Watermelon Metrics: Why Your Green Dashboard Is Lying to You
- Anne Adair-Major
- 4 days ago
- 3 min read
Early in my career, I learned that you can make any metric turn green if you try hard enough.
Fresh out of university, I asked the IT department director how I could add more value. He showed me an email that came in every week. Our department had one of the worst IT incident SLA metrics globally. As an eager graduate, I saw my chance to shine. I dug into the numbers and discovered something interesting: if you put a ticket on hold, the clock stopped. So I created a one-page process doc and rolled it out to the whole team with all the enthusiasm of a 20-something determined to impress.
Two weeks later, I bounded into the director's office. "Have you seen this week's SLA report?"
He opened the email. There we were. Bright green.
Classic watermelon metrics. Green on the outside, red on the inside. I hadn't improved service. I hadn't asked why tickets took so long, instead I'd just gamed the system.
Why watermelon metrics won't go away
Watermelon metrics are KPIs that look healthy on the surface but hide real problems underneath. The term comes from that visual: slice open the green and you'll find red.
What I did as a graduate was a textbook example of Goodhart's Law: when a measure becomes a target, it ceases to be a good measure. And it's not just a graduate mistake. I see this pattern regularly in enterprise transformations, 16 years later.
The board pack looks immaculate. Every RAG status is glowing green, every milestone is "on track." But walk around, and you'll find teams drowning in manual workarounds, ITSM processes no one has heard of, and every department inventing its own platform rules because the official process doesn't actually work.
The dashboard says success, but the P&L starts telling a different story.
What I look for instead
Real digital operations aren't about perfect metrics; they're about awareness. I'd rather see a red metric we can actually fix than a fake green one hiding a costly operational leak nobody's talking about.
That early SLA experience taught me something fundamental. I went straight to how to make the number look better. I didn't ask a single question about why tickets took so long. The metric improved, but the service didn't!
Now, when I work with organisations on their operating models, I start with key questions before I look at any dashboard:
What's teh vision and purpose?
What does the team on the ground say about how things actually work?
Where is money still being spent on things the transformation was supposed to fix?
Which metrics would change if you measured outcomes instead of activity?
Those conversations, not the RAG status, tell you more than six months of status reports.
How to spot a watermelon in your own reporting
If your transformation dashboard is all green but your OpEx is still climbing, that's a signal worth investigating. I find that the telltale signs are almost always the same: metrics that measure activity rather than outcomes, KPIs that nobody on the delivery team recognises as meaningful, and any process where the workaround has just become the standard way of working.
The next steps start with talking to the people who feel the pain, digging into the operating model and data, and being honest about what you find. Sometimes the most valuable thing a report can show you is a number that's genuinely red.
One thing to try this week
Pick one metric from your current reporting that's been green for months. Ask the team closest to the work whether it reflects their reality. If there's a gap between what the dashboard says and what they experience, you've found your watermelon.
Because the reds aren't the real problem. It's the greens that should be red.
If that gap turns out to be bigger than expected, it's the kind of thing I help organisations untangle.
