To write The Digital Metrics Field Guide I had to choose which metrics to include from an ever-expanding supply. Sometimes I wondered if there was a metrics nursery somewhere, like the star nursery in 30 Doradus.
So Many Metrics. Why?
When working on the Facebook metrics, I struggled with this question: Why were there so many measures concerned with engagement, interaction and sharing? Clicks. Social Clicks. Consumptions. Likes. Friends. Fans. Etc. Were they all needed? Why?
This stumped me for close to a month until the curtain drew back: The metrics Facebook reported appeared to reflect their assumptions about how social worked on their network, the features and functions they built into their platform, and their business model for selling advertising. I’m using Facebook as an example: the same was true for other social networks, sites, and engines.
Metrics Measure a Platform’s Business Model: The Endometrics Problem
I checked this notion out with Gilles Santini, a mathematician who, among his degrees, has a Masters in the philosophy of statistics and who created many of the industry-standard media measurement models. “Stephen,” he said, “these are ‘endometrics,’ Measures that come from within the system being measured. They measure the performance of a system in its own terms.” Bob Woodard, a former head of global insights for Campbell’s and erstwhile ARF colleague put it more bluntly: “Vendors measure what they want to sell you.” Aha!
A social network, site or engine that talks up its philosophy of “how advertising works on their platform” and is then joined by a chorus of agencies, consultants, or gurus, leads brands to work hard to optimize one or more of the metrics available from that platform to improve their chances for communications success on that platform. It made “perfect sense” to design Facebook campaigns to get as many clicks, likes, or fans as possible.
Business Models Change as Business Changes
Back in 2012 Facebook ran tests with Datalogix matching a Facebooker’s ad exposure on Facebook with their purchase data for 50 advertised brands. They found that 99% of the documented sales were from people who saw ads but did not interact with them. Brand advertising, Facebook VP Brad Smallwood announced, worked just like mass media advertising. Clicks, Smallwood noted, seemed applicable in some cases, such as in direct response campaigns, but it was challenging to understand their offline sales impacts .
Facebook built upon this finding over time and shifted its philosophy, business model, and marketing towards targeting, impressions, reach maximization, and frequency optimization, and away from the old engagement model. Just last month Facebook rolled out LIFT, a tool for measuring online or offline conversions that reinforces their current model.
Metrics That Matter Change When Business Models Change
The point is that philosophies and business models furnished by social networks, sites and engines change over time. This is not a bad thing and is to be expected: after all they are growing and evolving, constantly learning about their platforms and users, figuring out what works, adjusting their endometrics, and revising their sales pitches accordingly.
5 Steps to Overcome the Endometrics Problem and Improve Your Digital Measurement
Platforms adopting new business models make the case for change. What may be best for a platform interests, however, may or may not be best for a brand’s interests. Why? Each is pursuing its own business strategy. Here are 5 steps that help align your communications goals and measurement with the platforms used.
- Develop a crystal clear view of the philosophy and business model for any network, site or engine you are considering. Ask yourself these questions: How do they make their money? How do their metrics help them sell? Which of those metrics might be helpful to my brand given our objectives?
- Come to your own understanding of the ways a platform or combination of platforms should work for your brands. Most media plans combine networks, sites, and engines.
- Develop and agree upon a measurement framework for your communications goals, strategy, and tactics.
- Select and fit relevant vendor metrics to your framework.
- Periodically monitor and evaluate campaign performance. Delete, add, swap, and optimize metrics as results indicate.
These steps will help you select metrics that measure impact of your communications, tell your story to business partners, and provide the guidance you need to reach your goals. Don’t forget to revisit and reevaluate the platform’s business model to keep abreast of changes and modify your strategy and metrics as necessary.
Link of the Week: Facebook’s EdgeRank – List of Factors and Changes
Facebook’s EdgeRank algorithm undergoes periodic revisions to decide which items to place in users’ newsfeeds, and thus affects a brand’s potential reach. Buffer’s Kevan Lee maintains a very helpful list of algorithm factors and changes.