We spend money on television as well as digital, and the two don’t sit next to each other as neatly as the reporting pretends. Here’s the problem in a sentence: when the telly is on, more people search for us, more people come to the site directly, and a good chunk of that lands in the digital reports as though digital had gone and found them, when the television is what sent them.
You can see how it happens. A brand runs a TV burst, and over the following days brand searches climb, direct traffic climbs, even generic search lifts a little. All of that flows through paid and organic and shows up as digital performance. If you judge your PPC on the last click, you’ll conclude your PPC got suddenly brilliant the same week the television happened to be running.
The way out is unfashionable, because it means reporting some less impressive numbers. You establish a baseline: what the business does when the TV is dark. Then you can see the lift the television creates on top of it, and stop crediting digital with demand the telly generated. We’ve started using a TV attribution tool, Adalyser, to get a proper read on that contribution rather than guessing at it, which at least gives the argument some evidence to stand on.
What this changes in practice is the order you think in. Television, and brand-building generally, is doing a lot of the demand creation; digital is mostly capturing that demand and converting it. Both matter. But optimise the capture layer on its own and ignore what feeds it, and you’ll end up cutting the brand spend that was actually filling your funnel, because the spreadsheet told you digital was the thing that worked.
None of this makes digital less important to my job. It has made me slower to read a spike in my own numbers as proof I did something clever, when the likelier explanation is the TV schedule that week.
