What's holding your operating model back
Every marketing organization is ahead on some parts of its operating model and behind on others. This page breaks the model into six rows. Mark where you sit on each one and the line shows which row is holding the rest back.
Six rows, three positions on each.
Each channel keeps its own data, and getting it depends on knowing someone.
Some channels use shared data, but every cross-channel campaign is still a project.
Every channel works from the same data and feeds what it learns back in.
Full descriptions and examplesHide the full descriptions
Three positions
Held per channel
Each channel keeps its own data and its own numbers. Data that could be shared sits in one team's system, and getting it for a cross-channel campaign depends on knowing someone on that team.
Shared by some
There is common data and more channels use it. Consolidation has started, but which channels use the shared data, and how, still varies, and coordinating between channels is a project each time.
One shared spine
Channels take in and act on common data as the normal way of working. Segments and campaigns line up across channels, and channels still run their own experiments and feed what they learn back into the spine.
Falling behind
When the data stays in the channels, the rest of the rows cannot move far. Decisions get made on different numbers in different rooms. Agents cannot act across channels, because there is no shared data for them to act on. And roles stay channel roles, because that is where the data is.
In the four examples
In each channel keeps its own data. In the spine is finished while decisions and budget still sit where they always did.
Whoever runs the channel decides, which is quick and inconsistent.
Decisions that matter go up to a central team or a committee and come back slowly.
Written rules say who decides what, and which calls agents may make alone.
Full descriptions and examplesHide the full descriptions
Three positions
Case by case
Decisions sit with whoever runs the channel or the campaign, mostly because nobody has decided otherwise. It is quick and inconsistent, and the center usually hears about a decision after it has been made.
By committee
The decisions that matter go up to a central team or a committee. They come back consistent and slow, and people in the markets spend time waiting for them.
By clear rules
The decision rights are written and known: which decisions sit where, who can change a rule, and which decisions an agent may make on its own without a person checking first.
This is the hardest row to mark. The honest answer depends on the size of the organization, the industry and the culture, and one mark cannot carry all of it. Mark the position that fits most of the decisions that matter, and treat it as a rough shape rather than a verdict.
Falling behind
When decision rights are unclear, the other rows stall. Agents can only act as far as someone has said they may, so an organization can finish its data spine and still have the agents idle and the people arguing about who signs off. Budget follows the same rule. Money moves to whoever holds the decision, so this row and the next one tend to move together.
A local mark here is sometimes deliberate. Some organizations share the data and keep the decisions local, and if that is a choice, this row is not lagging, whatever the line says.
In the four examples
and both leave decisions with the channels. is the opposite, and the cost of that shows up in the roles row.
Once a year, split by channel, and moving it mid-year takes a sponsor.
At a scheduled review, and only when someone makes the case for it.
As results come in, inside limits that were agreed up front.
Full descriptions and examplesHide the full descriptions
Three positions
Annual, per channel
The plan is set once a year and the budget is split by channel. Moving money mid-year is an exception, and it needs a sponsor to push it through.
Quarterly reviews
Money can move during the year, but it moves at a scheduled review and only when someone brings a case for it. The cycle is shorter than a year, but a meeting is still the only way money changes hands.
Moves with results
Money follows results during the year, inside limits that were agreed up front. The plan says what the money is for, and how often it moves depends on what the results show.
Falling behind
The data changes faster than the budget does. Agents can propose a move, but the money is set for the year. A shared spine with annual funding underneath it produces very good reports about opportunities nobody can fund until next year.
In the four examples
and both still plan and fund annually, so the work moved and the money did not. moves money within the year, but only through the central team.
Each channel plans and ships its own, so one campaign becomes several.
The channels share a calendar and a brief, and one person keeps them in step.
Set up once, it runs across channels from one place, with agents inside each channel.
Full descriptions and examplesHide the full descriptions
Three positions
Channel by channel
Each channel plans, makes and ships its own work, so one campaign becomes as many campaigns as there are channels.
On a shared calendar
A shared calendar, a shared brief, a shared meeting. The channels still execute separately and the coordination is a person's job.
Run as one system
Campaigns and segments are set up once and execute across channels from one place. Agents handle execution inside channels and people coordinate outcomes across them.
Falling behind
Fewer things ship as one. Handoffs multiply, and an agent working in one channel cannot pass work to the next. This is also where a shared spine goes unused if the execution never reaches it.
In the four examples
runs each channel separately. and both run campaigns as one system, one because shared data allows it and the other because the central team requires it.
By channel, with job titles like campaign manager and system administrator.
Across more tools and more hands, flatter on paper and slower in practice.
Small teams own an outcome end to end, setting the goals while agents execute.
Full descriptions and examplesHide the full descriptions
Three positions
Channel teams
People are organized by channel and titled by it. A campaign manager runs campaigns; a system administrator keeps the platforms up.
Matrixed handoffs
People operate a growing set of tools across a growing surface, and the work passes through more hands than it used to. The team is flatter on paper and slower in practice.
Outcome teams
Small teams own an outcome end to end and supervise a set of agents. People set the goals, build the guardrails, and judge the work. Agents do the execution underneath.
Falling behind
The system gets stood up and nobody's job changes. People whose roles still point at channels keep pulling the work back into channels, whatever the spine and the tooling allow.
In the four examples
still has channel teams. In , one team has reorganized around outcomes and the rest have not. has left the market teams with thinner roles, since the work moved to the center and the people did not. On the marketing side the role names run campaign manager, then agent operator, then value engineer; on the marketing-ops side, system administrator, then stack wrangler, then context engineer. Those names are adapted from the State of Martech report. The argument about what the change does to the people, and where the new work sits, is made in Where the New Marketing Jobs Come From on this site.
They sit inside the tools and draft, and a person finishes everything.
They run the work in one channel, and a person reviews it before it goes out.
They run work across channels inside agreed limits, while people direct the outcomes.
Full descriptions and examplesHide the full descriptions
Three positions
Help a person
Agents sit inside the tools and help a person with a task. Everything they produce is a draft that a person finishes.
Work under review
Agents run the execution inside a channel, and a person reviews the work before it goes out.
Run inside limits
Agents run work across channels within limits that were agreed in advance, and people spend their time directing outcomes rather than doing the steps.
Whatever the position, the top of the work stays with people: brand and portfolio strategy, creative judgment, and risk decisions. Agents take on more of the execution over time, but they do not reach that level, and the limits that hold them below it are set before any agent runs.
Falling behind
The cost lands on people either way. Held too low, every piece of agent work waits on a reviewer, and the reviewer becomes the bottleneck. Pushed higher than the decision-rights row supports, agents act on permissions nobody actually granted.
In the four examples
In the agents run inside limits one team set on its own, a step ahead of the decisions row. In they stay inside the tools.
Or start from one of four made-up examples
Each channel has its own data, its own budget, its own team and its own tools, and makes its own decisions. The line is flat and low. Each channel would tell you it is doing fine.
The shared data spine is finished and campaigns run through it, but decisions and budgets still sit in the channels where they always did. Some organizations end up here by accident. Others run this way on purpose, sharing the data and leaving the decisions local.
Data, decisions and execution have all been moved to one central team, and that team has written down which decisions it makes and which it leaves alone. Budget moves, but only through that team. The people left in the markets have less to do, and the agents run only where the center can watch them.
One team has agents running most of its work inside limits that team set for itself. The rest of the organization has not changed: the budget is still annual and the roles around that team are still channel roles.
These are invented organizations, drawn to show what a line looks like. None of them is typical and none is a target. Pick one, then move the marks to match your own.
The Marketing Maturity Diagnostic on this site measures capability maturity. This instrument measures where authority, money and cadence sit, which the diagnostic does not.
The six rows come from operating-model work with client marketing teams and from running a data practice with its own P&L. Rows 2 and 3 are judgment, not research, drawn from the meetings where the argument was about who decides and who pays rather than about the technology. The four examples are invented for this page. None is a client and none is typical.
Reviewed September 2026