Steemit Web Analytics: Retention and Engagement. Part 1

in steemit •  8 years ago  (edited)

Morning steemers,

I'm currently mulling over investing in steem. In this process I'll be looking at several factors that will help determine if I move forward with that decision. I figured I'd share this information so anyone who has similar thoughts or questions has a resource. This data comes directly from steemd and accounts for the prior week from 8/2/16-8/10/16. Steemd tracks the daily, weekly, and total amount of active accounts by steemit user's steem total. 

Over the course of this post I'll be doing a bit of a dive into each metric and what I think this can/will/should mean for steemit. 

Why is this important for investing in steem? 

Well, steem is very much tied to the growth of steemit. Without steemit, steem is just another sheitcoin. 

This past week steemit saw a surge in daily active accounts. From an average of less than 4500 to a peak of ~6400. Substantial growth, but not random. The 6th and the 7th were the same days @dollarvigilante posted videos to his YouTube account which reached several thousand viewers. A combined 37,000 at the time writing. This is very likely to be a short lived bump, but will see our new average daily users still higher because of it. @dollarvigilante 's audience is very similar to what you'd expect to see on steemit. It's easy to imagine steemit has gained some committed followers because of him. We ended the week with an article from the International Business Times detailing the ideas of steemit and it's founders. 

This week, we saw average daily active accounts at approximately 10% of total accounts. 

The Pros: We can get substantial growth by engaging with outsiders who have even minor followings. 

The Cons: We gained another several thousand members who's ranks will likely swell the already gorged section of our site. Which means we'll be garnering very little variation in content. 

The Active Accounts End-of-Week chart tracks users on steemit for the week ending in that date. So data from 8-8-16 is from the previous six days and includes the final day. It's no surprise we see growth from 8.7 to 8.10. Between the two shoutouts mentioned above we of course received a ton of unique users day over day.

This week we saw average weekly active accounts at just over 28% of total accounts. Starting the week with a downward trend before rocketing up, on what I believe to be major influence from @dollarvigilante.

The Pros: It's going to be great if we're able to retain those users, but going into the end of the week we were seeing a drop-off in unique visitors. Something that could very well continue without fresh advertisement or new content.

The Cons: Only looking at one week's worth of data we have no idea what was driving those users away or if that trend would have even continued without the outside support steemit received towards the end of the week. 

Finally we have the above two graphs charted against total account growth. The "All" category will always show growth or stay stagnant. What we're looking for is times of growth across all three metrics. That would imply new users are coming in, returning, and are active daily. Crucial for the growth of a social media site.

Total accounts grew by a whopping 23%, something I expect to continue with each new major contributor, but I do not expect that to be the norm in a few months. 

What's next?

A look at steemle, Google Trends around steem keywords and our competitors, and where our site traffic is coming from.

Feel free to follow for more information on the state of the site. I plan on making this a weekly or bimonthly update for as long as I see value in investing time/money into steem.

Disclaimer: I am currently long steem. 


Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  

Interesting stats but I think some information is missing. For example, is the growth in active users a larger percentage of overall users, or does growth just correlate with the increase in number of accounts.

Likewise, attributing growth to user's like @dollarvigilante is premature. They certainly have the potential to bring in users, but it's going to be difficult to confirm that that's the case. I'm sure there was a post listing which accounts had brought in other users, though I don't remember how they calculated it.

Hey @bitcalm , I did post this about Jeff's Impact on STEEMIT in case it helps convince you that @dollarvigilante made a number difference.
@sharper Great post Up vote and can't wait for the next. I am now following.

Thanks streetstyle, let me know if you'd like to see anything else!

Growth is according to an increase in the number of accounts. And actually dropped in overall percentage of users. I did the calculation for it, but was unsure if I should include it. Now that someone has asked I know I should have.

I personally don't think it's premature, a lot of growth lines up with his arrival. All growth? No, and I should phrase it better, but we can definitely attribute some to him. Everything is difficult to confirm, but we can rationalize and use evidence to say one way or another.

I think a reason these users are not staying is not succeeding with early posts. While obviously it is unrealistic to expect success with your first post I feel it is very hard to get any traction if you only got 5-6 SP in your account.

I am not sure what the answer is, maybe a larger community with more specialised topics and a better distribution of the voting power

I agree with that probably being a reason. But it is a rather silly one. People don't leave other social media sites because they aren't getting paid. Regardless I'd love to be able to look at new user churn.

Specialized topics will definitely help with retention. The longer a user can stay on the site the more likely they are to mention it to others.

As I am new to Steem I like finding posts like this that may help improving overall understanding of steem, steemit... But yes this is a young system, need more time, build more historical data, establish control data to compare against, etc... I will be interesting to watch and grow with it

Agreed, and that's exactly why I'm doing this. I'm immensely curious about how the site will grow.

I think a lot of users may come on thinking they can get rich quick. When they see they have to actually contribute in some way, they scurry back to fb to mindlessly share someone else's creations. Just my opinion.