Fielded in July

Forty-one per cent.

Say that is what the tracker came back with, and that the calls were made in July. By October the figure is on a slide with the word awareness under it. By spring it is quoted by people who never opened the report, and when the next wave arrives a story is told about the gap.

What the slide never carries is the fortnight the calls were made.

A great many trackers here are fielded in July, because the fieldwork calendar was set in an office where July is an ordinary month. The Gulf cities have emptied. The customer the brand is planned around is in London or Cairo or a mountain town with the phone off. Whoever picks up a survey call in Riyadh in July is a particular person: the one whose leave has not come round, the one without children, the one who could not afford to go. A different sample from the one that will answer in November.

So the number is true and the comparison is not. Awareness among the people who answered the phone in July gets read, months later, as awareness among the market.

I have sat on both sides of a tracker, reading it as a planner and being judged by it as a brand manager, and the fieldwork date is the first thing to fall off. It survives the agency's report, in a footnote. It does not survive the brand's summary. It never reaches the leadership deck, where one figure carries the whole story and there is no room for the month it came from.

Which is how a July number ends up defending, or condemning, a September campaign it never measured.

A number's fieldwork date is part of the number. It is the first thing a deck drops, and once it has gone the figure will be compared with anything.