Next level community strategy
Whenever I hear the phrase low-hanging fruit, I shudder involuntarily. Same with “quick wins”. Far too often an idea presented as a “quick win” is neither quick nor a winning strategy. Especially when it comes to communities, winning strategies ignore low-hanging fruit.
Learning from old-school AI—chess computers
When I was in high school I got Genesis Chess Challenger for Christmas. It’s a nice chess board built on top of a sophisticated chess computer. You press a button hidden under each square to select a piece and press the button on the destination to show the computer your move. After thinking about it, the computer signals its move with LEDs highlighting the row and column. It also had 16 difficulty levels. According to the manual:
| Level | Approx. average time per move |
|---|---|
| 1 | 1 sec |
| 2 | 3 sec |
| 3 | 5 sec |
| . . . | . . . |
| 16 | Problem Level |
The 16th level is called “Problem Level” because it’s designed for endgame problems with only a few pieces left. The computer will keep working on the problem until it either finds a solution or the batteries die.
The other levels correspond to the number of moves the computer will look ahead before making a decision. Level 1 means it only looks one move ahead—its own next move. Level 2 means the computer considers each of its moves and each of the player’s possible responses. I got pretty good at beating these levels just because I was able to think a move or two ahead of the computer.
By level 3, the computer puts up a pretty good challenge because it never makes mistakes and I struggled to plan out my moves further than the computer. It only gets worse for level 4 and on. The computer perfectly plans so far ahead that humans can’t keep up.
But there’s a catch: the further they look ahead, the more time it takes computers to calculate each move. And since chess tournaments have time limits (3 minutes per move), my Genesis Chess Challenger would lose on time to a grandmaster. Winning chess computers require much more CPU power.
Loser Level 1
Mature communities inevitably get labeled as “toxic”. New users feel unwelcome and legacy users feel worn out seeing the same newbish behavior day in and day out. Community managers want more new users, so they write a post on the site asking the community to be nicer to the newbs. It never works. Sure, some people agree, but this strategy alienates many of the community’s most valuable contributors.
It’s not hard to figure out why it fails: nobody likes to be called “toxic”. Even when the post uses nicer words, nobody is fooled. Calling out unwelcoming behavior is an obvious solution, but it’s a Level 1 trap. It’s a strategy to lose existing members without helping new members. Thinking about how the community will respond first is a Level 2 strategy and that’s much more likely to work.
Instead of a “quick win”, take a little more time to understand the problem. Do some research. Ask some questions. Maybe you’ll find out that long-term members are tired of seeing the same questions asked over and over again. That might lead to better solutions such as a banner asking people to search the site or surfacing potential duplicates when people start to ask.
Aim for Level 3
Trying to fix the root problems is more likely to work than the naive solution. But odds are very good your first attempt won’t work. Far too often community managers define success by completed projects rather than improved outcomes. Instead of iterating, people tend to want to declare victory when the first solution is done.
I can easily beat a chess computer set to Level 1 because I’m ready with a response to any move the computer makes. Level 3 community management means having a plan if an initiative succeeds or fails. Can you fix flaws and try again? Would a different approach work better? Maybe you still need more research to understand the problem. Even ideas that succeed can be improved.
Beware of Level 16
If Level 1 doesn’t work and Level 3 is better, why not go to Level 16? This mistake is not as common as “quick wins”, but can be much more costly. After spending months preparing a plan, there’s no energy left for fixing the problems if it comes up short in execution. There’s a natural tendency to blame the community for not getting on board with the plan. And people get impatient for progress so they try gathering that low-hanging fruit for a quick win. That gives you the worst of Level 1 and Level 16.
Communities are complicated and unpredictable. That’s why they are so interesting! Community managers who hide in their ivory tower and only come out a few times a year to announce a major change aren’t likely to be successful. But if you can get into a cadence of iteration and planning ahead a few moves, your community will start to see where you are going. If you keep it up long enough, the community might even make your goal their goal too.
That’s a next level community strategy.
Comments