The Metrics That Really Matter Aren’t Always on Your Dashboard
Field Notes
September 8, 2025
Al Straughan
Field Notes is a running series of observations from inside Paladin. Think of it as a notebook — quick takes, patterns we’re spotting and lessons we’re learning in real time. It’s not polished thought leadership, it’s useful signals from the front line of performance marketing.
We spent a lot of time recently improving clients' analytics and making their data presentable in Looker dashboards. Our most recent projects were for very different clients; one is a group of retirement villages, the other an e-commerce furniture business with over 2,000 products.
The client motivation behind the projects was identical: they needed accurate marketing data to be able to make the right decisions and, linked to that, attribution was a nightmare with each channel claiming credit for the conversion.
The challenge with getting knee-deep in numbers is the clicks, impressions, session durations etc are all there, blinking at you like Christmas lights - and I use the analogy purposefully, They draw you in like a festive gift with the promise of something shiny, new and exciting inside. It's easy to get distracted - lost in the myriad of data and think everything is important.
It isn't.
Just because a number is big, colourful or easy to measure doesn’t mean it’s important. The challenge is differentiating between visibility and value.
Take impressions. They’re the easiest thing in the world to inflate - spend more, show up more, but did anyone remember you? Did they act? Or did you just become another digital billboard people looked past? Impressions might make you feel busy, but they don’t tell you if your brand is actually cutting through.
The same goes for vanity metrics on social. Engagement can look exciting, but half the time it’s bots, click-bait or comments from people who’ll never buy from you. I’ve seen brands celebrate a viral post only to realise it drove precisely zero sales. You can’t buy stock and cover overheads with likes.
The real work is tracking key 'decision metrics'. Things like repeat purchase rate, average order value, and cost per acquisition. These are closer to business reality and tell you if people found enough value to come back, or if you’re paying too much for each new customer.
Even softer measures like brand recall, sentiment, or share of search are more useful than chasing empty numbers, because salience is vital - if nobody remembers you when they’re ready to buy, you’ve lost before the game even starts.
The challenge? Decision metrics are harder to capture. They’re often buried under layers of attribution models, customer behaviour, and delayed signals. But if you’re serious about growth, that’s where you need to dig.