I am exploring statistics to examine the growth in activity on Steemit. What is the best way to measure growth? Today I will have a focus on user accounts.
In this post I will cover:
- Account Creation Activity
- Distribution of Reputation
- Distribution of Steem Power
- Introduce some analysis by account duration
Analysis by duration will lead to some key stats about where contributions for posts, comments and other activity on Steemit are coming from.
Creation Date
Lets first look at how many accounts are being added to Steemit each day. This is should give a good indication of the growth of the platform.
There has been strong activity in the last couple of months. This metric may be affected by people setting up ghost accounts, bots, setting up accounts but not using them etc. The daily number of accounts may in fact give a very misleading figure for real growth on Steemit.
Reputation
All new authors will eagerly try to get their reputation up. There are many accounts sitting at rep 25 so I have excluded these and the following graph shows the distribution of the rest of the accounts.
It takes a long time to get to the 70s club!
Distribution of SP by Account
It is interesting to see how SP is spread out among the community. This will be an interesting statistic to watch over time, as new users come onto the platform.
In the next graph I have summarised the SP in in each account by bandings.
In contrast to the amounts of SP in the bandings the numbers in each bandings is the completely reverse. i.e. a few accounts control the majority of the SP.
Duration of Accounts
Where are the contributions to Steemit coming from? Is it new users or veterans. I will post over the weekend some detailed analysis on this but today the next few graphs look at aggregate account data. I have grouped the data by number of months since the account was created and we look at the distributions of SP across Accounts by duration.
The majority of accounts are in the interval months 1-3. The last big spike was between 12-14 months ago, when the price of Steem was at its highest.
This last graph is the most interesting. The majority of the SP is still in the hands of the people who were here at the start.
Monthly Stats
I am building a dashboard of monthly KPIs for Steemit which will focus on the key indicators for activity and growth of the platform. Please let me know which graphs you would like to see in this Monthly Report.
Recent Related Posts
- Steemit Activity
- Machine Learning The Blockchain - Wordclouds
- Followmenomics "The Follow Me Index"
- Who does all the Chatting?
- Steem Payout Patterns
- Steemit Payouts and Author Reputation
- Post Payouts Since Hf19 Update
Thank you for reading this. I write on Steemit about Blockchain, Cryptocurrency, Travel and lots of random topics.
Well i think it will start to distribute but i am not sure how quickly this will happen.
The consequences if it doesn't are the majority of power remaining in the hands of a few, if that is the case other people will lose interest in steemit or fork the code.
Lets grow together :)
This post received a 3.5% upvote from @randowhale thanks to @tournego! For more information, click here!
thanks for your sharing
Thank you for the in-depth statistics @eroche
This is very beneficial for us all in planning our next steps :)
Very interesting information, I didn't know that so many accounts have a less than 25 reputation... I will be following these graphics in the future!
Quality post @eroche the last graph is very interesting. The early bird takes the worm I guess.
It is highly skewed by the Steemit account but that tells its own story. I think how the distribution on that graph changes over time will be very interesting.
very well done! resteemed
Thanks for the resteem.
sent you a message on discord
nice post. the reputation distribution is interesting. Hoping to hit 60 within a month...... I have no idea of the calculation for this, do you?
I am not an expert on this. It is on some sort of curve, so the higher you go the slower it builds up. I believe that it depends on getting upvotes from people with higher reputation.
"Head down and pull like a dog!"
yes I have defo observed that upvotes from those with higher rep than you helps
lol, from all I have been reading lately, probably 50% of those new accounts must be bots and fakes, if not even more than that...
That wouldn't surprise me. I think by looking at the number of unique posters per day you get a more accurate sense of the growth in the community. Most bots aren't posting, they are just voting and commenting.
Very useful graphs. Gives us a good insight. Is there a way you can get the total number of Accounts on Steemit quickly?
It takes a long time to get to the 70s club!
Will it be possible for you to make a separate viz for this showing how many months people took to cross 70s. Most of the people are not aware that the difficulty increases exponentially and they assume it will be a linear process.
The two easy ways you can get stats on steemit is to look at steemsql or steemdata. Both of these sources let you see the number of accounts quickly.
On your second point it would be interesting to look at how reputation has grown over time for some users but that data is not easily available. I do plan on tracking this data going forward so it should be possible in the future and a picture will begin to emerge.
Interesting read! What surprised me is that:
There actually the two that I'm currently the most interested in seeing in the monthly reports. Thank you for doing this :)
Not sure we have seasonal stats on Steemit yet. It is still so new with so many things happening each month. The accounts with the low reputation are ones that have been downvoted to oblivion. There are quite a lot at this stage.
Thank you. Concerning all the spam bots, it makes sense. Thank you for making these statistics.
I'd be interested in working out the bot %. Low reputation of course doesn't necessarily = bot. Interesting there are so many sitting around reputation=10
In the past you could get a sense of bot % by looking at the percentage of accounts voting at less than say 5% of their voting power, but with HF19 many people are now voting with lower power. I will have to come up with another way of identifying these.