Do larger payouts help or hinder the rate of adoption? A/B testing can answer this.

in steemit •  8 years ago  (edited)

Calling all data scientists, anyone who would like to crunch these numbers.

The question, do certain trending articles have a positive impact on the new account create rate for the date?
The question, do larger payouts have a positive correlation with a higher account sign up?

So if you do an A/B test, you can look at a day where payouts are within a smaller range but more dispersed and compare it to the day where there was the huge payouts on the trending line. Do more people join when they see higher payouts on trending or do less join? Or what about if the payout amount in Steem Dollars is hidden for a week and compare the results to find out if it helps or hinders adoption?

I'm not a data scientist or statistician but there are a lot of questions about whether huge payouts are good for marketing or bad for marketing.

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:  

It is more important to see what impact large payout has on retention ratio and not on sign ups. I have no doubt 10K USD payout attract new sign ups. The question is do they make most people leave when they realize those payouts were a marketing trick to get them in and virtually no one is being paid that regardless of the quality of the content they publish

  ·  8 years ago (edited)

Are you on slack? Craig grant is doing something interesting with pooling posts together basically he is using his name and promoting certain post which does two things make his "brand" larger and gives a more focused audience. You roland and few other of the more technical people are becoming my favorite content perhaps we could do something similar? Craigs post I did an experiment where i posted with the tag and without it. When i used the tag I received many more up-votes. This could get the developer and tech hobbyist community on steemit a chance to carve a niche section out for ourselves. Im not saying we use his tag but make our own maybe "techonsteem" or something.

  ·  8 years ago (edited)

I'd say it's too early to make this type of study. I think users sign up as soon as they hear about steemit and are convinced that it's worth trying it. Probably the big rewards make people talk more about it but a lot of smaller rewards also shows that anybody can make money. Anyway I think the biggest factor in the signup rate is word of mouth.

Why is it ever too early? Are you saying the population is too small or something to do meaningful statistics? I estimate we have anywhere from 20,000 to 30,000 accounts right now.

  ·  8 years ago (edited)

I just think the number of new sign ups compared to the number of high paying posts is not relevant right now and maybe never will be. People definitely signup because steemit pays but I dont think new comers go counting how many posts got more than $1000, how many got more than $500 etc...

@dana-edwards No one can speak from a place of authority on the subject. But I was in the as you call them "Economic Justice Warriors" camp until a UI bug caused large swaths of information to go missing from the website such as payment info.

Personal experience, seeing the big payouts before you get here is an attractor, but sets unrealistic expectations.

The people who get the big payouts rarely ever provide any value later on. The few who do get mega jackpot winings and then later can't scrape two cents together also get disillusioned quickly and forget that even 0.01 from this platform is more than they would be making at reddit or facebook.

As for me it spun my perspectives around. Without the payout information I just voted what I liked and ended up with MORE that day than I ever got trying to find whale bait.

If I were to design a UI for this I would leave some historical top10 highest payees list. But postings in general should not ever show the $ amount earned since it's no one's business but their own.

Incidently, upvoting and responding to anything with @dana-edwards or @stella-belle early is the fastest and most consistent way to turn a few hundred bucks a day for me.

Anyways, have you tried to dump the blockchain into neo4j to see what information can be extracted via higher dimensional analysis?

I think it will be better if the extreme earnings should be some lower and the lower ones some higher. This will keep people who not earn (much) in the platform. Beside this i think that the extreme high payouts scare new people away, thinking it's a scam / to good to be true.

This is exactly why I'm saying, collect the data, do the proper statistics, and you can answer this question. All I see is complaining and appealing to emotions like jealousy. I'm a rational individual who likes to see some numbers.

I would rather see 500 100$ posts then 10 5000$ posts thats for sure.

It's not about what you'd rather see personally. It's about analysis of the data to determine positive or negative effects on adoption. It could be that removing the $ amount from public attention is the best way to promote adoption. In that case it would be a UI problem and they could just keep the amounts hidden from view if it's having an unintended negative effect.

On the other hand my hypothesis is that it is having a positive effect. If the data teases that out then it should not only remain in the interface but should be displayed in a way to bring attention to it.

  ·  8 years ago (edited)

It is no good if people who are signing up don't stay because all the earnings go to a select few. We've also seen frustration with the small guy being unable to make any money spill into the press with "ponzi" and other accusations spreading in the crypto world - that is a bad development and is directly fueled by the high payouts.

There is a way for this platform to work better - but that requires users to be a bit smarter about what they vote for. I wrote an article about how to do it: https://steemit.com/steemit/@teatree/steemit-is-all-about-game-theory

I'm a rational person. Do the studies and let's get the numbers. Appealing to jealousy is all I can see from people saying that participants will leave in droves and it doesn't seem to be based on anything we know from psychology. According to psychology, intermittent reinforcement with variable ratio schedule is the most sticky of all. Facebook users can't leave in droves and don't abandon the platform because they don't get enough likes. They do leave when they find another platform which is more sticky but from the data I saw, Steemit is already nearly as sticky as Reddit.

Anyway, we need more data driven decisions rather than to just assume.

Yes I have that same question or probably similar. Check this out https://steemit.com/steemit/@thedailyhobbyist/how-much-steem-power-is-influential-when-voting

Its a tricky and broad subject.

Nice story informative stuff here i found.