¶01 The month everything stopped
Nobody buys marketing help because marketing is going badly. They buy it because delivery is going well, and delivery going well is what stopped the marketing.
The shape is always the same. A go-live moves. A rescue project lands on the same fortnight as a data migration. Two consultants are on site for three weeks and the third is covering their support queue. The work is good, the client is happy, the month is profitable. And the article that was half-written in week one is still half-written in week nine, the newsletter has not gone out since spring, and the follow-up you meant to send to eleven quiet prospects is still a note in someone's phone.
Nothing went wrong. That is the part firms find hardest to accept. A busy month is not a failure of discipline, and treating it as one leads to a plan that assumes a discipline nobody has spare.
¶02 Why it is always marketing that gives
In a specialist consultancy, marketing is nearly always assigned to people whose billable work has first call on their week: a director, a delivery lead, sometimes the founder. It is the one item on their list with no client waiting at the other end of it.
So when the week is over-committed, marketing is not chosen for the chop. It is simply the only thing that can be dropped that afternoon without a phone call. Everything else has somebody chasing it.
That makes the failure structural rather than personal. If a task only happens in the gaps, and the gaps close whenever the firm is doing well, then the task fails precisely when the firm can most afford to invest in it. Any plan that does not move production away from the people delivering the work will fail the same way next quarter, however good the intentions were in January.
¶03 What stop-start actually costs
The obvious cost is the missing output. It is not the expensive one.
The expensive one is the restart. When marketing resumes after a three-month gap, whoever picks it up spends the first month re-learning what the last one knew: which objections come up on calls now, which service is actually selling, what the offer wording is this quarter, which half-finished draft is worth keeping. That relearning is real work and nobody budgets for it, so it comes out of the first month's output. Three months off routinely costs four.
The second cost is compounding, or the absence of it. Search visibility, an email list that still recognises your name and a body of published work that answers real buyer questions all improve slowly and only while they are being fed. A firm that publishes six good pieces in six months is in a materially different position from one that publishes six good pieces in one month and then nothing, even though the output is identical.
The third is the one that shows up in the pipeline. Referral-led firms tend to discover a quiet quarter about two quarters after the marketing stopped, which is late enough that the two events are rarely connected. The busy month feels like the good news. The quiet quarter feels like bad luck.
¶04 The rhythm that survives
A rhythm that survives a bad month has three properties, and none of them is about working harder.
It is small enough to finish. One theme a month, not four. A theme that produces a handful of connected pieces from a single research effort costs a fraction of four unrelated pieces, because the expensive part of the work is the thinking, and the thinking is done once.
It is produced away from the delivery team. The firm supplies context and judgement, which are the two things it cannot outsource, and reviews the result. It does not supply the production hours, because those are the hours that vanish first.
It has one approval round, not five. An approval loop that requires three people to agree is an approval loop that stalls in a busy fortnight. One named approver, one round, a clear yes or a clear change.
None of this is a lower standard. It is the same standard, scoped to a month that can go wrong, which is most months in a consultancy.
¶05 One priority, not eleven
Most consultancy marketing plans fail at the point where they list eleven things. Eleven things is not a plan, it is a wish, and the first busy fortnight kills all eleven at once because none of them was ever the one that mattered.
One agreed priority a month is a harder document to write and a much easier one to keep. It forces the question most plans avoid: if only one thing ships this month, which one changes anything? Usually the answer is not the newsletter. It is the page that answers the question three prospects asked on calls last month, or the case study that has been waiting on a client's permission since February.
The test is simple. If the month goes badly and only one item ships, would you have picked that one? If not, the order is wrong before the month even starts.
¶06 What to ask a supplier to commit to
If you are handing this to someone outside the firm, the commitments worth having are the ones that survive your bad month rather than assuming it away.
Ask what happens if a month is not delivered. Ask whether anything can be published without your approval. Ask what happens if you want to stop at month three, and whether that is a conversation or a written commitment with a list attached. Ask who does the production when your team is on site for three weeks, and get a name rather than a reassurance.
For what it is worth, those are the questions Artifexa's own terms are built to answer: month one is pay on delivery, so an undelivered month is not invoiced; nothing is published without an approval round; and there is a review at month three against a written list of what should exist by then, where anything missing means you stop and months four to six are never invoiced. The full terms sit on the engagement menu and the delivery model is on how we deliver.
Whoever you use, hold them to that shape. A supplier whose plan only works in your quiet months has not understood the business they are selling to.