Not long ago, it seemed like there weren’t any reasonable measures for the health and effectiveness of communities; now it seems like there are too many! We often see eSupport and community leaders trying to grapple with page after page of bar charts, line graphs, and tables, unsure what to pay attention to, and unsure what to report up the management chain.
When we look at communities, we simplify matters by cutting things down to six measures—three activities, and three outcomes. Activities, and trends in activities, will tell us if the community is healthy. The outcomes let us know if they’re effective—or, more precisely, part of an effective eSupport strategy.
Activity measures tell us if the things we’re planning on are happening. Good activity measures don’t guarantee success, but poor measures are a good indicator that something’s wrong. (A cocktail party with 30 guests isn’t necessarily a good party, but one with only three guests is likely to be a bust.) Community activity measures tell us if people are participating in the conversations. If enough people are participating, that suggests they’re finding it valuable.
- Page views. We’re still not sure whether a tree falling in a forest makes any noise, but it’s a sure bet that a post that isn’t seen isn’t doing any good. More page views equals more opportunities for value creation. This is important to trend over time. It’s also interesting to see where the page views are coming from—your community site? Self-service search? Google and other Internet search engines? This can help you refine your marketing strategy, and perhaps help you fend off those colleagues who want to lock your community behind a paywall.
- Active contributors. How many people are not only registered and looking, but actively participating in a discussion (either starting a thread or following up) within the last thirty days? In the standard 90/9/1 model of community engagement, this measures the nominal 9%—although in the real world, the number is often significantly less than 9%. Trends are as important as the actual numbers for this measure.
- Posts per day by forum. This is what communities researcher Dr. Michael Wu refers to as “liveliness”—is there a good buzz? Are we at critical mass? For a specific forum, it takes at least five posts a day to be lively; those with fewer might best be merged with other forums until the topic attains sufficient momentum.
Outcomes: Gauging Community Effectiveness
Outcomes are the business results we are seeking from our community initiatives…and in fact, from all our eSupport initiatives. If the activities tell us the “what,” outcomes tell us the “so what.”
A challenge with outcome measures is that no one activity can “take credit” for the outcome. This is frustrating when trying to justify investment in a specific program, but it makes sense: wouldn’t it be odd if a self-service program and a communities program were trying to accomplish different outcomes? Shared goals encourage teamwork, and are a fact of life in the enterprise—no one group gets credit for company profitability, either (not that Sales won’t try.)
If teasing the value created by communities apart from other efforts becomes a paramount consideration, the activity measures at least provide a rough order-of-magnitude starting point. If there are 100 times as many community interactions as there are live chats, it’s reasonable to argue that the community program is a more significant driver of Net Promoter Score, while if there are ten times as many page views in the knowledgebase than in communities, the knowledgebase might be making a bigger impact.
- Deflection. The most easily quantified financial benefit from communities is contact (or case) deflection. Deflection in communities is measured just like deflection in self-service: it’s the percentage of people who are successful in accomplishing their goal, times the percentage entitled and intending to open a case, times the number of times people use the communities to resolve an issue. That is, Deflection = Success Rate x Escalation Rate x Sessions. While calculating success and escalation rate are worthy of a paper all their own, the quick answer is that you should call your community users and ask them about their last experience—were they successful? Did they / would they have escalated? Verb. sap.: escalation rates are lower—often far lower—than people assume.Note that communities deflect contacts by having a customer ask a question and get a helpful response. But it’s far more common that a third party—a “lurker”—comes along later, sees the exchange, and uses it to solve his or her problem. Accordingly, if community posts are returned by self-service search or by Google, it makes sense to calculate a blended deflection number across communities and self-service.
- Satisfied demand for support. Support is in the business of creating value for customers, not just closing and deflecting cases. So every successful interaction in the communities provides value. It’s hard to assign a dollar value to satisfying a customer’s need, but that doesn’t make it less real. Satisfied Demand = Success Rate x Sessions. As with deflections, this may be a combined number across self-service and communities.
- Loyalty. However you measure loyalty—typically Net Promoter Score, Renewal Rate, or Repurchase Rate—communities should affect this positively. Look for changes in loyalty especially when community activity measures have changed significantly within a given time period.
Six measures, two slides…and a very telling picture of just how your communities are going.
(Thanks to the Association of Support Professionals, who kindly let me repurpose this piece from a contribution I made to their excellent report Successful Support Communities.)
ps – to our KCS friends who aren’t out on the coasts, sorry if it seems like we’ve been ignoring you! DB Kay has a KCS Workshop coming up October 5-7, Chicagoland-style. We hope you can join us there…and please let your colleagues know.