There is a conversation that happens in almost every committee meeting where marketing results are reviewed. Someone presents the ROI of a campaign. The number is positive. The team nods. The investment is deemed sound.
What nobody asks is whether that ROI could have been twice as high.
ROI as a photograph
ROI measures what already happened. It is, at best, a photograph of past efficiency. But a photograph does not tell you whether the subject was running towards a cliff or the finish line.
To know whether a campaign is going to work before it ends, you need to look at three things ROI does not capture.
First metric: cost per customer acquired on first purchase
Not cost per lead. Not cost per click. The real cost of getting a new customer to make their first transaction.
This figure is the only one that matters at the top of the funnel, because it determines whether the acquisition model is sustainable. A customer that costs eighty-five euros to acquire and spends forty on their first purchase is not a problem if their lifetime value is six hundred euros. But if you abandon the relationship after that first purchase, those eighty-five euros are a loss.
Most marketing teams optimise cost per lead because it is the most visible metric on platforms. Cost of first purchase requires connecting marketing data with transaction data. That extra effort is exactly where the difference begins.
Second metric: dormant customer reactivation rate
A company with three years of operation almost certainly has a segment of customers who bought once or twice and disappeared. That segment is, in most cases, more valuable than any new audience you can acquire.
Why? Because they already cleared the barrier of the first purchase. They already trusted the brand at least once. The reactivation cost is consistently lower than new acquisition cost, and the conversion rate is higher.
Yet fewer than twenty per cent of the companies we work with have an active reactivation programme based on behavioural data. The rest invests almost its entire budget in acquiring new customers while letting previous ones go cold.
The reactivation rate does not appear on any standard campaign dashboard. It has to be built. That is exactly what separates reactive marketing from structured marketing.
Third metric: cross-channel cannibalisation index
When a brand operates across several channels simultaneously, there is a risk that is rarely measured: the channels are consuming budget from each other rather than adding up.
The most common example is the relationship between paid search and organic search. If a user who was already going to find your website organically clicks on a paid ad, you have paid for a conversion you would have received for free.
At scale, the same effect occurs between television and digital, between email and retargeting, between influencers and display. Without measuring the real incremental contribution of each channel, the attribution model you use will systematically overestimate the value of some channels at the expense of others.
What these three metrics have in common
All three require connecting data sources that typically live in silos: paid platforms, CRM, transaction data, web analytics.
They are not complicated metrics. They are metrics that require data discipline.
And that discipline, more than any tool or platform, is what separates the marketing teams that generate competitive advantage from the ones that simply report results.