Steem-Stats 18-03: Zappl

in #statistics7 years ago (edited)


sv-stats.jpg

Hello,

when contributing to a project, in this case the Steem blockchain and its platforms, it is vital to know how your plattform is doing.

So I have taken an extended look at the data provided with the Steem blockchain.

Regarding the collected data, all the posts on steemit that used the "zappl" tag were extracted via SQL.

Of course, this inevitably unsharpens the results to a certain amount. Every post on Zappl carries the "zappl" tag. However, it is possible that posts not posted on Zappl still carry a "zappl"-Tag. For example a video of me on D.Tube, which is about Zappl and therefore also has the "zappl"-tag. With more than 177,000 contributions per month in some cases, however, this statistical error is unlikely to have a major impact.

Info: All graphics open a new tab when you click on them, so you can easily hold several graphics next to each other to draw comparisons.
Zappl-Posts-E.png

On the first image it looks as if Zappl was hardly used for a while. But that's only partially true. More correctly, there was a real explosion in postings in February 2018 and March 2018. Every beginning is difficult and no one can expect a new website to take off from the get go. So slow and steady growth should be expected.

Therefore, a second image that only shows data till January 2018 and thus hides the sudden rise afterwards:
Zappl-Posts2-E.png

Here it is already more clearly recognizable that after an initial interest in the platform it went downhill again. I take till October to get a month performing better then the initial month of the plattform.

On the other hand: The number of posts does not tell us about the number of users using the service. But it is quite important to know how many unique users are using the service.

This also helps during the explosion month of February 2018. Do we have the same users just putting out mor content, or are there actually more users?
Zappl-Users-E.png

So there was also a big increase in the number of users in February 2018. A number that hasn't subsided at least a month later.

The third angle comes into play. How much revenue was distributed to the posts on the platform?
Zappl-Earnings-E.png

This is the first shadow on your plattform. Distributed revenues have not increased in line with user numbers. Although February 2018 is more than twice as good as March 2018 or May 2017, the posts were 108 times higher and the number of users 38 times higher.

Therefore, in the second part, we take a look at the "distributed" funds. Have these been distributed among the posts to evenly? This would of course be both desirable and important for a platform to grow. Disappointed users are likely to turn away too quickly.

In the first month, in May 2017, the ratio was as follows:
Zappl-Top-30-Mai17-E.png

The top 30 users, representing approx. 10% of all users, cashing in approx. 93% of all earnings.

Zappl collected the highest amount himself. We don't want to take offense at this point. It's still early. :-)

So let's move forward a little.

June 2017:
Zappl-Top-30-Jun17-E.png

July 2017:
Zappl-Top-30-Jul17-E.png

August 2017:
Zappl-Top-30-Aug17-E.png

Up to here, basically always the same game.

Basically the payments to often went to the same users... with corresponding variance, of course. I think that's a bad thing, and obviously I am not the only one thinking that.

Because in September there was a change.
Zappl-Top-30-Sep17-E.png

It may not be noticeable at once, but the account Zappl usually was in the 4-digit range. But in September it was only about 284,-. So why did it turn out so moderate in September?

To do this, we have to take a look at the votes that the corresponding accounts have received.
Downvotes-E.png

Here you can see that the Zappl-account got over 6604 downvotes. Why over 6604? The votes numbers shown reflect the so-called net votes. For example, if the account "steemquebec" has a value of 28, this could mean that he has received 28 votes up. But it can also mean that he got 128 votes up and 100 votes down. The number of downvotes for the account "zappl" may have been considerably higher. Compared to the previous months, it is safe to say that the downvotes have erased some of the revenue.

In October history repeats itself:
Zappl-Top-30-Okt17-E.png

As a result, almost 30% of all earnings is no longer distributed to the top 30 accounts. The "punishment" is again considerable:
Downvotes2-E.png

At least they had the desired effect in November. This time there were no big downvotes. Nevertheless, the "zappl" account has stopped at income, around 350,-.
Zappl-Top-30-Nov17-E.png

In the last month of 2017, the user "zappl" has again achieved a little more with 500,-... but at least the other 98% of non-Top-30 users get almost 35% of all earnings. Let's take that as a small success.

Let's fast-forward some more...

January 2018:
Zappl-Top-30-Jan18-E.png

February 2018:
Zappl-Top-30-Feb18-E.png

March 2018
Zappl-Top-30-Mar18-E.png

In the last two months (February and March), earnings have increased significantly compared to the previous months. Here the top 30 accounts got only about 30% of the earnings. This is, of course, still very much in itself, measured by the fact that these top 30 users represent only 0.25% of all users. More important, however, are the other 70% which are distributed to all other users.

And of course, it's not like the top 30 users are always consisted of the same user. With one exception of course, there also was a fluctuation, as the next chart shows:
Zappl-top30-top30-2.png

The current observation horizon is 11 months. In this respect... one account has always been represented.

In the right part of the image, there are some accounts that have been in the top 30 4 times but have not achieved a big financial advantage, as the next chart shows.

The income from this is as follows:
Zappl-top30-top30.png

So let's finish this review and see how it will develop in the future.

The account "zappl" recently has lost a large delegation. Of course, it is not clear if this was related to its behaviour.

Okay.

Sort:  

Nicely compiled data and translated ;)
Looking forward to more analyzed data on Steem.