Direct answer
Three things kill a marketing strategy at month three, and none is the strategy being wrong. The effects that build revenue are slow — Binet and Field's analysis of roughly 1,000 IPA cases found the fast, sales-activating effects fade quickly while durable effects build over many months. The strategy was recommendations, never decisions with an owner. And a disagreement at the top got settled by volume, not a test. Set the judgement date when you decide, not in the panic.
The pattern is so regular you could set a calendar by it. A new direction is agreed in January. February is busy and hopeful. In March the first numbers come in and they're neither good nor bad, which is what three months of anything looks like. Someone on the board asks a question. An agency pitch arrives with a fresh idea. By April the strategy is being "adjusted," which is the word companies use for abandoning something without having to say so.
Most people blame the strategy. Occasionally that's fair. Usually it isn't. The strategy was never given the one thing it needed, which was to be protected long enough to work.
What month three actually looks like
Marketing produces two kinds of effect, and they run on different clocks. Binet and Field's analysis of roughly 1,000 campaigns in the IPA Databank — published as The Long and the Short of It in 2013 — found that activation works fast and fades fast, while the effects that build durable revenue are slow, emotional and cumulative, and take many months to become visible in the numbers. At month three you are standing in the worst possible place to judge: the sharp early uplift from the launch has decayed, and the slow build hasn't yet cleared the noise. The dashboard shows a dip. The dip is the strategy working the way the evidence says it should. It looks like failure to anyone who wasn't told in advance what month three would look like.
That is the first killer, and it is the innocent one. The other two are not.
The three killers
| Killer | How it shows up in month three | What it costs | What prevents it |
|---|---|---|---|
| The clock | Activation faded, brand build not yet visible; the dashboard dips | A working strategy killed at its low point | Say in January what March will look like, and write it down |
| No owner | The strategy was recommendations; nobody's job is to hold it | Drift back to habit, one department at a time | Turn each recommendation into a decision with a name and a date |
| Volume | A leadership disagreement, settled by whoever is most senior or loudest | The decision replaced by a mood | Settle disagreement with a test, not a raised voice |
The second killer is the most common. A strategy that arrived as a deck — analysis, personas, a channel plan — contains no decisions anyone is accountable for holding. So when the sales director reverts to the old pitch in week nine because a prospect liked it, nobody notices, because nobody owned the new one. The strategy isn't rejected. It evaporates. Each department drifts back to what it did before, and by month three the only place the strategy still exists is the PDF.
The third killer is the one leadership teams don't like to hear named. Somebody senior disagrees — with the positioning, the audience, the message — and the disagreement is settled the way most disagreements in most companies are settled: by seniority and volume. I learned the alternative early, inside a global consumer-goods group, watching a marketing director evaluate an agency. He asked what they would do if he disagreed with work they believed in completely. They said they'd stand their ground and, if necessary, make the case louder. After they left he asked me whether I thought he wanted to work with someone whose answer to disagreement was volume. The intelligent answer, he said, would have been: let's set up the research to validate the position, or kill it. A disagreement about something testable should be tested. In month three, almost nothing is.
A strategy that can be killed by whoever is loudest in month three was never a decision. It was a mood.
From the book: an executive programme had thirty seats at €30,000 each and had never sold out. The decision we changed was upstream — who the programme was really for and what it actually sold — and a programme's intake runs on an annual cycle, so the seats get counted once. Judged at month three, that decision would have shown the same seat count as the year before and a dashboard nobody could read. That season all thirty seats were filled, with a waiting list at close. I can't separate the decision's share of that from everything else that moved in the market that year, and that isn't the point here. The point is the clock: nothing about the decision changed between month three and month twelve. Only the evidence arrived.
How to make a strategy survive its first quarter
Every strategy rests on four decisions — who this is for, why they'd pick you over the alternative, what exactly you sell them, and what you say to them. Call it the upstream audit. Make those four in writing, and then, on the same day, make three more: who owns each one, on what date it will be judged, and what evidence will be allowed to judge it. Name the leading indicators — which conversations, with which buyers, saying what — that should move before revenue does, so that month three has something honest to look at. And write down what "not working" would look like, in advance, so that a dip nobody predicted can be told apart from a dip everyone did.
Then hold the line. Not blindly — a strategy with a real judgement date can still be killed on that date, by the evidence you agreed to. But not by a board question in March, not by an agency with a fresh idea, and not by the most senior person in the room changing their mind over lunch. If that last one is the risk in your company, the disagreement is the thing to fix, and it's fixed by a test.
The reason this matters compounds. As I set out in why your campaigns never add up to growth, every abandoned direction resets the clock to zero, and the company pays for month one again. Why one senior adviser and a full agency team hold a decision differently is the subject of one advisor versus a full agency team. The complete argument — decisions made earlier and protected long enough to work — is Reports, Not Revenue.
- When you decide the strategy, decide its judgement date and its evidence on the same day. Write both down before anyone is tempted.
- Say now what month three will look like. If the answer is "a dip," the dip is not news when it arrives.
- Turn every recommendation into a decision with an owner. What has no owner will evaporate by week nine.
- Agree who may kill the strategy and on what evidence. Seniority and volume are not on the list.
- When a senior disagreement surfaces, propose the test that would settle it. If nobody can name one, the disagreement is about taste, and taste doesn't get a vote in March.
