The problem is almost never the tool
On the accounts we take over, the installation exists and the figures display. What is missing is for those figures to mean something.
Conversions often count false conversions
A click on a phone number that never connected. A thank-you page reloaded, counted twice. A form that counts the opening and not the sending. Until that point is checked in real conditions, by submitting an enquiry yourself from end to end, no decision taken on those figures is reliable.
Internal traffic distorts small volumes
On a site receiving a few hundred visits a month, the team looking at its own site is a notable share of the total. It is a two-minute exclusion, rarely done.
What the tool will never tell you
It measures visits, not customers. Reconciling with your invoicing stays manual and it is what counts: a traffic source can be excellent in volume and poor in revenue. So we always look at both sides.
On consent
Since consent banners, a share of visitors is no longer measured. Your figures are therefore structurally lower than reality, by a proportion nobody knows exactly. That does not stop you deciding, provided you compare periods with each other rather than believing in absolute values.
