deciding what to stop.

marketing activity accumulates in one direction. a channel gets added, a format is tried, a report is introduced, a platform is opened — and almost none of it is ever formally stopped. the result is a programme with fourteen active commitments and enough capacity for five.

the symptom is familiar: everything is being done, nothing is being done well, and nobody can say which activity is producing the results.

the reason this persists is that stopping something requires a decision with a name attached, while adding something only requires enthusiasm. one is socially expensive and the other is free.

— 01write down everything, with its real cost.

start with an honest inventory: every channel, format, report, meeting and recurring commitment, with the hours it consumes per month. include the hidden costs — approvals, coordination, the meeting about the report, the time spent deciding what to post.

this exercise alone usually settles the question. a team of two discovers it has committed to eleven recurring activities requiring roughly double its available time, at which point the conversation stops being about ambition and starts being about arithmetic. run the inventory with the people who do the work rather than the people who commissioned it, since the hours are only known accurately in one of those places.

be specific about who does each thing. an activity that depends entirely on one busy person is a fragile commitment regardless of its value, and fragility is a legitimate reason to cut something that works. the alternative is discovering it when that person takes two weeks off.

and count the things that produce no output at all — the weekly status call, the dashboard nobody reads, the deck that summarises the report. those are pure cost and they are the easiest things to remove. a recurring meeting that exists to discuss a report nobody reads is two commitments masquerading as one.

— 02cut by capacity before performance.

the instinct is to cut what performs worst. the better first pass is to cut down to what you can do properly, because a well-executed version of a mediocre channel usually beats a badly executed version of a good one.

so decide how many recurring commitments the team can genuinely sustain — usually fewer than four for a small team — then choose which ones. the constraint does the work that debate cannot. it also depersonalises the decision, which matters when the activity being cut belongs to someone in the room.

and be aware of the sunk-cost pull. an activity that took six months to set up, or that someone built personally, is defended for reasons unrelated to its results. the question is not what it cost to build, it is whether you would start it today knowing what you know. asking the question in exactly that form removes most of the emotion from the answer.

— 03the three questions that decide it.

for each item: does it produce something you can point at, would you notice within a month if it stopped, and is it the only thing serving that purpose. an activity failing all three should have been stopped already.

the second question is the sharpest. a large amount of marketing activity would be invisible in its absence, which is a clear verdict. if nobody internally and no customer would notice, it is not producing anything. the honest test is to pause it for a month rather than debating it — an unnoticed pause answers the question definitively.

the third protects against over-cutting. some low-performing activity is the only thing covering an important function — the only content reaching existing customers, the only channel producing referrals — and those deserve fixing rather than removal. the distinction is whether the purpose matters or the activity does — if the purpose matters, replace it with something better rather than cutting it and leaving a gap.

— 04stop properly, and put it in writing.

a stopped activity needs a decision with a date and a name, or it becomes a thing everyone assumes someone else is still doing. announce it internally, remove it from the calendar, and say what the freed capacity is going to instead.

that last part matters. capacity released without a destination gets absorbed invisibly, and in three months the team is equally busy with nothing to show for the cut, and the cut gets remembered as pointless. name where the hours go.

and schedule the review. quarterly is right for a small programme — half an hour with the inventory, asking the three questions again. it is the least popular meeting in marketing and the one that keeps a programme sane, because it is the only mechanism that removes anything, which is why it is part of how we run strategy engagements and why the reporting is deliberately short enough to read.