This is the latest post in my series on user retention on Hive, and a direct follow-up to yesterday's post about onboarding. In that I presented the data on user onboarding network wide, and tried to determine if users perform better and/or are more likely to stay with Hive depending on what account created theirs.
In today's post I will take a narrower focus and look at one on-boarding project. As far as I'm aware, OCD on-boarding project is the only one of its kind on Hive, where volunteers (whom are first vetted) are incentivized to encourage users to join using a referral system. These referrals get embedded in the metadata of the transaction that creates the new user's account, which allows the incentive system to work and also allows us to extract the data for some retrospective analytics.
Overall Stats
Here are the overall stats for the OCD program, with some network averages for comparison.
| Volunteers | Onboarded Users | Became Active First Month | Second Month Retention | Third Month Retention | |
|---|---|---|---|---|---|
| Overall | 60 | 745 | 548 | 345 | 210 |
| Rate | 12.4 per Volunteer | 73.56% | 62.96% | 38.32% | |
| Network Averages | 16% | 37% | 27% |
How Onboarded Users Performed on Hive
Since they joined, users onboarded by the OCD initiative have made 12,637 posts for a combined author reward of $53,692. The average author reward per post by an OCD onboarded user is $4.25, compared to about $1 network-wide. I have not gone to this level of depth, but it may be interesting to compare post length, votes and comments per post in the future.
Here are some charts showing the activity of users onboarded by OCD. First is monthly active authors, followed by posts per month, then author rewards per month. Each element is colour-coded according to the OCD volunteer that onboarded them.
Keep in mind that for most of the period below, Hive userbase, posts per month etc. were in decline.
Individual Volunteer Performance
Finally, let's look at how many users each volunteer has onboarded individually, how many of them made a post in first month, and how many users were retained into second and third months.
We can also look at the rates for the top 10 volunteers.
| Referrer | Total Onboarded | First Month Active Rate | Second Month Active Rate | Third Month Active Rate |
|---|---|---|---|---|
| @soy-laloreto | 124 | 82.26% | 63.73% | 38.24% |
| @starstrings01 | 68 | 77.94% | 73.58% | 52.83% |
| @readthisplease | 48 | 2.08% | 100.00% | 100.00% |
| @dodovietnam | 42 | 88.10% | 78.38% | 32.43% |
| @dimascastillo90 | 39 | 61.54% | 66.67% | 25.00% |
| @laloretoyya | 37 | 86.49% | 62.50% | 50.00% |
| @feanorgu | 35 | 54.29% | 26.32% | 36.84% |
| @tpkidkai | 33 | 81.82% | 66.67% | 48.15% |
| @merit.ahama | 29 | 48.28% | 50.00% | 28.57% |
| @samostically | 23 | 60.87% | 71.43% | 50.00% |
Conclusion
The users onboarded by OCD volunteers have much higher performance on Hive as well as much higher retention rates than the community as a whole. The initiative is limited in scale, but the users onboarded have an outsized impact. The overall group of active onboarded authors has plateaued after the first few months, but this is in the broader context of Hive user activity declining overall.
It is likely that the approach of volunteer on-boarding somewhat mitigates the problems Hive has that lead to low user retention. Questions remains of scale and scalability.
- Current scale is still quite small. Can it be scaled up?
- Will it still be as effective at scale?
- Are there any costs to this method of onboarding that will emerge at scale?
Thank you for reading, and I hope that you found some worthwhile insights in the data presented here.
My statistics and analysis posts take many hours each to research, chart and write, so if you find them valuable and of interest to other Hivers, I appreciate your support in sharing, commenting, and/or upvoting my work. If you're interested in these kinds of stats posts, click the 'follow' button on my profile, or subscribe to the Hive Statistics Community which features daily Hive stats posts from @arcange as well as less regular posts from myself and others.
Amazing, thanks so much for putting in the time and effort for this!
I want to take the time to input some more info on how the onboarding program currently works, perks it gives to onboarders and onboardees, and some other things that may answer your questions, maybe it may even encourage some new onboarders to apply for a position reading this and looking at the stats above.
