In the movie Moneyball, Brad Pitt is in a room full of veteran scouts doing what they've always done: debating whether to sign players based on how they look, whether they have a good swing, and whether their girlfriend is attractive. He asks the group, "What problem are we trying to solve?"
They've all forgotten why they are there. This happens in board meetings everywhere.
What is Performance Marketing?
There's a job title that's been spreading across marketing teams for the last decade: performance marketer. I've had this title, and it's total bullshit. This role is usually somebody running ads, email marketing, SEO campaigns, things that can be tied to a conversion (outcome) These also happen to be the bits most marketers find boring; despite what tech and growth bros suggest.
The Marketing industry lost the plot on this one
The word performance is doing no work here beyond making accountability sound like a specialism. It could just as easily be called:
marketing where someone checks if it worked.
The moment we accept that some marketing is “performance†marketing, we are accepting that the rest of it is just fun.
Why measure marketing at all?
You measure marketing when you want to know whether it's working. Most marketing operates in a space where worked is loosely defined as someone saw it or somebody said "That post was cool" on Slack.
Measurement forces a different question: did this activity move the number we said we cared about?
- If you spent 50k on content and photo shoots did it contribute your business goal?
- If you published 47 blog posts, did they contribute to your business goal?
- If you boost high engaging posts, did they contribute to your business goal?
Things that sound good in a deck are different from things that produce results.
Brand work, creative campaigns, and long-term positioning all have value. But if you can’t articulate what success looks like before you start, you're not doing accountable marketing.
Teams with tight budgets need to measure it more than anyone.
When every pound matters, you can't afford campaigns that “build awareness†without any way to know whether awareness is being built. Large teams with big budgets need it too. They just have the luxury of being able to absorb the waste.
The moment someone asks, is this working? is the moment accountability kicks in.
Before that question gets asked, you're operating on faith, taste, or tradition. All fine ways to make decisions, but everyone in a marketing department at some point should question their existence in this way or you're not marketing anymore, you're admin.
How measured marketing actually works
You define an outcome. You do an activity intended to produce that outcome. You check whether the outcome happened. If it didn't, you stop or change your approach until it does.
The difficulty is in the execution. Defining the outcome requires honesty about what you're trying to accomplish.
- Engagement isn't an outcome.
- Leads isn't an outcome without sales
- Awareness isn't an outcome unless you can show it produces something the business values.
What remains is simpler: did the thing you wanted to happen actually happen?
The measurement phase is where most efforts break down.
Measuring the right thing requires deciding what the right thing is, and that's uncomfortable. It means admitting that some activities don't matter. It means saying no to campaigns that sound great but produce nothing measurable.
Then comes the action phase.
This is where accountable marketing separates from accountable theatre. If the data says something isn't working, you should stop doing it. Or do it differently
If it says something is working, you do more of it. Most teams skip this step entirely. It sounds like this:
- We've got good engagement
- We're on page one, most traffic is organic
- We had x amount of traffic
- Our open rates are good
- We've got more leads this quarter than any other
All sounds good. But are they your business goals? What did people do after this?
When to start
Day one. When you know you'll have to report on whether something worked, you think harder about it.
Avinash Kaushiks Digital Marketing and Measurement Model starts with business objectives:
1. What is the business trying to accomplish?
2. How can marketing help?
If you can't answer it, you're not ready to measure anything, because you don't know what good looks like yet. Early measurement is also when you establish what normal looks like. You need a baseline. Without one, every number exists in a vacuum. A 2% conversion rate means nothing if you don't know whether 2% is good, bad, or exactly what you should expect given your product, price, and audience.
The mistake is waiting until you have enough data to start. You never have enough data. You always wish you started tracking earlier. Start now. Track the basics. Add sophistication later when you've earned it through volume and repeatability.
What to track (not everything)
- Did you hit that business goal over the time frame?
- Did the marketing do the things it was meant to do to help?
- Sales, leads, sign-ups, downloads, then work backwards
This looks like
- Didn't reach enough people
- Didn't convert enough people
- not enough click-throughs.
Don't try and provide a solution without a diagnosis. That looks like this:
- We need a CMO
- We need a performance marketer
- We need this tool
- We need an investor
The tools you need
A spreadsheet, to start. Before you buy anything, track everything manually for 90 days. Manual work clarifies thinking.
You'll discover that half the things you thought you needed to track don't matter, and the things that do matter are simpler to measure than you imagined.
- Sales were down this week - what happened?
- Sales were up this week - what happened?
- Sales were up again this week
Now you've done 4 weeks and got a benchmark, now you can set a target
If you're not measuring consistently with the tools above, buying more tools won't help. It'll just give you more dashboards to ignore.
Shutting out the noise
Tracking everything and acting on nothing. Measurement without action is pointless. Do something based on the data, or do nothing.
Confusing activity with outcome
Blog posts published, emails sent, ads run: these are activities. Outcomes are the things that happen because of them: Traffic, leads, sales, retention. If your report is a list of things you did rather than things that resulted, it's wrong.
Optimising too early.
You need volume before optimisation matters. If you're running emails to 50-100 people, split testing creative is pointless work on list growth, deliverability, open rates, the things that will give you the extra inch.
Blaming the measurement when the marketing doesn't work.
The data didn't lie to you. The campaign just didn't work.
Setting goals you can't influence.
Pick a goal you can track. It's more satisfying, either way.
Brand and accountability aren't opposed.
They're sequential. A strong brand creates conditions where all your other marketing gets easier. The mistake is treating the brand as immeasurable and therefore exempt from accountability. Everything is measurable if you're willing to define success in advance. If you can't define success, you're not doing brand marketing. You're making things and hoping someone notices.
- Measure brand lift.
- Track branded search volume.
- Monitor direct traffic.
These are all proxies, but they're measurable proxies you can plot over time and correlate with activity.
If you had to cut your marketing budget in half, do you know which half?
If yes, you're doing it right. If the answer is no, start measuring outcomes, then tie them to an activity.
There are high-performing marketing teams, average-performing marketing teams, and there is 50 feet of crap; The teams buried under it are the ones who can't answer the question:
What problem are we actually trying to solve?