"Community over code."

I overheard someone say this in a meeting recently; the speaker was arguing that for teams building a consumer web product, especially one that involves user generated content, nothing is more critical than attracting the right users. He was exaggerating to make a point but the spirit of the statement isn't far from the truth.

At the expense of pithiness, I would adapt the phrase to say that code, and by extension the product, is only as good as the kind of community interaction it nurtures. Yes, it's important to target a receptive seed audience in the beginning. But scaling requires that you drive the right type of behavior among the broader market.

Step 1: Aligning user motivation, behavior, and long-term value

Growing a strong community is a two step process with distinct challenges at each phase. The first step is identifying your early adopters or group of "must have" users. As a consumer startup it's tempting to view your target market as anyone with an internet connection, and while this may be theoretically true, it's not very actionable. Focusing initially on one segment is no guarantee of success, but there are plenty of supporting examples: Facebook began as a college only network; Yelp spent its first year and a half in the Bay area; Pinterest gained early traction among design bloggers; Quora got its start with tech-minded folks in the Valley, etc. [1]

The benefit of this approach is that it allows you to iterate your value prop with an audience who's more forgiving of missteps and eager to give feedback—so don’t be shy about asking for it. Much of the meaningful data you collect at this stage will be qualitative. Your goal is to understand motivations driving user behavior and how well they align with the value you want the community to create in the long-run.

Look for two kinds of misalignments in particular: The first is a gap between the source of uber value and actions users are taking, e.g. if the former is connecting people based on shared interests and areas of expertise, but most users end up following friends from Facebook, consider changing your onboarding flow.

A second gap can exist between the source of long-term value and people's motivations, e.g. if users are taking the intended action of writing reviews but their behavior depends on monetary incentives, it likely won't scale.

This is also the time to establish product signposts that signal what type of community norms and expectations are in play. Quibb, a network for sharing and discussing industry news, gives a friendly prompt to "Please capitalize when appropriate" if you begin typing a comment with a lowercase letter.

As another example, Yelp doesn't allow users to post reviews from the mobile app. You can jot down notes about your waiter's endearing accent so you don't forget, but you'll have to wait until you're in front of a computer before you can hit publish. It'll be interesting to see if this policy changes given where things are going, but it's not hard to understand the reasoning. You're about as likely to peck out a thorough, nuanced review on your iPhone keyboard as you are to find good sushi at a Brazilian steakhouse.

A lot of the iteration at this stage involves stress testing the implicit bets in your product vision. Take Medium as an example: One of the team's principles is to focus attention on high quality content irrespective of an author's popularity, the hypothesis being that people will appreciate compelling storytelling regardless of its origin. To that end, there are no public metrics like follower counts or number of shares; you can tweet or recommend a particular post but the latter is visible to just the author on an anonymous basis. Early versions of the product didn't even identify the author until the end of the post. The design has since changed to display a brief bio in the top left-hand column as the lack of context probably gave readers too little idea of what to expect before committing.

A final note about this initial phase of building community—don't forget to give back. Sandi MacPherson at Quibb has done a great job of being transparent with her thought process behind evolving the product.

Step 2: Crossing the chasm 

Once you've figured out how kickstart a sustainable contribution loop among early adopters, it's time to focus on the harder task of scaling. The biggest challenge here is that new members tend to differ radically from your seed audience both in terms of attitude and goals. They care less about the product's potential or long-term vision, and they're less willing to invest the time to figure things out. You need to make it obvious which types of behavior will lead to the best experience and deliver your lightbulb moment as quickly as possible. And in keeping with the 90-9-1 rule of internet participation, the majority of these users are here to passively consume versus actively contribute, so you should make the consumption experience as frictionless possible.

If you can offer a lightweight means of contribution as people are passing through, all the better (the hope being that you can gradually increase their level of engagement). Instead of implementing a standard upvote system to rank reviews, Yelp presents 3 dimensions—useful, funny, cool—to give more options for one-click interaction. Quora's "Thank" feature is in the same vein for users who want to express appreciation in a non-public way.

The best part of this community orientation is that it focuses design decisions around what results in the most value to the broader ecosystem instead of subjective opinion. The key is to start with a strong vision of how your community creates long-term value and then design for the behaviors that make it a reality.

[1] Instagram and Snapchat may have gone from 0 to millions of users almost overnight, but they’re also a slightly different type of product as mobile-first apps with significantly lower barriers for content creation.