At this point in time it can be scaled up, we at @ocd do manual curation only and then another vetting/checking process before votes land on nominated posts. We strive to give people rewards based on multiple factors to make sure they "earn" them which usually means there's a lot of effort behind the rewards to give them another "hidden" value. Compared to say some who may get post rewards easily and consistently no matter their activity or post quality/effort. This means that with the current somewhat stagnant user activity some of our daily voting power goes towards @hbdstabilizer as we don't want to just "overreward" users for the sake of spending all our daily mana. This means there's definitely room for growth if we see onboarding through our program or in general a lot of new users joining the platform that we try to find and curate. Other than that, plenty of other stakeholders/curation projects also curate or trail our votes to reward users we've onboarded.
On top of it all, we also adjust our votes based on the price of Hive to keep rewards somewhat fair and reasonable. If the price of Hive goes up a lot, as we've seen a couple years ago, our vote strength adjusts lower and if it dips quite low we may end up spending more mana and less on hbdstabilizer for instance. I believe this is something that in the event of a big increase of new users joining would affect the price and allow us to scale the project to cover both onboarded users either from our program or in general (we also run the @lovesniper initiative) along with our general curation activities of niche communities.
Your 3rd question, something that may be considered a cost and some of the perks to being an onboarder are: 1. Onboarders get the beneficiary settings placed on accounts they've onboarded, it's usually 3% to the onboarder and 1% to the account credit giver and 1% to the service that created the account. Keep in mind though, that this isn't set in stone and onboarded accounts can remove the beneficiary settings at any time. It is a nice bonus for the onboarders though which the new accounts don't mind since they wouldn't have received any rewards without having been onboarded but eventually when they've learned the ropes or decide to themselves can remove them. Secondly, onboarders in the program are also encouraged to guide the new accounts and teach them the basics of hive, this works well with the beneficiary settings because if they don't post they may not get any beneficiary rewards and at the same time to keep track of all accounts they've onboarded and give them a warm welcome, they can somewhat guarantee curation and the 3% beneficiary reward by nominating their posts to us for curation. There are however times where the onboarded users don't post even after having been guided, or they decide to plagiarise or abuse in some shape or form even after being warned about it, etc, that's why we also do a "1 time" reward by compiling introduction posts of onboarded accounts on the @ocdb account and sending the post rewards to those who onboarded them. This is to not make the effort placed on the onboarders of guiding and teaching the new accounts go to waste fully if it turns out to not be an active/decent user they onboarded. So the costs would in general be quite low and come mostly from the post reward pool.
We've fine-tuned the onboarding program over time. You can imagine that being in such a position to be an onboarder not only gives you some nice perks but it can also be a position of power. For instance an onboarder could pretend to be onboarding new users but it turns out that's just his alt account he's nominating for curation to double-dip/get easy post rewards. This is something we try to check as often as possible with a lot of eyes having access to the nominations to avoid. This is why vetting the onboarders and them having "something to lose" in terms of reputation is important. There are of course instances where the onboarder isn't at fault for new users abusing/plagiarising, after a certain amount of these things happening we may just part ways with the onboarder without consequences to him directly but we reserve the right to treat it accordingly if we find the onboarder was the one abusing on purpose in one way or another (such as downvotes for instance) which brings us back to only accepting onboarders that have a reputation and "something to lose". Another perk to keep onboarders from attempting such activities for rewards is that we also allow them to nominate their own posts in moderation if they are underrewarded authors we would have curated either way if they weren't involved with us.
We've also set certain thresholds on onboarded users curation so people wouldn't be incentivized to join through an OCD onboarder rather than to just create a Hive account on their own and get involved. Once a new user is onboarded their posts can be nominated for max 3 months to us directly, the votes that land on them are quite similar to other posts we'd curate in the community incubation, etc, but over time the amount of nominations of a particular onboarded author start to dwindle from 3 nominations/week in their first month to 1/week in their 3rd month and then no longer. We expect that after 3 months users ought to have learned enough about Hive to have gotten injected in the community and receive curation either from our incubation activity (i.e. nominated by other curators (not their onboarders)) or by other curation projects/stakeholders.
We've also been meaning to offer delegation rewards to active and effective onboarders who'd like to focus more on this aspect on top of all the other perks mentioned above but without exact stats like these it's been hard to manually keep track of everything, maybe we'll start it up now again if you'd be interested in updating this data once a month or so. :)
For anyone interested in this, feel free to reach out to us in our Discord under the #onboarder-application channel
Lastly, @ocdb has been generating accounts for hiveonboard for a long time now which many projects and users rely on. While the dev has unfortunately not been very active, we're looking into creating our own solution to it soon with some added features and to avoid some of the issues it's been having lately. The 1% beneficiary that @ocdb may receive from post rewards of accounts it creates go back to the delegators delegating to the account in daily rewards (you can sometimes see some hbd going out to delegators along with added hive cause of this).
