I sat through a great webinar last week put on by Compete. I imagine my reaction was similar to what The West Wing’s CJ Craig must have felt sitting through the Organization of Cartographers for Social Equality presentation. I thought I knew what I was going to see. And then I quickly realized reality looked a lot different than I’d expected.
And it started with two simple data points that weren’t the actual data, but tied directly to it. First, 35% of computers delete cookies. Second, the average number of different cookies for the same campaign among deleters was five.
Are these true of the campaign you’re running? I – obviously – have no idea. The important question is: Do you? Because it could have a huge impact on your metrics.
If you are using cookie-based metrics – and not accounting for how your users account for cookies – you may be greatly overstating your reach, greatly understating your frequency and, if you are conducting a brand study, may be artificially raising your control score.
The implications of any of those misunderstandings can steer you in a different direction. Not understanding cookies – and other variable inputs – can prematurely ruin your appetite for a further investment that might provide a greater return than you realize. It’s not that cookies are bad. Misunderstanding their context is.
The more we become dependent on data to make decisions, the more we must really understand what we do, what we don’t and what we can’t know
Interested in more stuff I find interesting? Follow me @casey_flanagan on Twitter.
SHARE:
