We need to proactively challenge community dogma with hard data.
Much of what we believe to be true, is a myth.
I used to think the best way to hire a great community manager was to find a great community and hire the manager. I was wrong. The best way to hire a great community manager is to find someone that's incredibly passionate about the topic and train them up.
I used to think that gamification was terrific for online communities, I was wrong. Studies show that gamification has minimal impact upon communities and could be negative in the long-term.
I used to think that Facebook was a great platform for building communities, I was wrong. It's now impossible to reach most of your audience on Facebook, they shift the platform often, and you can't customize important newcomer journeys.
I used to think it was possible and easy to convert many lurkers into regular members, I was wrong. The data shows that aside from a few anecdotal examples, this won't happen. People become lurkers due to a lesser interest in the topic or lack of initial engagement. The best way to tackle lurkers is to engage them in active contributions from the moment they join the community.
I used to think it wasn't possible (or necessary) to calculate the ROI of a community, I was wrong. It's essential. If you can't prove your community helps the organization (to a financial value), it should be scrapped.
I used to think personal welcomes were a powerful tool for converting newcomers into regulars. I was wrong. Personal welcomes don't scale, they're usually done badly, and there are far better optimization tweaks out there.
We're better at community building because we continually test our assumptions. We go out there looking for data/studies that supports or refutes our assumptions. We change our thinking based upon the evidence.
There are a lot of sacred cows we need to challenge. How we use our time, how/why we keep people in a community happy, who we need to focus our time on, what types of platforms work, the value of community guidelines, the benefits to organizations, group influence etc...All of these have incredible scope for improvement.
But first we need to accept that much of what we think is right about communities is false. If we can't accept that, nor feel comfortable testing our assumptions and changing our minds, we're not going to advance.