Steemit Retention Problem - Analysis of Steemit Database
After a recent analysis of new users from the Steemit database, there were signs to suggest retention on Steemit was rather low.
You can read that posts here: https://steemit.com/steemit/@paulag/analyzing-steemit-data-using-power-bi
So I decided to carry out a quick analysis to see how many users were active so far this month (July 17) with posts.
Analyzing the Steemit Database to establish an average retention rate for posting on Steemit
To do this I looked at only 1 table of data in the database, the Accounts table. This table will give us the date that the account was set up and also the last posting date.
I found that so far in July 35,656 accounts have made at least 1 post this month. I used Power BI to work with the data and preformed a simple count on the number of rows in the table that had a post date of 1 July 17 to date. This is only 14% of the total registered accounts.
That led me to another question, the people that did post, when did they register?
Would you believe only 4% of the total number of accounts registered in 2016 have been active with posts this month. That’s pretty poor retention
A whopping 40% of users that registered this month have made a post.
But if we remove the first 2 month and the last two months from the calculations we can get a more accurate reflection of retention. 14% of total accounts posting this month includes June and July which have high % of new user’s posts, hence inflating the overall posting retention.
So excluding the first two and last two month, the average retention for posting is 6%.
I did have some interactive reports with this data, however I can not embed them in Steemit, so here are some of the graphs I produced
Now that we can estimate the % posting retention, what about the users that have not posted this month?
The 86% of registered users.
Well these accounts are not all inactive. Some people don’t post, just vote (Some of which would be voting bots). I think my next analysis will be on the accounts that have not posted, see if I can get a better understand. Obviously I will share the results with you!
Why is retention so low?
I am a minnow, so my experience on Steemit is limited. But in the short time that I have been here, I have witnesses some stuff that would surely contribute to the low retention.
The first thing is how people are being directed to Steemit. There are many videos on Facebook and Youtube that sell Steemit as basically a get rich scheme. People are coming on looking for a quick few bucks. But to their disappointment, their first few posts don’t earn them much and they leave.
Next there is the learning curve. If you don’t come from the Cyrpto world there is a lot to get your head around. The money flow, Steem, SBD, Steempower, powering up, powering down can all be overwhelming for some. And then there is wallets and exchanges too.
And the last thing I will mention is whales and clicks. It can appear to minnows that whales vote for whales, if you are not in the click then you don’t have much of a chance.
Big thanks to @arcange for making this possible with SteemSQL!
It is hard to want to make a good quality post that takes hours, when you see some hot chick post a hello, and make a decent amount on her story. I'm curious what kinds of demographics are Steemians, and what floats everyone's boats.
I have been looking at the 'about' section on new profiles. Only 18% of new users in July have completed their 'About' section. I hope to do some text analysis on this later
Good analysis and very true. I recently created a flyer that was focussing on the "get paid" selling point. Other people have raised the issues you have in this post and I am recreating it with the "decentralised social media" selling point in mind.
I think this in the long run will help bring in higher quality users.
Thank you for your comment and feedback. I think I say your flyer, maybe you could post it here in this thread to help other minnows that are reading this post
Sure :) I'll get it altered with the new selling point and post it here.
Honestly I think that there are a lot of quality users but lots of them stay undervalued, just because... If I take a look at the trending section I hardly see quality, I see whale voting circles and sometimes crypto stuff.. The gems are not in the trending section, and that's a flaw in the system imo.. But we are in beta so I'm accepting this. I can't program a solution for it, so I'm accepting it..
It is great you are getting the info to keep us informed.. good post
thank you
Thanks for sharing. Very interesting. I'll have to look into the data sometime myself, as that's the type of thing I do for a living.
I love data
Nice post..maybe others create an account without knowing what to do about their log-in pass..and got lost their pass..
I would say more likely a huge number of accounts from 2016 are accounts that were created to exploit the fact that Steemit Inc. was giving out free Steem.
I was not aware of this. Pity they wont give me free steem now lol
I think that's a very possible scenario as well; I created an account for a friend via anonsteem and he had things really confused and almost lost access to his account ; but thanks to the password recovery thing he was able to restore the access ; still trying to help him figure out some things but new users can get all confused; yes
Big time confused. I would say I am good with tech, and as an accountant, I am good with numbers and stuff, add this to working online for 7 years and I am still confused (but getting there)
you sound like you are learning your way around here just fine @paulag ! I didn't think you were a minnow from reading this until you mentioned it. Nice post, many of the same thoughts I've been pondering too.
This is true too.
6% is low for a social media platform. I think you are right that those people who believed that steemit is get rich scheme left after getting frustrated with the result of their first few posts.
Im hoping that the people who are still here will continue to contribute. :)
there was a big spike in intake in June. A lot of this can be attributed to 1 person that joined in may, and although he is getting the name out there the quality of the user that he brings is not what steemit really wants ( people that think they can get rich quick).
6% is very low, but remember, this is still in beta and also the points raised by @moataz above are very true
There are also some accounts that are obtained for future value. Such as microsoft, intel, coca cola etc.
Congratulations! This post has been upvoted from the communal account, @minnowsupport, by paulag from the Minnow Support Project. It's a witness project run by aggroed, ausbitbank, teamsteem, theprophet0, and someguy123. The goal is to help Steemit grow by supporting Minnows and creating a social network. Please find us in the Peace, Abundance, and Liberty Network (PALnet) Discord Channel. It's a completely public and open space to all members of the Steemit community who voluntarily choose to be there.
If you like what we're doing please upvote this comment so we can continue to build the community account that's supporting all members.
Great work @paulag!
I believe that a good percentage of these accounts were created automatically by some software, but it'd be very hard to detect those accounts.
I also believe that many people register accounts and forget all about them; or at the very least lose interest in Steemit; as some people had to wait for days if not weeks to get approved; I think that's a major problem that puts off a lot of new members; but we should wait and see how this issue will be addressed in the next HF.
A starting point in detecting these accounts would be voting, and delegated steem but no posts.
Very interesting, I think this could also be due to human nature, many want to see results fast, and when this does not happen lose heart and move on to the next promised gold rush. SEO was what drew me to Steemit originally and the opportunity of learning about crypto currency, the community and the huge learning curve is what keeps me here now. Thank you for the post.
I enjoy to learn myself, but not everyone wants that. Thank you for your feedback