When marketing metrics disconnect from goals

Too often, marketers measure the wrong goal, measuring diagnostic metrics instead of actual goals.

My friend and mentor Tom Webster recently shared a fascinating article about the computation of the calorie.


Already an inexact measure, scientists now suspect measuring the calorie may be far more inaccurate. Counting calories alone may not help a dieter achieve their weight loss goals.

Why do dieters count calories? For the same reasons we marketers count email open rates or clicks on ads: we seek measurements we can understand and influence easily.

Dieters need to make several changes to lose a pound of body weight, and many variables can confound that measurement. The calorie is easier to count and easier to influence. A weight loss seeker can not eat a cupcake or eat an apple instead and see an immediate change in calorie counts.

Many variables can confound a sale or even an inbound lead. We can much more easily influence clickthrough rates on our ads by changing the bid or updating the copy. We can much more easily understand and influence email open rates by messing with a subject line.

Diagnostic metrics like calories or upper funnel metrics aren’t inherently bad. Choosing to eat a salad instead of a bacon triple cheeseburger impacts our health and weight loss. We’d be fools to believe otherwise.

All other variables being equal, a 5% clickthrough rate on our ads is better than a 2% clickthrough rate. We’d be fools to choose the lower clickthrough rate with equal lower funnel metrics.

Where we run into trouble is when a diagnostic metric uncouples from our goal.

As researchers are finding, our bodies process different foods in different ways; two steaks can have the same number of calories in them from a theoretical perspective, but we digest fewer calories from one than the other. Our weight loss efforts – our goal of reducing body weight – can end up drastically different despite the same theoretical number of calories on a label. Our metric has uncoupled from our goal.

Marketers are in the same boat. Our goals of conversions or revenue can end up drastically different when our diagnostic metrics disconnect from goals. For example, we can send the same amount of traffic from an ad to two different landing pages; page A converts at 25%, while page B converts at 10%. In this example, page A’s ads can have a lower clickthrough rate than page B’s ads and still drive more revenue. Again, our metric has uncoupled from our goal. The same metric gives different results.

How do we keep our diagnostic metrics and goals aligned? All reporting should have our actual goals in it. Whether we’re doing an email marketing report or a social media report, our end goal – conversions – should always appear. We may need more sophisticated analysis tools to correctly attribute the upper funnel to lower funnel goals, but doing so ensures our diagnostics have not disconnected from our goals.

Diagnostics are valuable. We can understand them easily, change our tactics, and see fast results. We need them. We cannot rely solely on them; always keep your absolute goals in line of sight.

If you enjoyed this, please share it with your network!

Want to read more like this from ? Get updates here:

Marketing Blue Belt Preorder

Order your 2016 Marketing Planning Framework

Why share of voice no longer matters

Share of voice is nearly useless as a measurement because the media landscape is larger and more fragmented than ever, but share of voice metrics fail to take the landscape into account.

I’ve seen no fewer than a dozen dashboards and Powerpoint slides recently which reference share of voice as a marketing KPI. Other than making things up, I can’t think of a worse KPI for marketing.

First, share of voice is a function of media, not marketing. It belongs in the realm of advertising and PR.

Second, share of voice is a nearly useless measurement in today’s media landscape. The average share of voice conversation goes something like this:

“Out of 3128 social media conversations mentioning us and our competitor, our brand had 15% share of voice. We are (awesome/terrible)!”

Why is this nearly useless? Share of voice suffers from what we measurement folks call denominator blindness. Denominator blindness is a lack of perspective on our part. For example, we might read a headline in the news which says “150 vaccinations last year had serious side effects!”. What’s left out of the story is the denominator: 150 out of 10 million annually. When you apply a denominator, suddenly the story becomes far less compelling.

How does denominator blindness impact share of voice?

Consider the above example. Suppose we were a local coffee shop and we were measuring our share of voice against a major chain coffee shop. We netted 15% share of voice out of the mentions of us vs. our competitor, or 469 mentions. That’s great, isn’t it?

Except… on the topic of coffee alone, hundreds of thousands of people talk about coffee daily:


Our competitor AND our shop combined amount to less than 1.5% of the conversations about the topic. That’s one of the denominators we’re blind to – and it’s not the biggest one.

Let’s expand the denominator further. By recent estimates, we are sending more than half a million Tweets a minute. We watch almost 3 million videos on YouTube a minute. We update Facebook 300,000 times per minute. We load more than 100,000 photos to Instagram a minute.

469 mentions of our coffee shop are insignificant compared to the vast, ever expanding media universe.

Share of voice made a great deal of sense when there were 3 television networks, a handful of local radio stations, and a few hundred newspapers. We could accurately measure our portion of the entire media universe at the time. Today, with apps like WhatsApp and Facebook Messenger sending millions of unobservable messages, combined with public social and digital feeds, we can no longer know what the total landscape is, much less measure our portion of it.

What should we measure? Continuous improvement – kaizen, in Japanese. If we netted 469 mentions today, try for 470 tomorrow. Focus on what we can do to grow our tiny patch of land, our tiny empire, a little more every day, every week, every month.

We compete for the attention of our audiences against our competitors, against apps and games and mass media and the rest of the world clamoring for the same slice of attention. Rather than worry about whether our competitor has a bigger slice, worry about holding onto and growing the slice we have.

If you enjoyed this, please share it with your network!

Want to read more like this from ? Get updates here:

Marketing Blue Belt Preorder

Order your 2016 Marketing Planning Framework

Don’t make a meal if no one eats the snack

Avoid investing in high-effort content creation if the idea gets no traction in low-effort content forms.

Consider our content creation options. Some of our options take relatively little effort and production time:

  • Social media update (text)
  • Blog post
  • Photo
  • Illustration

Other forms of content require a lot of production time and effort:

  • Audio
  • Video
  • Interactive experiences
  • Software

All these forms of content begin with an idea, with a story we want to tell. When we’re deciding what our content strategy should be, what formats should we use?

The answer is the now-maligned snackable content. Let’s bring our ideas to life in the formats requiring less production time and effort. If no one appreciates or engages with our ideas in a quick photo on Instagram or retweets our idea, we should reconsider investing more time, effort, and resources in it. Conversely, if we can use our analytics and data to identify our best stuff, we can invest our time, effort, and resources in relatively ‘sure bets’.

Here’s an example from my Twitter analytics, sorted by total post engagement.


I should consider turning the top performing short content – text and photos, mostly – into longer form content because it’s proven popular already.

We can even kick it up a notch by examining our competitors, identifying what’s most popular in their content, and then doing a topically-related but unique spin if we can.


Don’t make a meal if no one eats the snack. Instead, find the top performing content snacks and turn those into meals.

If you enjoyed this, please share it with your network!

Want to read more like this from ? Get updates here:

Marketing Blue Belt Preorder

Order your 2016 Marketing Planning Framework