A prospective client recently asked: “We know our audience better than you, we can do the research ourselves, why would we need your help?”
It’s a really good (and fair) question. We will never know an audience as well as you. You’ve been working with them for years.
But research isn’t about what you know, it’s about what you don’t know. Better yet, it’s about distinguishing what matters from what doesn’t.
The curse of knowledge is very real here and people tend to repeat the same mistakes. These mistakes usually include (in order):
- Using research to support rather than refute existing ideas. By far the biggest problem is gathering data to support, rather than refute, existing ideas. Science is the process of coming up with a hypothesis and trying to disprove it. I’d estimate around 75% of people sabotage their own audience research trying to support whatever idea they had originally for the community.
- Averaging the answers. This happens when you build a community around the average of responses instead of around a topic a small niche of members are intensely passionate about. This leads to [communities about topic] which struggle to gain traction instead of “Communities for data scientists who want to take data visualization to the next level”. You’re not looking for a community everyone finds acceptable, you’re looking for the thing that lights a spark under a small segment of your audience.
- Homogenizing the community. This is the act of building a single member profile. This is silly. You have newcomers, experts, part-timers, hardcore advocates, different geographies, and personalities. The research phase should identify the different groups (we usually have 2 to 5) and the core passions, fears, and motivations of each.
- Treating interviews like surveys. If you’re asking all members the same questions without diving deeper into responses, clarifying answers, finding out what they get really excited about and picking up on non-verbal cues, you’re basically running a very inefficient survey. Interviews are about getting to the deep emotional reasons to why people do what they do.
- Bad data analysis. We correlate the time a community manager spends on each activity with the results they’ve achieved from it. Usually, we can cut 75% of activities without any noticeable impact on the community. They can then spend this time working on big wins. Most people collect a lot of data but miss the key correlations/analysis that turns it into actionable insights.
- Not developing and testing options. The end result of this isn’t a big document, but validated steps to pursue. The analysis reveals several options to test, but the process of testing them takes time to get to the right answer. The end result is ideas which have been tested and are ready to go.
- Collect data you don’t need. This happens often in surveys. Organizations try to drop questions in which are interesting but have no obvious influence on how we develop the community.
- Breaking the law. Laws vary by country (especially GDPR in Europe). More than a few organizations have accidentally broken the law in the audience research phase.
You can certainly go through this process without outside help. We’re certainly not cheap and I doubt others doing this work are either. Our rates for this process typically range in the $20k to $35k region for a project that takes around 6 to 8 weeks.
However, this sort of research is the difference between a community which explodes to life and one which struggles to gain traction for years. It can save you hundreds of thousands of dollars, and many years, trying to get it right. It helps you ensure you’re building the best community possible. It’s a community which is almost guaranteed to succeed.
If you do go it alone, treat it internally like a $20k to $35k project. Smart companies know the success of the community doesn’t hinge upon the technology, it depends upon deeply understanding the needs of members. So give it the gravitas it deserves. This process takes time and requires more than just a lone community manager trying to get it done in her non-existent spare time.
Make sure the end result of the research is a powerful community concept which has been tested with the audience and shown to really resonate/drive participation. Don’t launch the community until you’ve validated a powerful concept and proved you can sustain activity on a small scale.
