Table of contents
Quick answer: Scope a marketing KPI scorecard backwards from decisions: list the calls it must inform, pick a handful of leading numbers that are measurable this week, give each one owner, a target and a floor, then fix a weekly cadence.
Last verified: 2026-09-11
Start from decisions, not from available metrics
Most scorecards are assembled from whatever the tools export, which is why they get ignored by week six. Scope the engagement the other way round: write down the decisions the leadership team actually makes on a weekly and monthly rhythm — reallocate budget, pause a channel, hire, change an offer, escalate a pipeline gap — and then ask which number would change each call. Anything that informs no decision is not a scorecard line.
That constraint is what keeps a KPI advisory engagement from becoming a dashboard project. A KPI exists to trigger behaviour, and a balanced-scorecard layout only works when each line has a target, an owner and a consequence attached.
Insist on leading, measurable numbers
Revenue and closed-won belong on a financial report, not on a weekly marketing scorecard: by the time they move, the week that caused them is gone. Weekly lines should be upstream — qualified conversations booked, opportunities created, cost per qualified lead, proposal volume, publish or test throughput — the measures that move before the money does.
Then test measurability honestly. Each candidate must be pullable this week from a named system with the same result twice: GA4 events for on-site behaviour, the ad platform for cost and volume within its own conversion windows, the CRM for pipeline. If a line needs a manual spreadsheet reconciliation every Monday, scope the measurement fix first and add the line later. Scoping a metric you cannot yet measure is how scorecards die quietly.
One owner, one target, one floor
Every line gets a single named owner who can actually move it. Shared ownership produces the weekly ritual of nobody answering. Beside the owner sit two numbers: the target for a normal week, and the floor that triggers a conversation. Without a floor, a red line is just a colour.
Keep goals and measures separate. Ambitions and stretch outcomes live in an OKR or a marketing plan; the scorecard tracks the steady operating numbers that tell you whether the plan is being executed. Mixing the two turns the weekly meeting into a strategy debate.
Write the scope of work in deliverables
Put the engagement in writing as artefacts with dates, the way any competent statement of work reads: the decision inventory, the metric definitions document, the measurement gap list, the built scorecard, and a facilitated cadence for a fixed number of weeks. Name what is out of scope too — attribution rebuilds, CRM migrations and dashboard design are separate work and will otherwise swallow the engagement.
Where the gap list turns into real remediation, route it: conversion tracking for missing conversions, analytics for definition and reporting structure, server-side tracking only where collection is genuinely lossy. That keeps the scorecard engagement short and the fixes accountable.
| Scope item | Deliverable | Typical duration |
|---|---|---|
| Decision inventory | List of weekly and monthly calls the scorecard must serve | Week 1 |
| Line selection | Five to fifteen leading measures, each tested for measurability | Week 1 |
| Definitions | Written definition, source system and query per line | Week 2 |
| Gap list | Measurement defects blocking any chosen line | Week 2 |
| Built scorecard | Sheet or dashboard with owner, target and floor per line | Weeks 3-4 |
| Facilitated cadence | Weekly meeting run to the same agenda, then handed over | Weeks 4-12 |
Budget the cadence, not just the build
The build is the cheap half. A scorecard becomes real when someone chairs the same meeting, in the same order, for a quarter — long enough that missing numbers get embarrassing and owners start pre-empting the question. Scope facilitation explicitly, with a handover date and a named internal chair.
Scope the effort in weeks and named artefacts rather than in a lump: the build, the facilitated quarter, and any measurement repair the gap list exposes are three different commitments, and pretending they are one is how engagements overrun. We quote after scoping, once the gap list is known — the help library covers the process, and delivery sits under data intelligence.
What goes wrong
The failure mode: too many lines. Forty metrics, nobody can recite them, and the weekly meeting turns into a data-quality inquest. If the team cannot name the lines from memory, the scorecard is a report, and reports do not change behaviour.
Second failure mode: lagging-only lines. A sheet made of revenue, closed deals and quarterly totals cannot be steered weekly, so the meeting becomes a retrospective with no available action.
Third: scoping lines you cannot yet measure. Every unmeasurable line trains the team to treat the whole sheet as approximate. Cut it, fix the measurement, add it back in a later cycle. Constraint thinking helps here — see theory of constraints — and access hygiene, including MFA on every admin account per CISA guidance, has to be settled before anyone can pull numbers reliably.
Frequently Asked Questions
How many lines should a marketing scorecard have?
Five to fifteen. The working test is whether the team can recite them without looking; beyond that the weekly conversation loses focus.
Is a scorecard the same as an OKR set?
No. OKRs carry ambitions for a quarter; the scorecard tracks steady weekly operating numbers. Keeping them separate stops the weekly meeting becoming a strategy debate.
What if the numbers we want are not measurable yet?
Scope the measurement fix as its own item and leave the line off the sheet until it can be pulled twice with the same result.
How long before a scorecard changes behaviour?
About a quarter of unbroken weekly cadence. The change comes from the repetition and the named owners, not from the sheet itself.
Sources: GA4 events, Google Ads conversion windows (Google); CISA MFA; KPI, Balanced scorecard, OKR, Statement of work, Marketing plan, Theory of constraints (Wikipedia). Verified 2026-09-11.
