The first 90 days when you take over a marketing mess

The temptation is to cut the obvious right away. That is almost always a mistake. A campaign that looks absurd is sometimes the only one feeding another.

For two weeks, we take inventory. Which accounts exist, who owns them, who has access, which contracts are running, with what notice period, and who actually decides what today rather than on the org chart.

This work looks administrative. It reliably surfaces at least one thing nobody in the company knew: a forgotten account still spending, a contract on automatic renewal, or access held by a former supplier.

It is measurement.

Until the chain between the ad and the invoice is verified end to end, every number in the account is a well formatted opinion. We have seen accounts where half the reported conversions were map directions and page views. The number was climbing, leadership was reassured, and no lead had ever arrived.

So we check in this order: does the event fire, does it fire only once, does it correspond to something with value to the business, and can it be traced to a sale.

Judging performance before that step means arbitrating on false data. It looks decisive and it is expensive.

One page, every dollar of the marketing budget on it, and next to each line what it produces.

Three categories always appear.

What produces and can be proven. Usually smaller than the company believed.

What may produce, but whose measurement is broken. That is the biggest block, and that is where the work pays.

What produces nothing and nobody defends. Those lines get cut without a meeting.

At this point we cut what is indefensible, and we write the decisions down rather than discussing them.

Writing changes everything. A decision spoken in a meeting gets renegotiated the following month. A written decision, with its reasoning and its review criterion, holds. And if it turns out to be wrong, you know exactly why it was made, which is far more useful than remembering who proposed it.

This is also when we meet the suppliers in place. Not to replace them. To give them what they were missing: a counterpart who decides. A good agency with no arbiter drifts toward what is easy to deliver. With an arbiter, it becomes good again.

The final deliverable is not a report. It is a one-page plan and a decision structure.

For each budget line: what we expect, on what date we look, and what would make us stop. A line with no stop rule will be renewed forever, because nobody will ever have a formal reason to remove it.

We do not rebuild the site. We do not change agency. We do not launch a new channel.

Those three decisions are sometimes right, but none can be made properly before the numbers are reliable. Making them early replaces one mess with another, having spent a lot and destroyed the one thing that was actually missing: continuity.

It is not a rise in sales. Three months is often too short for that, especially on a long sales cycle.

It is that by the end, anyone in the company can answer the original question: where did our last ten customers come from, and what did each one cost.

Part-time marketing leadership, with a written scope and a mandate that lightens as your team takes over. See our marketing leadership offer

Should advertising stop during the diagnostic? Only what is expensive and indefensible. The rest continues, otherwise you lose three months of data at the exact moment you need it most.

Will our agencies take it badly? The good ones will not. A competent supplier prefers a client who decides to a client who never replies. The ones who take it badly identify themselves, and that is useful information.

What if the mess is internal rather than from suppliers? That is the most common case. The work is then less about cutting spend than about naming who decides what. It is more delicate, and it is often where the real gain sits.