Dividing a marketing budget in a way that survives the year

Most marketing budgets get set in two moves: a total, then twelve equal months across four or five channels.

Both moves are defaults rather than decisions, and both cost money.

Split by time horizon first, not by channel

The most useful division is by when the return arrives.

Immediate return. Paid search, remarketing, anything capturing existing demand. Money in, enquiries out, within days. Predictable and it does not create new demand.

Medium horizon. Content, search positions, email programmes, the website. Months to mature and then they compound.

Long horizon. Brand, reputation, relationships, the things that make everything else cheaper and that never attribute cleanly.

A reasonable starting shape for most established services businesses is roughly half to immediate capture, a third to medium, the remainder to long. A business that needs work this quarter weights the first more heavily. A business with a full pipeline and a growth plan weights the others.

The failure mode is putting everything in the first bucket, which works until you need to grow and discover there is no compounding asset underneath.

Then hold back a reserve

Ten to fifteen percent, uncommitted.

It funds the thing that started working and deserves more, the opportunity nobody forecast, and the month when something breaks.

A fully committed budget cannot respond to its own results, which makes the measurement pointless.

Shape the months to the demand

Even monthly spend is only correct in a business with even demand, and almost nobody has one.

Load the spend into the weeks before the buying season, allowing for the sales cycle. Reduce it in the periods when nobody is deciding. Use the quiet months for the work that has lead time rather than going dark.

Australian markets have a hard December to January stop in most sectors. Planning around it beats discovering it.

Protect the things that compound

When a month is tight, the cuts come from the medium and long horizon work, because it is the easiest to stop and the slowest to show damage.

Which is exactly why it should be the most protected. Stopping content and search work for a quarter costs about two quarters to recover, and the cost does not appear until later, by which time it is attributed to something else.

If cuts are necessary, cut the immediate-return spend first. It restarts instantly.

Ad spend and fees are different lines

Keep them separate, always.

Agency or internal cost is a fixed operating expense. Media spend is a variable that should move with performance. Blending them into one number makes it impossible to see which part is working, and it makes the inevitable conversation about scaling spend much harder.

Build in the cost of production

Budgets routinely fund the media and forget the creative.

Photography, video, design, writing, and the website changes needed to support a campaign. Running paid traffic to a page nobody had budget to fix is one of the more common ways a campaign underperforms for reasons unrelated to the campaign.

Review quarterly, not monthly

Monthly reallocation is reacting to noise. Most channels need a quarter to produce a reliable signal, and shifting budget every four weeks guarantees nothing ever gets long enough to work.

Quarterly, with the reserve available in between for genuine opportunities, is the balance that holds.

The number itself

If there is no basis for the total, work from what a customer is worth and how many more you want, rather than from a percentage of revenue.

A percentage of revenue is a comfortable number and it is disconnected from the objective. What you can afford to pay to acquire a customer, multiplied by how many you want, is a number you can defend.