What a good CMO does not do: five decisions that protect growth
After a decade as a CMO, my best decisions were things we did not do.
Nobody celebrates the campaign that never launched. But that list is most of the job. A senior marketing leader is paid as much for the spend they stop as for the spend they approve.
Here are the five no's I come back to most, and why each one protects growth.
1. We did not scale a winner until we knew why it won
A campaign that wins without a reason is a coin that landed heads. Scale it and you are betting the budget on luck repeating.
Before more money goes in, the team should be able to say what drove the result: the audience, the creative, the offer, or the timing. If nobody can, the next step is a test, not a budget increase. More on this in when to scale your ad budget.
2. We did not keep a channel alive because it had been alive for two years
Age is not evidence. A channel that has been in the plan for two years has defenders, a slide and a history, which is exactly why it rarely gets tested.
Every channel should earn its budget again each quarter. The ones with the longest history deserve the hardest look, because they are the least likely to have been questioned.
3. We did not trust a dashboard that could not survive a holdout test
Attribution tells you which channel touched a sale last. It does not tell you whether the sale would have happened anyway. The only way to know is to pause the channel in some regions and see whether total revenue moves.
If a channel's dashboard looks great and it has never been through a holdout, treat the dashboard as a claim, not a fact. See how a holdout test cut a channel and raised profit 42%.
4. We did not hire to fix a problem nobody had named yet
"We need a growth hire" is often a way to avoid naming the actual problem. If the constraint is retention, a performance marketer will not fix it. If it is positioning, a new channel specialist will spend money proving that.
Name the constraint first, in one sentence. Then decide whether it needs a person, a project or a decision. The order is in how to run a growth audit.
5. We did not approve a plan the team could not say out loud without the deck
If the plan only makes sense with forty slides behind it, the team will not execute it the same way twice. A plan people can say in two sentences is a plan they can act on, defend and change when the data says so.
How to use this list
- Write your own five no's for the next quarter, before the planning meeting.
- Attach a reason to each one. A no without a reason turns into a yes the first time someone pushes.
- Put them in the plan document, next to the things you will do.
- Review them at the end of the quarter. Which no's held, and which ones should change?
Bottom line
Say no with a reason. Write it down. Repeat.
FAQ
What does a fractional CMO actually do?
A fractional CMO sets marketing strategy, owns the growth number and leads the team or vendors part time. A large part of the role is deciding what not to do: which channels to cut, which winners to test before scaling, and which hires to delay until the problem is clear.
How do you decide which marketing channel to cut?
Test whether it is incremental. Pause it in matched regions and compare total revenue with regions where it kept running. If revenue barely moves, the channel is not adding much, however good its dashboard looks.
When should a startup hire its first marketing person?
After the main growth constraint has a name. Hiring before that often brings in someone whose skills fit a different problem. A short audit or a fractional CMO can name the constraint first.
Why is saying no important in marketing leadership?
Budget and team attention are limited. Every campaign, channel or project that continues without a reason takes resources from the few things that actually move growth.