The strategy-execution gap: why most marketing plans die before they ship — and a 90-day fix

The size of the gap

The HBR study has one more finding that explains the rest. Nearly 90% of managers believed leaders communicated strategy often enough, and 84% could rely on their own boss and direct reports — but only 9% could rely on colleagues in other units all the time. Strategy does not fail at the top or in the team. It fails in the hand-offs. A solo operator has the same hand-offs, between Monday-you who plans and Thursday-you who has an inbox.

Five ways a marketing plan dies

Having watched a lot of plans die, including my own, the causes are boringly consistent.

1. It was a list, not a strategy

"Post three times a week, publish two articles, run a webinar" is a to-do list. A strategy says which of those matters most and why, so that when time runs out — and it always does — you know what to drop. A plan with no priority order dies the first busy week, because everything gets dropped equally.

2. It had no constraint

LocaliQ's 2026 survey found 60% of small businesses spend one to ten hours a week on marketing and 52% have a monthly budget under $1,000. A plan that does not start from those two numbers is fiction. Most plans are written as if the constraint were ambition.

3. Nobody approved it — including you

There is a difference between writing a plan and deciding to do it. The deciding step is where you say no to the other things. Skip it and the plan competes with everything else on equal terms, and loses to whatever is urgent.

4. Planned and shipped were never compared

If you do not write down what you intended and check it against what happened, the plan can fail silently for months. This is the mechanism behind the 67%. It is also the easiest one to fix.

5. The learning evaporated

The campaign ran, something happened, and three months later you cannot say what or why, so the next plan starts from zero. Evidence that is not kept is not evidence.

The one metric: planned versus shipped

If you track a single execution number, track this. Each week, write down the marketing actions you decided to take. The following week, count how many actually went out. Divide.

A healthy ratio for a solo operator is 70–85%. Below 50% and you are planning fiction — cut the plan, not the effort. Above 95% for several weeks and you are planning too little, or only the easy things. The number is uncomfortable, which is why almost nobody keeps it and why it works: it is the only marketing metric that measures you rather than the market.

AchieveIt's 2025 data makes the same point at organisational scale: 87% of leaders say real-time monitoring improves execution, and the organisations relying on manual data collection were the ones reporting slowdowns. For one person, "real-time monitoring" is a column in a spreadsheet you look at every Monday.

The 90-day cadence

This is the execution system I use. It needs no software beyond a calendar and a sheet, and it maps directly onto how the better growth platforms structure their loops, so if you later buy one you are not starting over.

Three hours to set up, about an hour a week to run. It is not sophisticated. The 67% failure rate is not caused by a lack of sophistication.

Where Next Best Action comes in

The hardest step in the cadence is number 3: deciding, each week, what the strongest move is given everything you now know. It is the step people skip because it is cognitively expensive and there is no forcing function. It is also the step software is now good at.

"Next Best Action" is a decision-science idea — it descends from the OODA loop and became a marketing term through customer-journey systems that pick the one best offer for each customer. Gartner expects that by 2027, half of business decisions will be augmented or automated by AI agents for decision intelligence. The 2026 growth platforms apply the idea to the business itself: given the goal, the constraint, the evidence and the results so far, what is the single strongest thing to do next, and why?

SEnuke AI's version — the reason it is worth mentioning here — attaches the reasoning to the recommendation, and puts an approval gate after it. That is step 3 of the cadence, packaged: the system proposes, you decide, and the decision is logged. Whether the proposals are good is what nobody outside the company can say before 8 September, and my review is explicit about that. But the shape is right, and the shape is what most plans are missing.

For teams and agencies: the hand-off problem

HBR's finding that only 9% of managers can rely on colleagues in other units is the agency problem in one number. Strategy lives with the account lead, execution with the content person, measurement with whoever has the analytics login, and the client sees none of it until the monthly report. The fix is the same cadence with two additions: the ranked initiative list is visible to everyone, and the approval step is a named person, not a meeting. Growth platforms that offer client-facing approval and white-label reporting — SEnuke AI's Agency plan does both — are essentially selling that visibility. You can also build it with a shared sheet and discipline; the platform is buying the discipline.

Frequently asked questions