STATS | VIEWS | READS | READING | MEDIUM
Why the Stats Vary So Wildly Each Day
Explaining the daily volatility of Medium’s Stats
Yesterday I had a net gain of 147 views in my total ‘Prev 30 days’ views. That happened because my words attracted 327 views on November 10 (header) and on October 12 they attracted 180 views (image below). 327–180 = 147.
What is the connection of those two days? They are the latest completed day of the current 30-day period and the last day of the previous one, respectively.
Those are the only two days that define whether each new day will contribute a loss or gain to the total views (and reads/fans). The intermediate days do not matter at all in the daily count. Not directly anyway.
Each day only the first and the last day of the 30-day window matter. The first day, on the far left, is the one that will retire to the previous 30-day period when your current day is over.
If your first day is stronger than your present day you get a drop in total views. If it is weaker you get a rise. It’s as simple as that. And the difference between them is what solely defines your total views/reads count right before the close of each day.
If both days are the same -e.g. both gained 300 views- then the total views will not change at all (they will when the present day ends, of course; I mean the last few minutes of your present day, right before it closes).
As you can see in the second screenshot I had two abnormally strong days on Sept 14 (412 views) and Sept 19 (409 views). When those two days ‘retired’ I had a severe drop in views.
In contrast when my very weak Sept 18 (74 views) day retired I had a steep rise in views. This happens because Medium’s stats are rolling, not per calendar month.
Medium set them up that way so that we do not see ZERO Views / Reads / Fans at the first few minutes of each new month and lose heart. It would also be mentally draining to start each new month from absolute zero.
When my four very strong days of November (header) which are in the ~400 views range retire in late Nov — early Dec I will also get a drop of views, but only when all four retire.
They are all roughly at the same level, so my drop will come when the last of them, Nov 3 (403 views), retires. That’s because the next week is on average 100 views weaker.
However if Dec 2 gains more than 403 views I will see no drop in total views when Nov 3 retires for the reasons I explained above.
I hope I explained this clearly and simply, but if you have any questions or observations please let me know in the comments. I was inspired to write this quick explainer after reading this post of Marta Henriques:
I started ‘spamming’ her comment section with my usual long comments and then decided to explain the behavior she observed with a new article. For the sake of Marta and her readers too :)
That’s a wrap everyone. This is my previous meta article. Not relevant at all in any way to this one, apart from both being meta. And both being about Stats. Of Medium. Perhaps they’re a bit relevant after all: