Marketing capability model
Definition
Marketing capability model
Marketing capability model is the levelled list of things a marketing team must be able to do, with each capability scored from basic to advanced. A business capability model does this across the whole company; this one is the marketing version of it.
The live business capability model page explains the instrument: name, description, owner, level. Read it once, then stop — this page is about the marketing list, not the instrument that holds it.
The scoring method in general belongs to a maturity assessment framework, which sets the levels, the evidence and the review cycle for any function. A marketing capability model is the content such a framework scores.
Key takeaways
- A marketing capability model names what the marketing function must be able to do, and scores each capability by level.
- The capability list is marketing specific, so a generic corporate taxonomy will not score a marketing team usefully.
- A level 2 and a level 4 differ in evidence — not in effort, and not in how busy the team feels.
- Capabilities cut across job titles, so scoring by organisation chart produces a flattering, useless result.
How it works
A marketing capability model works by fixing a short list of capabilities, writing an observable description for each level, then scoring the team on what it actually produced last quarter.
The gap between the current level and the target level becomes the plan. It tells you what to hire for, what to train, and which capability is cheaper to buy from a partner than to build in house.
| Capability | Level 2 looks like | Level 4 looks like |
|---|---|---|
| Audience data | One list, manual exports, duplicates tolerated | Governed profiles, consent recorded, refreshed nightly |
| Campaign measurement | Last click reports, argued over monthly | Agreed definitions, incrementality tested, disputes rare |
| Content production | Occasional briefs, freelancers found per job | Standing briefs, named owners, a reuse target |
| Channel operations | One channel run well, the rest copied | Each channel owned, tested and funded on evidence |
| Partner management | Providers briefed by email | Scorecards, quarterly reviews, exit plans written |
Levels are descriptions — not marks out of five. Each one names behaviour a reviewer can observe, so the level is arguable from last quarter’s output alone, without asking anyone how mature they feel.
Target levels are not all level 4. Picking the top level everywhere turns the model into a wish list, so most teams set level 4 for the two or three capabilities that differentiate them and level 3 for the rest.
Capabilities are not teams. One capability sits across three job titles and one person often carries two, which is why a model scored by organisation chart flatters everybody and changes nothing.
Agile marketing is usually scored as its own capability, because it changes how every other one is run: short cycles, standing teams, and a backlog in place of an annual brief.
Role pages show what a high level looks like in a person. A growth marketer carries experiment design, a digital marketing specialist carries channel execution, and a social media marketer carries community and publishing.
Examples
Capability models earn their cost when the score changes a decision. Four uses are common: a data plan, a measurement plan, a sourcing decision, and a review borrowed from an existing assessment guide.
Scoring audience data. A retailer scores level 2: one customer list, exported by hand, with duplicates nobody owns. The level 4 target names a governed profile store with consent recorded, which turns into a two quarter data project rather than a vague ambition.
Scoring campaign measurement. A business to business software firm finds last click reporting that every department disputes. Moving to level 4 means agreed metric definitions and a tested incrementality read, which is a measurement project with a named owner.
Sourcing against the gap. Where the gap is capacity rather than judgement, teams buy it instead of building it from scratch.
Most capability reviews are run by outsiders anyway. The United States Bureau of Labor Statistics notes that although some management analysts work for the company they analyse, most work as consultants on a contractual basis.
Borrowing an assessment shape. Formal assessment guides already exist for specific practices, and their structure travels well.
The United States Government Accountability Office published a guide in November 2023 setting out best practices to assess the adoption and use of Agile processes. It was written for federal auditors — the shape is recognisable to anyone scoring a capability.
Related terms
A capability model sits beside the instrument that defines it and the roles that carry it. These five terms are the ones most often mistaken for it, and each bullet says what that page covers instead.
- Business capability model: the generic instrument, applied across a whole organisation rather than one function.
- Agile marketing: a way of working that is usually scored as a capability in its own right.
- Digital marketing manager: the role that usually owns the channel operations capability end to end.
- Lifecycle marketing manager: the role carrying retention and audience data capabilities across the customer journey.
- Growth marketer: the role that evidences the experiment design capability at higher levels.
FAQ
What is a marketing capability model?
It is a short list of the things a marketing function must be able to do, with each capability described at several levels and scored against observable output. The score drives hiring, training and sourcing decisions.
How many capabilities should the list have?
Between eight and fifteen at the top level. Beyond that, processes have crept into the list, and processes multiply in a way that capabilities do not.
What is the difference between level 2 and level 4?
Level 2 means the capability exists and works when somebody pushes it. Level 4 means it runs to agreed definitions, produces evidence, and keeps working when the person who built it leaves.
Who should score the model?
Somebody outside the team being scored, with the capability owner present to supply evidence. Self scoring without evidence produces a flattering picture that changes no decision.
How often should it be rescored?
Annually for the whole list, and after any reorganisation or large sourcing change. Rescoring more often costs more than the movement it detects.
Providers that can evidence a specific marketing capability can show that record in the Outsource Accelerator hubs.
Independent
