See this article by Matt Binder for some discussion of this data.
I'm happy to share the full data set if anyone wants to verify the numbers below. I've also uploaded subsets that make it easy for anyone to spot check.
In each case I've arbitrarily picked a prime number that would provide a sample of around 10k accounts, and then selected all accounts from Musk's followers that have the specified property and whose Twitter ID is divisible by that number (6,521 for the zero followers list and 6,257 for the zero posts list).
This makes it possible to spot check my data collection by confirming that these accounts follow Elon Musk and have either zero followers or zero posts. You still have to either trust me that the divisors I've chosen don't somehow pick out Twitter IDs with special properties (they don't), or verify this for yourself.
Note that in a very small number of cases (1.3% for followers and 0.6% for posts) these accounts have either gained followers or made posts since the most recent snapshot in the original data set, or have been suspended or self-deactived. These changes are indicated in the CSV files.
This data is based on a scrape of Elon Musk's Twitter follower list that began at 17:24 CEST yesterday and completed at 10:14 CEST this morning (17 August 2023). All of my software respects Twitter's rate limits, so the profile snapshots used here are up to three weeks old (the oldest were taken on 27 July 2023). In general more active accounts will have more recent snapshots.
The follower list used here contains 153,209,283 user IDs. A small number (5,968, or 0.0039%) were suspended at the time of their last snapshot, and a slightly larger number (39,485, or 0.026%) were self-deactivated. All numbers below use as a total the 99.97% of accounts on the follower list that were active at the time of their last profile snapshot.
Most accounts have almost no activity. 38,035,372 (just under a quarter) are using the default profile image. Over 40% (49,711,202) have more than three numbers in their screen name. The average number of followers is 187.3 and the median is 1.
| O1 | O2 | O3 | Median | O5 | O6 | O7 | |
|---|---|---|---|---|---|---|---|
| Followers | 0 | 0 | 0 | 1 | 3 | 12 | 55 |
| Following | 2 | 7 | 18 | 36 | 67 | 123 | 297 |
| Tweets | 0 | 0 | 0 | 2 | 7 | 34 | 326 |
Over 65 million (42.6%) have zero followers, and 62 million (40.8%) have zero tweets.
| 0 | <10 | <100 | 100-1k | 1k-10k | 10k-100k | 100k-1m | 1m+ | |
|---|---|---|---|---|---|---|---|---|
| Followers | 65,312,713 | 111,996,490 | 139,404,177 | 11,713,719 | 1,806,720 | 210,410 | 26,373 | 2,431 |
| Following | - | 43,882,195 | 108,569,333 | 39,468,803 | 5,085,324 | 39,313 | 1,040 | 17 |
| Tweets | 62,500,641 | 100,253,312 | 124,791,622 | 16,407,179 | 9,507,922 | 2,355,928 | 100,803 | 376 |
(Of Musk's followers, @BEMANISoundTeam has by far the most tweets: 7,099,740, almost twice as many as the closest account.)
Around a quarter of Musk's followers created their accounts after he took over Twitter.
| O1 | O2 | O3 | Median | O5 | O6 | O7 | |
|---|---|---|---|---|---|---|---|
| Account created | 2014-09-28 | 2018-01-08 | 2020-08-19 | 2021-11-13 | 2022-04-30 | 2022-10-29 | 2023-03-29 |
453,662 (0.30%) are marked as Blue subscribers. My current estimate is that at least 10k and probably fewer than 15k accounts are using the "Hide my blue check mark" feature, so I think it's unlikely that more than 460k of Musk's followers are subscribers. This is probably around half of all Blue subscribers (I've currently identified 834k).