Content marketing has a measurement problem, and pretending otherwise helps nobody. Every client I work with now accepts, in principle, that publishing useful content is worth doing. Then the finance director asks what the return is, and the room goes a bit vague.
The vagueness is understandable. A paid search click and a sale an hour later is an easy line to draw. A blog post that someone read in March, remembered, and acted on in July after two more visits and an email in between; that is a real chain of events, and our tools are genuinely bad at drawing it. So people either give up and call content “brand” and stop trying to measure it, or they over-claim and pretend a spike in traffic is money in the bank. Both are wrong, and the second gets budgets cut the moment someone checks.
Here’s roughly how I try to do it without kidding anyone, least of all myself.
Start by deciding what the content is actually for, because different content has different jobs and you can’t measure them the same way. A buying guide is meant to help someone choose and buy; you can and should hold it to a commercial number. A piece explaining how something works, aimed at people nowhere near a purchase, is doing a different job, and judging it on immediate sales will make you kill the very things that fill your funnel.
For the commercially-minded content, the tools have got better this year. Google Analytics added multi-channel funnels and assisted conversions, which for the first time let you see the channels and pages that touched a conversion without being the final click. It is not perfect and the attribution model underneath it is a blunt instrument, but it beats giving all the credit to the last click, which is what most people were doing until recently. If you’re measuring content on last-click alone, you are systematically undervaluing everything that introduces people to you.
For the top-of-funnel content, I’ve made peace with softer measures. Did the right people find it. Did they come back. Did they subscribe, share, or come closer to us in some traceable way. These aren’t pounds, and I’m careful never to dress them up as pounds in front of a finance director, because the fastest way to lose credibility is to present a fuzzy number as a hard one.
The bit I’m still wrestling with is time. Content compounds; a good guide keeps earning for years, and none of our reporting tools are built to show a return that accrues slowly over a horizon longer than a quarterly review. I don’t have a neat solution to that. I mostly ask clients to judge a body of work over a year rather than a page over a week, and hope the patience holds when the monthly numbers wobble.