Thanks again and thanks to our current onboarders for their awesome efforts!
Ohhhh I see that's how it works
Thanks a lot for this profound and easily understandable explanation @acidyo now I know how it works around here. Most of the onboarded users only joined and become active for the first three months but after that, they lost interest. This is something we could try to find out ways how to let them stay longer and participate in the discussion.
Hello!
There is one important thing I have noticed with the people who come in, and that is that not everyone stays. I've looked for ways to retain them, one way or another, but it's impossible, so I make comparisons with the university. When I started at the university there were many of us in the first year, but as time went by, many left for multiple reasons, here I think the same thing happens: not everyone follows the rules, many people arrive wanting easy money.
On the other hand, I have a filter, if you don't write your introduction and send it to me before, you don't get your ticket for the account creation. This way I know that at least they are committed to create their first publication. Otherwise, my statistics would be completely different. But the truth is that I prefer quality authors to numerous authors.
Thanks for the stats.
You have done a great job, congrats!
Thank you ☀️
The user retention growth trend is exponentially decreasing one instead of an expected exponential increase. It just shows that the people onboarding do not make this platform a tool to engage with others or interact with its existing content.
This highlights a fundamental problem in this platform.
The times when you earned 1000$ for a post are not coming back. It was the privilege of early adopters, when price was high due to speculation but participation was low, so reward pool was split between few accounts. Now even if Hive reaches 100$, it will do so only as a result of higher adoption, which means bigger reward pool will be split between even bigger amount of users. Unless you are living in a country where 1$ a day makes or breaks your budget, you will likely never see substantial rewards for posting. The value of Hive as a blogging platform is not in that you can earn for your posts, but that you own your account and your content, and you cannot be censored.
There is zero correlation between earnings as an author and amount of HP you have. Even if you could point out some examples where there might be such connection, it is rather result of "social standing" (such account is likely to be an old one and other whales might know the user personally and might want to vote for their posts) and it is definitely nowhere near linear - whales are six-seven orders of magnitude bigger than plankton, the differences in post rewards span 4 orders at best (between .10 and hundreds of HBD).
The big accounts earn on curation as part of their investment - they would actually most likely prefer to just earn that passively as effective curation takes time, so they frequently end up redirecting votes to hbdstabilizer or delegate their power to services that do curation for them. Unless you employ the bot it is hard to squeeze your voting mana properly. The plankton falls under dust vote threshold (and rightfully so, otherwise it would incentivize Sybil attacks) therefore earning nothing on curation. It is actually "the middle class" that is in best position, since you earn the most on curation by digging out articles that are so good, yet underappreciated, that other people would not mind voting for even outside optimal voting window.
'Quality' content in so far as it gets attention and engagement and inspires participation. If we're not optimizing for attention, but instead optimizing for PhD articles worthy of being published in Nature then we are in a very archaic and frankly non-current social meta. Quality =/= results. In the words of smooth, you can put all the effort into creating quality, and for that, you get a gold star. What matters is results. In society, we don't reward people for high effort, we reward people for merit - whether or not the absolute effort required to achieve that is a lot or not is beside the point.
Automated voting is an issue, but like your point 3, so is removing the incentive for people to 'work' for their rewards, be it curator or author. A curator is literally anyone and everyone who has stake and is participating in the proof of brain mechanism. This is also a value contribution as crowd consensus is a big part of how we reach value determination of anything. By encouraging idle participation, we are actually incentivizing people to walk away from the underlying social elements offered and embedded onchain to keep the wheels turning.
I have mentioned elsewhere that I find rent-seeking incentives to be a gradual but growing headwind for advancing the overall value of Hive. Hive that is printed as a result of rewards is coupled with 'work done' through crowd assessment (curation) but also the time and effort of the creator to attract eyeballs (posting). (As mentioned above, the attracting attention part is some what the point of the attention economy, and quality as a metric should be determined by this outcome more than anything else). HBD on the other hand is the same capital input, with no additional work done and yet, nominally earning more than HP holders who accept market volatility tail risk, time-duration risk. Balancing it out without touching the APR would require increasing the time it takes to withdraw from savings. I'm talking a minimum of 13+ weeks since that is in line with the time duration risk of HP. Otherwise, the APR needs to adjust to at most, the amount HP holders can earn through legitimate activity which is currently ~ 12%.
