Constructing Marketing’s Key Performance Indicators (KPIs)

Been listening in on a conversation between Gary Angel and Eric Peterson regarding web analytics KPIs. Eric says:

Key performance indicators should be included or excluded from a hierarchical reporting strategy…based on the likelihood that the indicator will spur some type of action in the organization when the indicator unexpectedly changes…[although] the action the organization would take, when unexpected change occurs, is never precise.”

Gary’s take is more context-oriented:

A report becomes actionable by using KPI’s to provide the business context within which an action can be identified or deemed worth trying. The more relevant context a report provides, the more likely it is to be actionable. KPI’s are the context builders that make up our view of what’s important and what isn’t.”

Although their focus is on Web analytics (I think), their points are relevant to Marketing’s broader quest to define the right set of KPIs. My take: Constructing the “right” set of KPIs is not an either/or decision between actionability and context. CMOs should look at their KPIs as a portfolio of measures, each of which should meet at least one of the following criteria:

  • Explanatory ability. Perhaps closest to Gary’s description of context, does the KPI help explain why what happened happened? (sorry, you may have to read that twice).
  • Predictive ability. Does the KPI enhance the firm’s ability to predict what will happen in the future?
  • Behavior change. Does the act of measuring the KPI incent people to act or behave in a particular way (that management believes is desirable)?

When evaluating a set of potential KPIs, marketing execs need to put each metric up against these criteria and ask themselves (at least) two questions:

1) Does this metric help us explain, predict, or change behavior? If not, then that metric may not be a good candidate for the KPI list.

2) Is the set of metrics balanced between the criteria? If all of the proposed KPIs are explanatory, then it’s likely that reports will be “rear-mirror” focused, leaving senior execs frustrated that Marketing isn’t looking ahead.

What about metrics that simply describe what happened? I’m not saying that these shouldn’t be measured. But they don’t clear the bar to qualify as a Key Performance Indicator.

This is important because too many marketing dashboards and scorecards are nothing but laundry list of metrics that senior execs often get from finance or within their own LOBs. To be effective, Marketing’s reports have to go beyond the standard metrics and add value to execs understanding of the business (i.e., explain, predict, or change behavior).

Hat tip: Jim Novo
Technorati tags: , , ,

3 thoughts on “Constructing Marketing’s Key Performance Indicators (KPIs)

  1. I looked at Gary’s, Eric’s and your own post. Mostly they make my head hurt. Can’t just about any metric explain or predict something? Surely the real question is how to prioritize. My own approach is to prioritize based on financial impact (specifically, change in aggregate lifetime value), as I’ve been writing about in gruesome detail this week. But I’m sure there are other ways.

  2. ok… but if aggregate lifetime value isn’t utilized at a particular firm, would you still advocate for its inclusion in the list of KPIs? Financial impact may or may not work, because not all KPIs will necessarily have a direct financial impact.

    As I think more about this (don’t ask WHY I would be thinking more about this), I think you’re on the right track thinking about prioritization.

    After establishing a candidate list of KPIs (hopefully that meet the criteria I’ve laid out), the marketing management team still needs to address whether or not they can measure the metric, and if not, what would it cost to measure it?

    In the end, here’s what my guidance to a CMO would be: Pick the set of metrics that will best help marketing to INFLUENCE senior management. Influence investment decisions, influence business direction, and influence management’s perception of the impact that marketing is having on business results.

    [Now go get an aspirin for that headache]

  3. Ron, just to clarify my position: KPIs should be components of aggregate LTV, not aggregate LTV itself. Any change in a KPI can then be associated with a change in aggregate LTV, and reports can prioritize the changes by ranking on this value. My underlying point is there are lots and lots of potential KPIs, so you need some way to figure out which ones are important right now. Having KPIs that are all linked together in a LTV formula also provides the “context” that Gary Angel was talking about, if I understand him correctly.

    Yes my approach limits you to KPIs with direct financial impact. But I’d argue that everything important can somehow be linked to financial impact, so this isn’t as restrictive as it sounds. Plus, undertanding those links is itself a very worthwhile activity.

    But enough about me. I’m intrigued (in a good way) at your idea of “influence” on business behavior as a key criteria for selecting KPIs. Influential measures could definitely be important, and yet wouldn’t necessarily have anything to do with financial impact. The example that leaps to mind is projected management bonuses, which certainly influence behavior but usually involve amounts that are immaterial in terms of the overall business. I guess the challenge becomes, would you want metrics that influence behavior to be metrics that are not also related to value? That could lead to conflict between personal interests and business interests, which is surely not your intention.

Comments are closed.