Exploratory Analysis shows DApps activity is gaining market share on Steemit.com

Last month I published an article showing the downtrend in benefactor reward in 2018.  If you missed that post you can check it out here.

https://steemit.com/utopian-io/@paulag/exploratory-analysis-of-benefactor-rewards-shows-dramatic-impact-of-downturn-on-dapps-1540645248384

Benefactor rewards changed over the year for a number of reasons.  

1. Less users

2. Drop in the price of STEEM

3. Some apps changed their rewards %

Because of this the charts show in this analysis did not reflect the usage of these app and so I wanted to have a quick look at how the much the user base and the number of posts has changed over the year from the different apps. 

 Repository

https://github.com/steemit/steem   

Aim of Analysis

• Explore and visualize the activity from apps posting root comments to the steem blochchain.
• Establish what % of root posts are made via Steemit.com and how this has changed over the year.
• See what apps are growing in terms of root posts.

Exploratory Analysis

 

In the chart above we can see in the blue the distinct number of apps posting to the blockchain each month.  As December is only a part month, we can really discount this as it may not show a true reflection.

Beneath this shows the distinct number of authors posting to the blockchain each month.

The number of authors per app decrease considerably more than the number of apps posting.  Using an animated scatter plot this is considerably easier to see.

This first scatter plot show the number of posts made to each of the apps each month plotter against the number of authors.  We can clearly see Steemit way out ahead.  Esteem and Busy also shine threw. 

If we remove Steemit, Esteem and Busy we can get a clearer image of the other popular apps.  You can see Zappl doing well and then disappearing and also Dlive. Dmaina has dramtic reductions as does Utopian. Steempress seems to increase as many other decrease and you can also see the rise of parkito, actifit and some of the other new apps around October. 

Lets look now at these same metrics on a year to date view.  Again it is very clear to see how far out ahead steemit.com is when it comes to author activity. Esteem and Busy are also very clear to see.

Removing these 3 accounts we can get a better view of the other popular apps.  We can clearly see @steempress racing along from May with the numbe of posts but not increasing in the number of users.  Zappl stops moving in June and Dlive in September.  Dtube, Steepshot and Dmania all doing well.  

Removing Zappl, Dlive, Dtube, Steemshot and Dmania now in addation to Steemit, esteem and busy lets take another look.  Unfortunate this would not upload as a gif so I had to take a still from month 12.

  

As well as the apps that are preforming well, from this we can see a new cluster of apps that are on the rise.  A table view can be seen of the top apps sorted by year to date posts displayed by month.

 

If we sort this no by % per month we can see that in Jan Steemit.com made up 68.8% of all root posts made and now its standing at 51.61%

    

Conclusion

Steemit.com remians the most active app posting to the steem blockchain.  We can also see a rise in the activity of the other apps.  This is very encouraging and I hope to see the gap between Steemit.com and other apps close more in the next 12 months.

Its fantastic to see apps llike steempress, dtube, partkito, actifit and dlike take a market share of the posts in the last few months.  Its also interesting to see Dmania make a comeback.

I would love to see the same analysis done with comments, however the query for the year is way to big for me to analyse in this way using the tools I have.

Data and query

As always I connected to Steemsql managed by @arcange and I used Power BI to preform the analysis.  The M code used to gather this data was

let

    Source = Sql.Database("vip.steemsql.com", "DBSteem", [Query="select author, permlink, created, json_metadata#(lf)from Comments#(lf)where CONVERT(DATE,created) BETWEEN '2018-01-01' AND '2018-12-31'#(lf)and depth = 0"]),

    #"Changed Type" = Table.TransformColumnTypes(Source,{{"created", type date}}),

    #"Parsed JSON" = Table.TransformColumns(#"Changed Type",{{"json_metadata", Json.Document}}),

    #"Expanded json_metadata" = Table.ExpandRecordColumn(#"Parsed JSON", "json_metadata", {"app"}, {"json_metadata.app"}),

    #"Filtered Rows" = Table.SelectRows(#"Expanded json_metadata", each true),

    #"Inserted Text Before Delimiter" = Table.AddColumn(#"Filtered Rows", "Text Before Delimiter", each Text.BeforeDelimiter([json_metadata.app], "/"), type text),

    #"Filtered Rows1" = Table.SelectRows(#"Inserted Text Before Delimiter", each true),

    #"Removed Errors" = Table.RemoveRowsWithErrors(#"Filtered Rows1", {"Text Before Delimiter"}),

    #"Filtered Rows2" = Table.SelectRows(#"Removed Errors", each true)

in

    #"Filtered Rows2"

 

Shameless Promotion

 Like what we are doing?  Dont forget to vote for @steemcommunity as your steem witness.  You can vote for our witness using SteemConnect here: https://steemconnect.com/sign/account-witness-vote?witness=steemcommunity&approve=1