What Is Process Intelligence in Marketing?

1. A definition worth being precise about

The phrase is short, so it is easy to stretch. A precise version:

What marketing work is happening. A product launch, a repositioning, a budget shift, a campaign that is underperforming and a quarterly plan look similar from a distance — documents, meetings, decisions. They are different kinds of work, with different prerequisites and different ways of going wrong. Naming the kind of work correctly is the first act of process intelligence, and it is often where teams disagree without noticing.

What should already exist for it. Every kind of marketing work rests on a few things: an agreed audience, a positioning, an objective, a way to measure, a constraint the brand has committed to. Which ones matter depends on the work. A budget shift without an agreed objective is a different problem from a creative refresh without one.

What is missing. A gap only matters in relation to the decision. The same missing measurement plan is irrelevant to a copy refresh and decisive for a budget review. Process intelligence weighs each gap against what is about to be decided, instead of counting empty fields.

What path is reliable now. Reliable is a property of the path, not of the plan. Given what the team knows today, some steps will hold whatever is learned next and some will have to be undone. The useful next step is one of the first kind.

2. What process intelligence is not

The term already means something in other parts of the business, so it helps to draw the borders.

It is not process mining. In operations and finance, "process intelligence" usually refers to process mining and its extensions: reconstructing how a process actually ran from the event logs of the systems it passes through, then finding its variants, delays and deviations. That is valuable, and it can map parts of marketing operations too — approvals, content production, launch checklists. What it cannot see is the substance those steps rest on: whether the positioning a campaign depends on was ever agreed, or whether the audience definition changed in a meeting nobody logged. Process intelligence in marketing reasons about that substance, not about the path the work took through the tools.

It is not project management. Status tracking answers "is it done, and is it on time?". Process intelligence answers "is it ready to be relied on?". A campaign can be on schedule in every tool the team uses and still rest on a positioning nobody signed off.

It is not analytics. Analytics tells you what happened: what the campaign delivered, which channel moved. Process intelligence asks whether the next decision is ready to be made on what the analytics shows — and whether the numbers being compared were measured the same way.

3. What it reasons over

To read where a piece of work stands, process intelligence needs more than the work itself. It brings four things together.

The objective. What the work is meant to achieve, in the terms the business agreed. Without it, "ready" has no meaning: ready for what?

Company Memory. What the company already knows and has decided — past choices and their reasons, evidence, assumptions, constraints. A plan that contradicts a decision taken two planning cycles ago is not a fresh idea; it is an unacknowledged reversal. Memory is what makes that visible.

Marketing expertise. How experienced marketers approach this kind of work: what a sound positioning choice depends on, what a launch plan usually lacks, what to check before moving budget between channels. Olymuse brings this into the work as expert process lenses — structured ways of examining a problem the way a seasoned practitioner would.

The evidence actually available. Not the data the team wishes it had, but what it can see today, how recent it is and how it was measured. A decision that needs comparable numbers cannot rest on two channels measured on different rulers.

Any reading worth having also has to account for the way the team works today. Two teams can face the same question with very different foundations: one has a shared measurement practice and a documented positioning, the other is building both while it works. The reliable next step is not the same for them, and advice that ignores the difference is right in general and useless in particular.

4. Four questions, answered differently each time

In practice, process intelligence comes down to four questions. The questions stay the same; what counts as a good answer depends on the kind of work and on what the company already knows:

1. What are we trying to do? Name the kind of work and the objective it serves.
2. What do we already know? Bring in the decisions, evidence and constraints that bear on it.
3. What is missing? Separate what this decision depends on from what would merely be nice to have.
4. What should happen next? Choose the step that makes the work more reliable from here — which is sometimes a decision, sometimes a test, and sometimes a conversation that should have happened earlier.

The value is in answering them before the decision, while changing course is still cheap.

5. What it returns: a reading, not a verdict

The output of process intelligence is not a score and not a green light. It is a reading of the work, with four parts:

• What is ready — the inputs that exist, are agreed and still hold.
• What is missing — named, with what depends on each gap.
• What would be premature — the moves that would commit the team before the work can support them.
• The next reliable step — with the reason it comes next.

And sometimes the honest reading is that there is not enough to go on. When the evidence does not support a recommendation, the right answer is to say so and to say what would change that, rather than to fill the gap with a confident guess.

6. A worked example

An illustrative example, not a customer story.

The product team wants to bring a launch forward by three weeks, and marketing is asked a simple question: can the campaign be ready?

It sounds like a scheduling question. Process intelligence starts by naming the work more precisely: this is a change to a launch, and a launch depends on an agreed positioning and audience, a message that matches the offer, the material sales will use, a way to measure the first weeks, and the brand commitments that apply.

Then it looks at what is known. Company Memory shows that the positioning and the audience were agreed at the last planning review, with the reasons written down. It also shows that the pricing message depends on a pricing decision that is still open, and that an earlier launch slipped because sales material arrived after the campaign had started.

The reading the team gets is concrete. Ready: positioning, audience and the core creative. Missing: the pricing message, because the decision it depends on has not been taken, and the sales material, which nobody has started. Premature: booking paid media against the earlier date, because those placements carry a price message that does not exist yet. Next reliable step: put the pricing decision on this week's agenda, and plan the earlier date in two stages — announce on owned channels on the new date, and start the paid push once the pricing message is agreed and sales has what it needs.

The head of marketing still makes the call, and might choose to hold the original date instead. Either way, the choice is made on a clear picture, and the reasons are recorded for the next time a date moves.

7. A test you can run this week

8. What this means for Olymuse

Marketing Process Intelligence is one of the five parts of Olymuse, the marketing operating layer that sits above the stack a team already runs. Because it behaves as a liquid intelligent layer, the reading starts from wherever the work is — a launch date, a draft plan, a question from the CEO — rather than from step one of a template. It brings the objective, Company Memory, expert process lenses and the available evidence into that moment, and returns the reading described above. When the evidence is not enough to say what is ready, it says so instead of guessing.

It does not decide, and it does not act on its own: the marketer stays the decision-maker, and any action moves only after a person approves it.

Frequently asked

Is process intelligence in marketing the same as process mining?

No. Process mining reconstructs how a process ran from the event logs of the systems it passes through, and it can map parts of marketing operations. Much of marketing's decisive work, though, happens in meetings, documents and conversations that leave no log. Process intelligence in marketing reasons about whether the things a decision depends on exist and have been agreed.

Does process intelligence tell the team what to do?

It tells the team where the work stands and which next step would make it more reliable, with the reasons. The decision stays with the people who own it — and when the evidence is not enough, the reading says so.

Is this just a checklist?

A checklist expects the same answers from every piece of work. Process intelligence keeps the same four questions but starts from the kind of work and the decision in front of the team, so what a launch needs to be ready is not what a brand refresh needs — and the answer depends on what the company already knows.

---

This guide goes one level deeper into an idea introduced in our second launch article: Marketing Is Not a Linear Workflow. That Changes What AI for Marketing Should Be. · Olymuse is in founder-assisted private alpha. Apply for the Private Alpha →