Most marketing teams have plenty of numbers. Dashboards, exports, platform reports, a spreadsheet someone updates when they remember. What they usually do not have is a scorecard.
A scorecard is a short list of numbers that tell you whether marketing worked this week. Each number has one owner, one goal, and one weekly moment where somebody says on track or off track out loud.
That last part is the whole trick. A dashboard is something you look at. A scorecard is something you answer for.
Why dashboards do not create accountability
A dashboard invites browsing. Forty metrics, no hierarchy, no names attached. When everything is visible and nobody owns anything, a bad number can sit in plain sight for a month.
A scorecard forces a decision. You pick the handful of numbers that matter, you attach a person to each one, and you review them on a schedule. The number cannot hide, and neither can the owner.
Choosing the metrics
Start with five to fifteen numbers. Fewer than five and you are probably missing a stage of the funnel. More than fifteen and the review turns back into a dashboard.
Run each candidate metric through three questions:
- Can you measure it weekly? Quarterly numbers do not belong on a weekly scorecard.
- Can one person influence it? "Revenue" is a company number. "Qualified leads from paid" is a scorecard number.
- Would a bad week here change what you do next week? If not, it is trivia, not a metric.
Mix leading and lagging. Lagging numbers, like cost per qualified lead or closed deals, tell you what already happened. Leading numbers, like ads entering testing or pages shipped, tell you what is about to happen. A scorecard made only of lagging numbers reports on a game that is already over.
For a paid media team, that might look like spend pacing against budget, cost per qualified lead, new ads entering testing, and landing pages shipped. For a content team: pieces published, pages indexed, organic conversions, internal links added.
Do not chase perfect. Pick the numbers, run them for a month, then fix the list. A decent scorecard reviewed weekly beats a perfect scorecard that never ships.
Every number gets one owner
One name per line. Not a team, not a channel, not "shared."
The owner is not always the person doing all the work behind the number. The owner is the person who reports it, explains it, and raises a flag when it turns red. When a number belongs to a team, it belongs to nobody, and you find out in the meeting that everyone assumed someone else was watching it.
Owners should own numbers they can influence. Handing your designer the cost per lead line creates frustration, not accountability. Hand them "creative concepts entering testing per week" instead.
Set goals that produce a clean red or green
Every number needs a line. Above the line is green. Below it is red. No yellow.
Yellow is where accountability goes to die. A number that is "sort of close" gets a pass for weeks on end, and later nobody can say when it actually broke. Pick the line, even if the first version is a guess, and adjust it after a few weeks of real data.
Set weekly goals from the annual plan. Take the yearly target and divide it down to a week. The math is simple on purpose. Anyone on the team should be able to trace the weekly line back to the number the company committed to.
Live data beats self-reported
There are two ways to fill in a scorecard. Someone types the numbers in every week, or the numbers pull themselves from the source systems.
Self-reported works on day one and decays from there. People forget, people round up, people paste last week's figure when they are busy. Nobody is lying. The number just drifts away from reality, and the review slowly becomes a meeting about opinions.
Wire the scorecard to the source instead. Ad platforms, analytics, the CRM. When the number comes from the system of record, the meeting argues about what to do, not about whether the number is right. This is exactly the plumbing we build in analytics and reporting. Reporting that writes itself is the difference between a scorecard that lasts a quarter and one that lasts for years.
Some numbers resist automation, like "sales conversations had." Keep those self-reported, but mark them as such, and hold them to a written definition everyone agreed on.
The weekly 20-minute review
The review is a standing meeting. Same day, same time, twenty minutes, hard stop.
The agenda:
- Read the scorecard top to bottom. Each owner states the number and one phrase: on track or off track. No stories yet.
- Drop every red number onto an issues list. Not solved. Listed.
- Pick the one or two issues that matter most. Give each a next step, an owner, and a date.
The discipline is refusing to solve problems during the read-through. The moment someone starts explaining a red number on line two, the meeting is over for lines three through twelve. Read first. Then discuss the short list.
Twenty minutes sounds impossible until you notice how much of a typical reporting meeting is spent finding the numbers, disputing the numbers, and narrating slides. With live data and one-phrase answers, twenty minutes is roomy.
What to do when a number goes red
Red is not a verdict on the owner. Red is information. Treat it that way and people stop hiding bad news.
When a number goes red, the owner brings three things to the review:
- What happened, in one or two sentences.
- Whether it is a data problem, an execution problem, or a goal problem.
- A proposed next step.
Those are three different diagnoses with three different fixes. A data problem means tracking broke, and the fix is plumbing. An execution problem means the work did not happen or did not work, and the fix is a change in activity. A goal problem means the line was set wrong, and the fix is resetting it out loud, on purpose, not by quietly ignoring the miss.
One red week is a conversation. Three red weeks in a row is an issue that gets a real plan and a date. A number that stays red forever with no consequence teaches the team that the scorecard is decoration.
Where this fits in the bigger system
A scorecard is one layer of the operational system underneath marketing. It sits on top of clean tracking and automated reporting, and it feeds the weekly cadence that keeps work moving. That layering is the core of how we approach marketing systems.
We also coach teams through this exact build: picking the metrics, wiring the data, and running the review until it sticks. That is what scorecard coaching is for.
Start with the layer underneath
Most teams that struggle with scorecards do not have a metrics problem. They have an operational layer problem. Reporting is manual, definitions are fuzzy, and the data cannot be trusted enough to review weekly.
That is what the Performance Audit is built to find. We map how your marketing actually runs, from tracking to reporting to cadence, and hand you a dollar-weighted plan you keep whether or not we do the work together.