Quarterly marketing planning: a session that produces decisions

Aug 22
3 min read

Updated: 2 days ago

Introduction

Annual plans are obsolete by March and monthly plans are too short to change anything structural. A quarter is long enough for a real test to finish and short enough that a bad decision is not expensive.

The useful unit is therefore the quarter, and the useful output is not a document. It is three decisions.

1. Quarterly marketing planning needs four inputs prepared beforehand

The session fails when it starts with data gathering, because gathering turns into discussion and the decisions never arrive.

Bring four things already written down: spend by channel for the quarter, enquiries by source, conversion rate from enquiry to customer, and average transaction value. One page.

If any of the four is unavailable, that absence is the most important finding of the quarter and belongs at the top of the agenda.

2. Start by reviewing what you said last time

Open with the previous quarter's stated change and the number attached to it.

Did it move? By how much? Was the cause identifiable, or did three things change at once?

Skipping this is how businesses run the same initiative four times without noticing. It is also uncomfortable, which is precisely why it goes first rather than last.

3. Separate what happened from why

The most common failure in these sessions is jumping to explanation.

Enquiries fell 18%. That is what happened. Whether it fell because of seasonality, a competitor, a broken form or a paused campaign is a separate question, and it usually needs checking rather than debating.

Write the facts on one side and the candidate explanations on the other, then agree what would distinguish between them.

4. Decide what to stop

Stopping is the decision that never gets made, so it goes before the decision about what to start.

Any channel or activity that has run for two quarters without a measurable result is a candidate. So is any tool nobody has opened, and any report nobody has acted on.

Freeing time and budget is what makes the next quarter possible. Without a stop decision the plan is purely additive, and additive plans collapse under their own weight by week six.

5. Choose one change, with a number

Then the central decision: the single thing you are trying to move, expressed as a from and a to.

One change, not three. The constraint is not modesty — it is attribution. Three simultaneous changes produce a result nobody can explain, which means the quarter teaches you nothing that transfers to the next one.

State the number, the current value, and the target.

6. Agree the evidence in advance

Before anyone leaves, write down which number will be checked, where it comes from, who pulls it, and on what date.

Also write the stopping condition: the result that would make you abandon the change rather than extend it. Deciding that afterwards is impossible, because by then you are invested.

This single paragraph is what turns a plan into a test.

7. Keep the session to two hours and the output to one page

Longer sessions do not produce better decisions; they produce longer documents.

The output is: what we stopped, the one change with its number, the evidence and the date, and who owns each. Anything else discussed goes in a notes file nobody is obliged to read.

Circulate it the same day. A plan that arrives a week later has already lost the room.

8. Do not re-plan mid-quarter

Between sessions, the plan is fixed unless something is actually broken.

New ideas go on a list for the next session. This feels rigid and it is the mechanism that lets a test finish — most abandoned initiatives were not wrong, they were interrupted by a newer idea in week three.

The exception is genuine breakage: a form that stopped working, a channel that collapsed. Fix those immediately and note them as facts for the next review.

Conclusion

Prepare four numbers beforehand, review last quarter's stated change honestly, separate what happened from why, decide what to stop before what to start, then commit to one change with a number, an evidence plan and a stopping condition.

Two hours, one page, owners attached — and no re-planning until the next session, so the test actually gets the quarter it needs.