Capturing the Right Campaign Data Part 1

Many companies, when they think about the campaign data they’d like to capture, suffer from a syndrome familiar to anyone dining at an all-you-can-eat buffet: their eyes are too big for their stomach. Like hungry diners, companies start out piling on the goodies (Let’s measure this! And that!), fully intending to come back for multiple rounds, only to find that their appetite and capacity to digest is quickly challenged.
When you ask marketers at these companies what they’d like to know about their campaigns, the first answer is often, “Everything.” Now, of course, that’s not actually true. When you press, there are always variables that are more important than the others, and that rise to the top in priority. Part of the challenge in determining how to capture the right campaign data is figuring out what data is interesting, but not critical — and beginning to treat those variables as the wrong campaign data. Don’t believe me? In the data game, tradeoffs are important. Anything that isn’t critical to measure shouldn’t be measured. “Nice-to-Have” information is nothing but a distraction from information that is Must-Have.
When it comes to deciding which reports make the cut, the technicians are the ones who need to make the hard choices — those should be the people driving the business. By that, I mean anyone whose job it is to optimize spend, or effectiveness, or manage revenue across channels. If a given piece of data isn’t driving insights or changing the company’s marketing, measuring it is just an intellectual exercise. The business leaders should determine the data targets so that the technicians can become the team in charge of data excellence.
Once your business has made the hard tradeoffs, then you need to come up with a taxonomy for identifying new campaign variables and rules to ensure that things don’t go awry. Sadly, 99% of the world still does this with spreadsheets. That’s not a best practice. Getting off spreadsheets will remove a chronic bottleneck, making your efforts scalable, and will allow more teams to engage in the work of campaign tracking — ideally without ruining everything.
You’ll also need a structure for consistently naming and containing all the metadata. You’ve chosen your critical dimensions; now how do we make sure those reports stay clean, regardless of how many teams (including outside agency teams) have access to them? Keeping the data clean will require building sufficient controls so that the people doing the important work of creating campaigns have the option to complete their work, but not the ability to knock things askew. Ideally, you’ll want to tailor the input environment so that marketers only see the variables related to their work. Rather than overwhelm them with dozens of rows of input (many of which may not apply to them), offer them a view where they are able to input every bit of data within their purview and not a whit more.
Last thing: for campaigns to work, there needs to be a transmission from the campaign landing page to the analytics vendor, and it has to be perfect. If any single character or keystroke is awry, the baton is dropped and you literally get nothing. It’s not a decrease to ~90% accuracy — it goes from 100 to 0. It’s tricky for the generalist to do right (and it’s a topic for another day).