So I’m not personally a fan of our new micropayment experiment, but I also feel that experimentation is far and away the best way to figure out what works. I also think that it’s essential that Hubski figure out a a way to pay for its expenses because right now, we’re enjoying the hospitality of a handful of dedicated volunteers serving us all out of the kindness of their hearts and as great as they all are, we can’t expect them to continue forever.
I’ve been knee-deep in the business of aggregators for a while now and am a reasonably clever person, so I know that this is harder than it looks. But I’ve been knee-deep in the business of aggregators for a while know and am a reasonably clever person, and I honestly think Hubski has a better chance than anybody else. To whit:
- Hubski encourages ownership. Your posts are your posts, your relationships are your relationships, your conversations are your conversations and there’s no intermediary between what you share and what everyone else sees.
- Hubski encourages original content. The makeup of the site favors the creation of things as opposed to the theft of things. Because discussions move slowly, “karma” is deprecated and context is important, the repost hit’n’run approach common on Reddit and Digg is much less favored here.
- Hubski favors pseudonymity over anonymity. A member on Hubski is a part of a social fabric whose absence alters the makeup of the site for all who interact with them. Reputation matters more on Hubski. Relationships matter (at all) on Hubski.
In short, members have a stake in Hubski. And while I have no illusions that this stake will make anybody rich, I’m confident that it can be made to pay the bills.
What follows are five ideas, none of which are proven, all of which may be stupid, all of which can be tested… after a fashion. They aren’t mutually exclusive and they aren’t equally lucrative but they’ve all got some precedent out in the world so maybe we could try them out?
__________________________________________________________
(1) THE SOMETHING AWFUL MODEL. I did some back-of-the-envelope calcs on what Reddit “costs.” I might share them later. Long story short, Reddit is profitable at about $2 per username per year. Something Awful has always used a signup fee to cover costs, pay salaries and disincentivize trolling; act like an asshole and you have to pay your Ten Bux again.
- The idea: determine a sustainable “cost per account” for Hubski and charge members that much as a one-time fee. It needn’t be immediate - I’d probably let everyone operate with full privileges for a month, for example, and then hit them with the fee after 30 days.
- The drawback: people will likely feel shaken down after 30 days unless the fee is nominal. Textra recently went from free to paid and the internet was alight with indignant rage… until everyone determined that Textra wanted 99 whopping cents.
- The test: Well, this one is more of a Survey Monkey thing than anything else. I suggest doing three or four quarterly surveys of the user base to see what Hubski is worth to them. We needn’t know publicly what Hubski’s “member cost” is but it would be useful to correlate member age with member price sensitivity. Ask everyone how long they’ve been on Hubski and what one-time price they’d be willing to pay to keep using it, and ask it more than once. Then consider that survey behavior doesn’t perfectly correlate with actual behavior but hey - it’s data points.
____________________________________________________________
(2) THE NPR MODEL mk has said he doesn’t want to “beg like PBS” but the fact of the matter is, patronage models are time-tested. As a “thoughtful web” there are likely other sites and services that would be interested in sponsoring Hubski for the exposure to an erudite and drama-free audience. Hubski isn’t 100% pure thought leaders, but as Reddit hemorrhages user base the smart ones aren’t going to Voat. I’ve done pledge drives for public radio in the past and when you have a concrete goal and a concrete way to fulfill it, people make the connection between value received and money required.
- The Idea: show Hubski’s “actual costs” (could be a dummy figure) over time and incite people into paying for it.
- The drawback:
- The test: You know, the fact that my upper left hub wheel and “feed” go to the exact same place. And we’ve got the “you’ve got satoshis!” code built out already; can we turn that wheel into a “health” indicator? And when I click on it, can I see, like a projection of funding going out 12 months in a happy graph? ‘cuz I want to be able to use my funny money to fund shortfalls. And you know what? I want to be able to donate in other people’s names. And I want the option (not the requirement) of showing that I helped keep the place running. Let’s try it with fake money and see what happens.
______________________________________________________________
(3) THE VALUE ADDED MODEL So at $2 per account per year, it becomes pretty obvious that one month of Reddit Gold pays for two accounts. If one person in 15 buys a year of Reddit Gold, Reddit is profitable. Unfortunately what you get with Reddit Gold is bullshit. However, it demonstrates the Pareto Principle pretty well: you don’t need everyone on board, you just need the ones that are paying.
- The Idea: entice Hubski’s more active members to underwrite the site for everyone else through perks.
- The drawback: Coming up with perks good enough to entice people to pay for something they’re getting for free.
- The test: I noticed that my “interactions” graphs are gone. I suspect they were a casualty of site influx. Put them back… for a fee. Come up with some schemes that aren’t f’n hideous and let me use them… for a fee. Hey - let me inflict my choices on my posts… for a fee. See what that sort of two-level culture does when the distinction is artificial. Turn 20% of the current member base into “patrons” at random and see what happens.
_____________________________________________________________
(4) THE TARGETED ADVERTISING MODEL Something that has always bugged me about Reddit’s self-serve is that Reddit has your complete interest profile. They know what you upvoted, they know what you down voted. They know what subreddits you’re subscribed to, they know which ones you have blocked… yet when it comes to advertising, you get straight, conventional CPM thinking.
- The Idea: determine how successfully Hubski can target ads, while also giving members control over their ad experience.
- The drawback: Ads.
- The test: forwardslash was talking about messing about with Markov chains and the like… I’d like to see Hubski extract my shares and my follows, first and second degree, to get a weighted chain of “interests” based on my behavior on Hubski. Then I want Hubski to use that data to create a query… and maybe give me something off Google Image search or the like. Better yet, show me what it’s extracted, allow me to edit it, and get several searches, then let me pick one out of three that allows me to select my own “ad experience.” If you can make this work, you’ve got an ad platform that blows the doors off anything else available… well, more on that in a minute.
_____________________________________________________________
(5) THE SELF-CURATED ADVERTISING MODEL Building on the above, targeted advertising is better than scattershot advertising… but it’s still advertising. And it’s something I’m not in control of. If you give me three choices of the ad experience I get, I’m engaged. But if you give me three choices of the ad experience people see on my posts, I’m even more engaged. Better yet, rather than serving me ads give me the option to buy my own ad experience on a monthly, quarterly or yearly basis. Anticipating that you’ll make 50 cents off my CPM this year? Let me buy out of all ads for a dollar. Let me buy my friends’ ads. Let me buy pretty pictures instead. Hell, let me sell a marketplace of “things that aren’t ads” to put into your “ad network.”
- The Idea: take the above one step further and increase engagement.
- The drawback: Ads.
- The test: continue the fun and see what comes of it.
_____________________________________________________________
I switched from days to overnights yesterday. I’ve had two or three one-hour naps but I’m definitely sleep-deprived. As such, I’m sure there are stupid ideas here and I invite anyone and everyone to shoot them down. I also invite anyone to build on them or suggest something entirely new. We all benefit if we can make Hubski self-sufficient in a way that doesn’t suck for everyone, and there’s nobody better at figuring that out than Hubski members.