaging in this way.
For some odd reason, Medium has decided to classify articles and comments as the same thing. They are all listed under “stories.”
When authors use Medium and they want to find a previous article, they go to the Medium backend and look under published stories.
But if authors are regularly responding to comments and commenting on other people’s posts, this area gets flooded with comments that author has written instead of articles.
So although an author may want to engage with their readers, they know that if I do that their account will become more difficult to navigate.
Only Medium has the true data, but I suspect that if they dig into it they will see that authors are not especially responsive to article comments — and I suspect that this very small UI issue may be creating this not-so-small effect.
This is a great example of how a seemingly trivial UI issue can have a big effect on how people use a SAAS platform. It is antithetical to the purpose of Medium to disincentive people from commenting but that is the effect the UI produces. This makes Medium less of a social network and more of a traditional publication site like Buzzfeed where the authors submit their content and then wash their hands of it.
If Medium is truly invested in building communities around authorship, then even the smallest details of their UI should drive users to engage with each other — and reward them for doing so as much as possible.
To me, the solution seems obvious. Add a tab for Responses in the Medium backend. To make it easier, I even mocked it up in the image below to show how it would look.
If you look under the stats section, you’ll see that these comments/responses are already separated out. That means that they are tagged as separate post types somewhere in the Medium database. Thus the development effort required to implement this fix would be relatively small.
So if they are already tagged as different post types, why is it like this. I’m guessing it may have something to do with metrics. If all comments are considered “stories”, then Medium can boast a number of “authors” far in excess of what they actually have by counting many casual readers/commenters as authors. Again, only Medium has the true reason why or the data to explain it, but from my experience at SAAS companies, that would be my best guess.
A small tweak like this could potentially increase user engagement in what we in Growth Hacking call a low-cost high-reward experiment. There is no real danger. They could implement it for a few thousand users to start and see what happens. And while the danger is low, the rewards are potentially huge.
With a platform of the scale of Medium, even a small increase in user engagement (like being able to have real conversations with authors) could result in monstrous growth for the platform.
With Linkedin trying to push into their territory with their native publications (where you can directly engage with authors), squeezing out some additional growth might not be a bad move.
To all the growth hackers and UI/UX designers out there, I hope this very specific case study was useful or at least interesting. One of the great things about our industry is discovering how small changes can result in big behavior shifts — and I think this example exemplifies that well.
Hi, I’m Sean. I write about data science, COVID-19 statistics and risk, habits, mental health, mortality, and occasionally politics. I have degrees in both physics and data science and I’ve worked at particle accelerators, NASA JPL, a research nuclear reactor, and I’m now in the startup world. I’m a data geek who enjoys contributing to Towards Data Science, The Startup, Hackernoon, and Dialogue & Discourse.
Find me on Twitter and Linkedin.