It’s always nice when someone working behind the scenes allows a peek behind curtain. A Medium support agent did that for me over the weekend.
I previously presented some musings on the impossibility of figuring Medium Partner Program earnings. Therein I demonstrated how the four aspects of “engagement” — page views, reads, fans (claps), and responses — could not reliably be used to estimate a story’s earnings.
Medium’s Compensation Black Hole
Is it fair that authors don’t know how they’re being paid?
That was before this little revelation provided by said support agent.
It’s (Almost) All About The Claps
In the main, my previous analysis still stands, but I was wrong about “engagement.” I discovered this when I asked Medium to explain a curious discrepancy between two of my stories. Let’s call them Story A and Story B. Their stats are as follows:
- Story A has 2.1K views and a 42% read ratio. It is regularly viewed and read, yet it has only earned $1 over its lifetime. For a recent week, it earned $0 although it was viewed every day that week.
- Story B has 179 views and a 36% read ratio. It’s viewed and read somewhat less regularly than Story A, yet it has earned $23 over its lifetime. For the same recent week, it earned $6.52.
Makes no sense, right? The difference, our friendly neighborhood Medium rep explained, is claps.
- Story A has earned 162 claps from twelve people. But four of those are not paying members, so the count for earnings purposes is 139 claps from 8 people.
- Story B has earned 335 claps from 29 people. Three of these are not members, so its count for earnings purposes is 313 from 26 people.
That’s where I previously went wrong. “Engagement” means claps and, to a lesser extent, time spent reading a story. Views don’t count for anything. Reads can be treated as a proxy for time spent reading and fans as a proxy for claps. End of the day, though, claps are the primary signal.
But wait, you say. Story B received well over twice as many claps as Story A. How does that translate into 23 times more earnings? It doesn’t directly, but there’s more to the story. Remember, Medium distributes a share of each member’s monthly contribution in proportion to their activity on Partner Program stories. Suppose I interact with three stories during the month, as follows:
- Story 1: 5 claps
- Story 2: 3 claps
- Story 3: 2 claps
Medium will take a fraction of my $5 membership —for sake of illustration, let’s suppose that portion is $1.00 — and distribute it as follows:
- Story 1: $0.50
- Story 2: $0.30
- Story 3: $0.20
Now suppose a more exuberant clapper responds to the same three stories thus:
- Story 1: 30 claps
- Story 2: 18 claps
- Story 3: 12 claps
What will the payouts be? Exactly the same, because we both allocated our claps the same way: 50%, 30%, 20%. The number of claps doesn’t matter. The fraction of a user’s total monthly claps received by a story is all that matters. You’ve no doubt noticed how some people give 1 clap for a story they like and others give 50. I tend to give 10 for a story I like. I’ll downgrade a bit for stories I like but which have flaws and upgrade for something really good. We can probably assume that most clappers are roughly consistent in their clapping, so a story’s fan count can stand as a proxy for claps.
Stories with more fans are likely better-earning stories, bearing in mind that not all fans are paying members. A story with a lot of non-paying fans won’t necessarily perform as well as a story with mostly paying fans. You can tell the difference between paying and non-paying members by their profile pictures. Open the clap list for any story. Member photos circled in green are paying, those not circled in green are not.
Remember Story A? It’s garnered 2.1K views and a 42% read ratio. That’s 936 reads. Yet out of 936 people who thought the story was worth reading in its entirety, only twelve responded with claps. That’s horrible. In fact, measured on fans as a percentage of reads, Story A is my worst-performing story, aside from three that received no claps at all. Yet it’s my most viewed and most read story overall. How does this happen?
The answer is found on the story’s stats page. Story A has garnered 96% of its visits from external referrals, the overwhelming majority of those being from Google. It’s a very successful story, the top ranking Google result for “how to use writing prompts.” But readers who find it that way are by and large not Medium members, so they can’t clap for it. By contrast, Story B gets 78% of its traffic from Medium and almost none from Google. Its readers are almost all potential clappers. My top-performing story, fan-wise, gets 39% of its traffic from Medium and only a quarter from Google.
Friend links and search engine optimization are great for attracting widespread audiences, not a bad thing, but you might be fooled by the resulting view and read statistics. Reaching a wider Medium audience — through a big following and placement in popular publications — can lead to more fans. But come payday, the only thing that matters is how paid Medium members interact with your stories. And therein lies the problem for anyone wanting to understand their Medium income.
Waiting for Transparency
Untangling the influences behind the stats and getting to what matters for payments is bad enough. To make matters worse, Medium makes the relevant stats opaque. Even knowing that claps are the primary signal used in calculating payments, we have no easy way to tell how many members clapped for a given story in a given pay period, nor is there any way to compare clap performance among our full list of stories.
In my email exchange with Medium support, I offered a suggestion regarding how they might begin to address this deficiency. In return, Medium acknowledged that many writers have asked for more transparency in the payment system. That’s good to know. We can all hope they are serious about improving reporting. Until then, think of the available stats as falling into two broad areas:
- Stats describing reach: Your views and reads tell you how many people find your story what percentage of them spend enough time with it to read the whole work. These people may be guests on Medium, non-paying members, or paying members, so this has nothing to do with how much you’ll earn. Use story stat pages to learn how readers are finding your work.
- Stats related to earnings: Although claps are used as Medium’s primary signal in payment calculations, the fan count may be a more reliable indicator of earnings, because each fan’s interaction with your story counts only as a percentage of their total claps for the month. Neither fans nor claps are guaranteed indicators, however. At best, they are estimators.
Until Medium comes up with something clearer, that will have to do. But it’s probably enough for now.