Analysis of New Steemit Users Aug Interim Report – Steemit Business Intelligence
Attracting new and quality users to Steemit every day is key in the success of the Steemit Platform and ecosystem. Retaining these users is equally important.
Remember this report for the month of July?
Well I have just looked at the data to run an Interim Report for August and I am slightly surprised at the changes
In the first two weeks of Aug 16,973 new accounts were registered on the Steemit Platform. That is almost 31% of the total number registered in July. But it is almost 25% less that the number of new registrations for the first two weeks of July.
Mondays and Saturdays were the most popular days for new enrollments in July. This has now changed to Wed and Fridays, with Sundays having the lowest number of enrollments in both months.
What is also disappointing to see is that only 39.37% of new accounts have made a post, compared to 44.85% in Aug. This drop could be of concern so I will make sure I monitor it.
22% of all posts made by new accounts are blog posts, 78% are comments. This is consistent with the July report along with the % of accounts that contain a link to a website or have the about profile complete.
Lets look back now at the 54K new users from July and see how many of them are still active.
First lets look at how many of these accounts were active for voting until 14 Aug. The Green represents the % accounts active when it comes to voting and the yellow represents the % of accounts not active. By the 14 Aug less than 17% of accounts set up in July were still actively voting.
Now lets look at how many of these accounts were actively posting until 14 Aug. Again the green represents the % of active accounts when it comes to posting and yellow is the non active accounts. What is very different from this chart to the previous is that with the voting, almost every account set up in July has voted at least once. Where as 50% of accounts set up in July have not yet made a post.
It is also interesting to see that only 10% of accounts set up in July are still posting mid Aug. Which is less than the number of accounts that are voting.
Many of these accounts that are actively voting and not posting could be bots. What do you think?
Did you like this analysis? Is there something else about new users you would like me to include in next month’s report? Have you any questions on any of the above?If so then comment below.
Steem On Everyone – I am really enjoying carrying out these analysis….follow me for more!!!
Hi Paula, thanks again for some great analysis and stats. Please do keep them coming.
The drop off rate is rather worrying. I guess posting drops off faster than voting as it is easier to vote than to post. And the time v reward equation would weigh towards a quick read and upvote v an hour or so making a post.
These numbers would support the graph I have found recently on https://steemdata.com/charts . If you look at the first graph on there of Daily Active Users there appears to have a been a rather cataclysmic drop off around 6 August.
I do really hope there is some glitch on the data feed on this. Do you have the facility to do a comparable data analysis like this?
Hi @pennsif, I have been extracting data from steemdata for my own analysis and it seems the datafeed to that graphic is lagging behind the blockchain.
I am not sure when this will be remedied. @furion did a post about it a few weeks ago. https://steemit.com/steem/@furion/updates-on-steem-python-steemdata-and-the-node-situation
I believe @paulag sources her data from elsewhere which is more up to date.
That is reassuring. I was hoping the data was not correct.
from what I can see there has been no glitch in the data :-(
Eeekkks
This sort of thing is natural in platform growth. There will some articles that spread the word, all of a sudden there is a stampede of people come to see what the latest thing is, do not understand it at first and then leave. I did the same a year ago.
Expect waves of people. Sadly, also expect waves of bots as there is profit to be made with these.
This post has received a 1.04 % upvote from @drotto thanks to: @banjo.
nice, really nice post.
It takes a lot of time and commitment to blog on steemit. Many of us have thought it's just too much effort for too little return. I suspect that most new users find out that's it's not a get rich quick scheme and go back to their old bad habits of Facebook, Twitter, etc. Those of us that get it though stick around and build our accounts.
I actually think a percentage of those people will come back. One of their friends will mention it, or they will read about it on a website, or something else will bring them back for a second try. Come winter when the sun goes down earlier and it's cold outside people have more free time on their hands. My bet is Nov. & Dec. we will see a big wave of new & returning users blogging about the holidays.
Interestingly some time back I did an exercise to download data from steemwhales.com. I only got up to around 80k acounts. I messaged the developer of that site and the reason for a much lower number of accounts than registered was
At the time there were over 160k accounts registered. This suggests to me there are a lot of inactive accounts on Steemit, so looking at registered accounts is not very robust metric. Its much better to look at active users.
Active users are very important. Understanding why people become unactive is also important in improving steemit for everyone. I would love to find a way to engage those unactive people, catch them before they drop off....
I think there were so many 'non-standard' accounts being created in July that it's hard to compare across the two months. I've been impressed with all the interconnected spam accounts that @sherlockholmes and others have found. I recall posts during this rapid growth where at least one person was bragging about making lots and lots of accounts that match celebrity names, with the hope of selling those accounts when Steemit hits the mainstream. Simply counting accounts doesn't account for nefarious intentions associated with many of them. Maybe a better count is the number of active authors over a week or a monthly period.
Hi Haphazard-hstead and thank you for taking the time to read and comment on the analysis. yes a lot of spam and non standard accounts do appear to have been set up in July. but this is the first time I have heard about anyone bragging!