This was published on 9/30/14 on hatch, unedited except for a small redaction. See Hatching Inside Medium for context on this collection.
Yesterday, 12 people from Medium product development met for 5.5 hours to answer the question: “What should we focus on next?” Our aim was to define a set of priorities for the next quarter for each team, as well as to identify important projects that did not fit squarely in any team’s domain.
The theme of the day can be summed up thusly:
Make Medium obvious.
This is not meant to be a pun relating to our former company name. It’s an acknowledgement that Medium is awesome — and it’s actually working — despite the fact that it is hard to understand. Our collective acknowledgment is that it feels like Medium is rolling downhill with the brakes on. It seems possible to remove a lot of friction, making us roll (and grow) much faster.
It’s Hard
We know that putting thoughts down on the page is hard. I hope that one day we can tackle making the writing process itself easier — or least the editing process — but we can accept that friction for now.
Where we feel we introduce unnecessary friction is in many other parts of the system:
It’s hard to understand the formatting options in our editor.
It’s hard to know who will see your story when you publish.
It’s hard to manage, submit to, and understand collections.
It’s hard to find the stuff you want to read or even understand what type of content is on Medium.
It’s hard to understand why stories show up in your PRL in the order they do.
It’s hard to navigate around the site (especially on mobile).
It’s hard to understand what Medium is, who’s it for, and what should be published there.
Those are just a few of the things that should be easier than they are today.
That’s why, while we started the day saying we need to talk about both incremental improvements and game-changing ideas, we ended up only talking about improvements. And very few of the improvements we discussed involved adding things — they were mostly about making things that are already there make more sense.
You could feel a palpable sense of relief — excitement even — about the idea of cleaning up and fixing various broken bits of the Medium experience.
Here’s some of the specifics, broken out by area:
General
Our goal by the end of the day was to have high-level priorities for each product team — Writer Success, Discovery & Delivery, and Interaction. We made a fourth category, called “General.” It was for the stuff that fell in between or across teams. Giving a forum to talk about these issue was one of the reasons for having our big meeting. And the issue that came up in this category seemed to be some of the most resonant (if daunting) during the day.
Specifically: Our site navigation is broken. It obfuscates what Medium is and what is possible. And even if you know that, it makes getting around feel tedious. It hasn’t evolved in a long time, and we decided to make fixing this is a major priority. This means looking at the entire site, mapping how it fits together, and reconstituting it in a more coherent way.
This means questioning things like: the left-side shelf, the metabar, post-to-post navigation (read next), and other bits of Medium that keep one part connected to another. In other words, not the middle of story pages or listing pages, but the chrome of the site — which means every page will be affected. This is a high priority, because it’s the framework in which everything else lives. Done well, it makes it easier to add other stuff. In the state it’s in now, it weakens everything — like a bad foundation of a house.
We discussed redesigning the navigation mobile web first. Mobile web is the vehicle for a huge percentage of our traffic, and we often treat it as a second-class citizen. By designing for the small screen, we’ll not only optimize that experience, we’ll force ourselves to simplify and prioritize. We’ll then scale it up to desktop web and translate it to the app.
Because this project doesn’t fall in any one of the project team areas, we are going to have to figure out how to resource it — largely from design.
Discovery & Delivery
The Discovery & Delivery team will be working to make Medium make sense from a consumption standpoint. One of the biggest things we can do in that respect is making the logged-in PRL be more comprehensible. That’s what the reverse-chron PRL project, already underway, is about.
In addition to this, D&D will be attempting to give readers — logged in and not — a better understanding about what is on Medium and helping them explore it. One idea everyone got excited about was a category or topical based approach to organization. E.g., be able to browse the best/current stories in, say, Politics or Science, as well as Ferguson or Apple Watch.
We don’t know if we’ll actually tackle this category idea and, if so, how posts would get assigned to categories. (There was a consensus that user tagging wasn’t necessarily the place to start.) But I’m a big fan of the idea of topics potentially having editors that are curating the best stuff under each banner (some of which bubbles up to the homepage). If not this, we’ll find some other way of making it dramatically easier to explore and find the delightful content that is being published every day.
One of the reasons this is such a high priority is because if you don’t understand the system, it’s hard to be excited about publishing there. I suspect improving the reading experience will have a significant effect on publishing metrics.
Writer Success
When it comes to writing and publishing, again, we feel like a lot more clarity — and a lot less friction — is needed.
Educating first-time writers about how our editor works is one area of improvement. Making the process more efficient for repeat writers is another.
Interaction
Lastly, Interaction is going to make sense of interacting. Actually, they said: “Make an interactions action system.” Our feeling is that Notes are cool, but broken in many ways. Responses show promise, but there should be a clearer relationship to Notes. And recommends and paragraph-level likes are also interesting but need to be tied into a clearer, more cohesive system.
It’s Easy
Nothing on the list above is mind-blowing on its own. In fact, much of it is mundane. However, when you put it all together, you get:
It’s as easy to write as possible and clear how to apply just the right amount of formatting and production value.
It’s easy to understand how your story is distributed, build your audience, and know how you’re doing.
It’s easy to find compelling stuff to read, no matter your interests, which compels you to visit often.
It’s easy (and fast) to move around the site, from any device; therefore, browsing Medium is fun, fast, and engaging.
It’s easy to engage with writers and readers to whatever degree you’d like.
When I imagine this, I see a Medium I love.
We’re not done inventing
To be clear, none of the above should imply we’re not interested in game-changing ideas. At the company level, we are continuing to invest a lot of resources into ████████████████████. And each Medium product team will continue to ideate and think about how to experiment with new stuff. But there is a pendulum in any creation process, which swings from “flare” to “focus.” Given we’ve been trying lots of new things lately — and a lot of those things are working! — right now, for the next little while, it seems right to buckle down rather than blow things up.
Once we’ve tidied things up a bit, we’ll have the opportunity to invest on top of a robust framework, rather than bolting things on to a precarious one.
What’s Next
Dustin is starting to tackle the site nav project immediately. Each team is defining and setting their own priorities with this new shared understand in mind. I’m very excited about the near future.
We have a lot to do before the end of the year. It’s going to feel really good when we do it, and I’m confident it’s going to have a huge impact.
