I'm a little out of practice, having been gone for the past 2 months on maternity leave. I haven't been blogging, and only infrequently tweeting. But that doesn't mean I haven't been thinking about social business. As the folk wisdom goes, sometimes taking a step back gives you a fresh POV.
One of the best perspectives I got came from the worst part of my maternity leave: a 3 day stint at the Children's Hospital, where they had the most impressive analytic dashboards.
In the pediatric ICU there are monitors everywhere, with all the right real-time analytics. At first, I was in awe of the gold 'standardness' of it all - they know exactly what to look at to monitor the health of their patients. One dip or rise and the alarms sound. If only we knew the equivalent of heart rate, blood pressure, and body temperature in social business...
But what I learned is, even when you've got the right metrics, it's still complicated. So as you continue to chase the holy grail of business metrics, here are some things to think about.
- Don't set thresholds too aggressively. The monitor in our room buzzed all the time. Initially, I panicked each time. Inevitably, a nurse would come in and turn it off. No exam necessary. No big deal. They had set the thresholds for various measures as early warning signals- the beeping wasn't really cause for concern. Eventually I learned to ignore it. When your metrics aren't calibrated to levels that require action, it leads to non-response.
- Connect your sensors to the right surface (think: people, department, products). My son's blood pressure gauge was on his ankle. With any kicking, the monitor was overcome with noise. He kicked all the time. Again, I endured the meaningless beeping and worse, an eye-opening flatline would appear on the monitor. The signal was great when it worked, but it was hard to read patterns over time with so much missing data. If you want your signal to come in loud and clear, be sure you're accessing a reliable location.
- Single measures mean little in isolation. Each of the measures monitored was a very strong indicator, but never did movement in one call for action. Action was the result of a configuration of activity across measures, over time, and as a result of consensus across nurses and doctors. Here was the biggest lesson: 'metrics that matter' actually means more than identifying the right metrics, the holy grail-- it's about finding the ones that act in concert to yield something meaningful, having the right people monitoring them, and instituting the right processes for crisis response.
Now, this is NOT a critique of the medical practice. This particular hospital had it down- they knew exactly how to 'listen' to their dashboards. Having internalized tactics like the above, they knew exactly when to intervene.
This is, instead, a lesson about valuing what it is that you're currently measuring. The right metrics will come; for now, make sure you're making the most of your current dashboards.
More to come!