Every operations meeting starts the same way now. Somebody opens a dashboard. The dashboard tells everyone the channel is healthy, the campaigns are converting, the funnel is working. The room nods. The meeting moves on.
Except in the corner of your eye, you can see the regional sales director shaking her head slightly. She isn’t disputing the numbers. She’s watching a different picture entirely.
A quarter century of building and fixing direct channels for FMCG and manufacturing businesses tells me one thing about dashboards. Long enough to be wary of any dashboard everyone agrees with. The number on the screen is rarely wrong. What’s wrong is what the dashboard isn’t measuring, what it’s hiding, and what it doesn’t know how to ask.
The vanity metric problem
Most dashboards are built around marketing metrics because that’s what the people commissioning the dashboard wanted to see. Sessions. Conversion rate. Cost per acquisition. Add-to-basket rate.
These aren’t bad metrics. They tell a true story about what the marketing team is doing. The problem is that they tell almost nothing about whether the business is making money.
Sessions can rise while average order value drops. Conversion rate can improve because you’re discounting harder. Cost per acquisition looks fine until you compare it against the lifetime value of the customers you actually kept. The dashboard says everything is on track. The operating team can feel that something isn’t quite right.
The attribution trap
The second thing dashboards hide is who actually deserves the credit.
A direct channel can look like it’s growing 20% quarter on quarter. The dashboard celebrates, as it usually does. But if you look at where those orders came from, half of them are repeat customers who would have bought anyway. Another quarter are people who saw a TV ad your brand team commissioned six weeks ago. The actual incremental contribution of the D2C channel itself might be a fraction of what the dashboard shows.
Attribution is a thorny problem. Attribution models try to solve it. They mostly don’t. They allocate credit across touchpoints but rarely satisfy anyone with operational responsibility for the channel. The honest answer is that attribution is hard. The most useful thing a senior leader can do is treat all single-source attribution numbers with mild scepticism.
The cannibalisation question
For FMCG and manufacturing businesses, the dashboard hides one other thing more dangerously than the rest. Channel cannibalisation.
Your D2C channel is growing. The dashboard says so. What the dashboard does not say is whether the customer would otherwise have bought through Tesco, or through your dealer network, or through the trade account that pays your overhead each quarter. If you’ve shifted £10m of revenue from retail to direct, and the direct margin is structurally lower because you’re absorbing fulfilment costs you used to push to the trade, you might be celebrating a number that’s actually eroding the business.
This isn’t an argument against D2C. It’s an argument against trusting the D2C dashboard to tell you what D2C is doing to the rest of the operation. The whole picture lives outside the dashboard, in the trade reports, the wholesale numbers, and the conversations the sales team is having with accounts that have started ordering less.
What the sales team already knows
This points to the most useful diagnostic in any data review. The sales team usually already knows something the dashboard doesn’t.
Sales teams live in the operating reality. They hear when a major account is unhappy. They know which accounts are building their own direct alternative. They can tell you which products are stocked, which are being deprioritised, and which buyers are about to change roles.
None of this appears in a dashboard. All of it matters more than the dashboard. The sales team are paid to worry about these things. The dashboard wasn’t designed to worry. The senior judgement that turns data-driven into data-informed is largely the discipline of weighing the dashboard against the conversations.
Data-driven versus data-informed
There’s a fashion at the moment for declaring a business ’data-driven’. The intention is usually good. The implementation can be a leader hiding behind whichever number is least uncomfortable to defend.
Data-informed is harder. It means treating the dashboard as one input alongside the operating context, the commercial relationships, the brand equity question, and what the people actually doing the work are seeing. It means accepting that the senior judgement still matters, and that the dashboard is a tool to support that judgement rather than replace it.
The teams who get D2C right are not the ones with the prettiest dashboards. Pretty dashboards are ten a penny. They’re the ones who can read what’s on the dashboard, compare what they’re hearing from the field, and act on the difference.
Three questions, every time
When the next operations meeting starts, and somebody opens the dashboard, the most useful contribution anyone in the room can make is not to interpret what’s on the screen. It’s to ask what’s not on it.
Three good questions, every time.
What is this number not measuring? Who would have bought anyway? What is happening elsewhere in the operation that this dashboard cannot see?
The answers that may not be on the dashboard, may well be in the room.
If you’d like an honest external view of what your current reporting is measuring and what it’s missing, that’s part of what we do.