Many others enter individually and when they notice that their content is not supported, they leave, others enter investing money and when they notice that nobody turns to see them, they leave, they need a community committed to support these users who enter by 3rd persons and need help to know more about the ecosystem, nobody deigns to leave them a message inviting them to a community or inviting them to create content in the appropriate communities, they enter lost and leave lost.
This type of post can be quite time-consuming to create, and you've done an excellent job of providing us with information that most of us would have spent hours learning or researching. Onboarding is an important activity that needs to be examined and improved upon. Personally, I believe that those who are currently working on onboarding are doing an excellent job, and going forward, everyone should be more involved, not just those who have volunteered. Besides OCD, I think that all other community leaders should also help promote promote Hive more through various activities. This is something that Hiveghana is working hard on every day. We are steadily increasing our numbers and aim to grow even more significantly before the end of the year. While focusing on bringing in new users, it is equally crucial to consider how we can retain them. In my opinion, it is essential to concentrate on strategies to achieve user retention as well.
wow amazing work m8! much value and insight!
Muito obrigado!
Heya Demo definitely your charts are something that I do look forward to reading. Being an onboarder's a challenge as some have different motivations after they joined Hive. we try to at the very least have them connected to the community and hope that they will stay but outside Hive life is still strong.
You already know I love your work and it inspires me to do my own tracker as well within the PH community. Keep this thing up as it is truly helpful!
This is my first time knowing what the OCD does and how it works
Thanks for the explanation
I don't understand why that should be? It seems like we have more curators and incentives than ever...
This post has been manually curated by the VYB curation project
This comment was made by a hive-archeology bot running under the control of @pibara
The goal of this comment is to act as reward proxy for up-voting valuable timeless content on HIVE for what the one-week upvote window has closed.
The bot script is currently pre-beta.
Sorry, I don't understand. What am I supposed to do?
You aren't supposed to do anything. Maybe this (automated) post should have other wording to make that clear. It's this bot creating a comment so I can up-vote posts and comments outside of the 7 day window. If you have any suggestions to make the post less confusing, I'm all ears.
It was after 2021 Hive bull run. A decline afterwards is fairly inevitable. That said, we have chronically low new user retention at all times which has never been resolved.
I remember that everyone seemed to go quiet when Hive was truly inexpensive, but now sat wise we're right back where we were during the bull run? Even those who don't truly believe in Hive have an excellent incentive to post...
As for me, I never went inactive in the comments. And never stopped posting except for when I had to for offline reasons.
!PIZZA
Thanks for sharing 🙏
Hello @demotruk
Thanks for the amazing stats again, reading your posts and the comments gave me an insight of how OCD works, given that i was onboarded by OCD it was nice to know their mode of operations.
This comment was made by a hive-archeology bot running under the control of @pibara
The goal of this comment is to act as reward proxy for up-voting valuable timeless content on HIVE for what the one-week upvote window has closed.
The bot script is currently pre-beta.
Nice, but if I now upvote your comment it's basically a self vote 😅
PS! These stats are a bit of a puzzle to me, wonder what happened there.
It's skewed because they had a very low first month rate. The later months are compared to the first, in order to make it more directly comparable to the rest of the stats I have used. This can be skewed when users take more than a month to become active after joining, such as happened for this account's onboarded users as well as @steemmonsters in the post yesterday.
I cap it at 100% rather than let the outlier push up the vertical axis in any charts.
Ahh okay makes sense, thought it meant 100% of onboarded users became active the 2nd month at first which doesn't make sense to begin with since that number represents total accounts they onboarded til today.
goood!
Congratulations @demotruk! You have completed the following achievement on the Hive blockchain And have been rewarded with New badge(s)
Your next target is to reach 65000 upvotes.
You can view your badges on your board and compare yourself to others in the Ranking
If you no longer want to receive notifications, reply to this comment with the word
STOPCheck out our last posts:
Support the HiveBuzz project. Vote for our proposal!
$PIZZA slices delivered:
@wrestlingdesires(4/15) tipped @demotruk
very entertaining :)