Success message
Error message
The spark for this week’s post comes from a study by Germann, Lilien and Rangaswamy (2013): Performance implications of deploying marketing analytics. It’s a paper that gently (and by “gently” I mean “with the force of a polite academic sledgehammer”) reminds us that marketing analytics is still wildly underused—and that this underuse has consequences for decision making across organisations, not just for marketers but for the audiences they’re trying to reach. One of the standout findings: in a survey of 587 executives from large international companies, only 10% said their organisations regularly used marketing analytics. Ten. Percent. That’s the same proportion of people who claim to “love” kale.
The biggest reason for avoiding analytics? Leaders felt it slowed the business down and caused “analysis paralysis”—a term popularised by Peters and Waterman in In Search of Excellence (1982) . In other words: “We can’t make decisions because we’re waiting for the data.” A noble excuse, and one I’ve personally used to avoid choosing a restaurant.
But for me, analysis paralysis is less about waiting for data and more like standing in front of a supermarket shelf with 47 types of peanut butter. Too many options. Too much information. Suddenly you’re questioning your entire identity. As discussed in the earlier post on Choice Overload, (https://thedecisionlab.com/biases/choice-overload-bias) the real challenge isn’t what to report—it’s finding the sweet spot between having enough data to tell a meaningful story and not overwhelming decision makers to the point where they mentally check out and start thinking about lunch.
Image source Lab Bratz – It’s all in the way you look at it
The Trouble With Vanity Metrics (a.k.a. “Look, Ma! Big Numbers!”)
One of the core tasks in social media marketing is reporting on what’s happening across platforms. High level results—likes, impressions, shares, retweets—give management a quick snapshot of activity. They’re shiny. They’re comforting. They’re the marketing equivalent of fairy lights: pretty, but not especially functional. But what do these numbers mean? And are they even useful?
Understanding what to report—and how to report it—is essential. As Avinash Kaushik notes in his article Digital Marketing and Measurement Model, all communication (including social media campaigns) should be grounded in strategy. That strategy should include clear goals and objectives, and reporting should align with them. Vanity metrics can support this, but only to a point. In most cases, you need to dig deeper. Much deeper. Archaeology level deeper. Simple charts, clean visuals, and a clear narrative beat a 12 tab spreadsheet every time. And if your report includes terms like “impressions” or “mentions,” a glossary can save your audience from quietly Googling under the table.
The Three Layers of Social Media Analytics
To make sense of social media data, it helps to think in three layers:
• High Level Reporting — What happened? • Analysis — What’s going on in the data? • Interpretation — What does it actually mean?
High Level Reporting (Vanity Metrics)
This is the surface level stuff: impressions, engagement, likes, shares, retweets, video views. Management loves vanity metrics because they’re easy to understand and give the impression that “something is happening.” But without context, they can be wildly misleading. For example: “Facebook comments are up 200% this week!” Amazing! Unless, of course, all those comments are people telling you your product is terrible. Or maybe last week was simply catastrophic and this week is… slightly less catastrophic. Numbers without context are like movie trailers: they can make anything look good. That’s where analysis comes in.
The man has spoken! - Image source www.analytics.com
Analysis (Examining the Data)
Analysis means breaking down metrics to look at trends over time, performance against benchmarks, comparisons across platforms, and so on. It tells you what is happening. But it often doesn’t tell you why. For that, you need interpretation.
Positively trending charts make people happy, but what do they actually mean? - Image source : Google Images
Interpretation (The Deep Stuff)
Interpretation is the most under resourced and undervalued stage of the analytics process—and yet it’s the one that actually drives decisions. It’s not just spotting trends; it’s understanding their implications. It’s identifying what the data tells you, what it doesn’t tell you, and what you need to fill in the gaps. As Connor Jeffers (2020) points out, many businesses don’t invest in dedicated analytics interpretation because they don’t see it as a priority. But interpretation is where assumptions get challenged, theories get tested, and evidence based narratives emerge. Without it, decision makers are left to guess—and guessing is rarely a winning strategy.
In a nutshell - Image Source : Google Images
A Real World Example (Featuring Champagne and Disappointment)
During a recent campaign, a video posted on social media was declared “the most successful yet.” Why? Because it had the highest number of views in six months. Cue high fives and champagne. But once someone actually interpreted the data, the truth emerged:
• The previous campaign ran on YouTube. • The new one ran on Facebook, where autoplay inflates view counts. • 95% of viewers watched less than 5 seconds.
In terms of the campaign’s actual objective—getting people to watch the full video—it was one of the worst performers to date. This is why interpretation matters. Without it, the next campaign would have followed the same flawed path, fuelled by misleading vanity metrics and misplaced optimism.
Data, it is all in the interpretation. Reducing cheese intake saves lives!! - Data sources noted in image