Marketing plan 2027: the template I'd actually use

Search for "marketing plan" and you get a thousand templates. Most follow the same pattern: a cover page, a SWOT, a structure of vision, goals and strategies, and then forty pages nobody returns to after September. I know, because I've written a few myself. The problem isn't that the templates are bad. The problem is that they're built to be presented, not to be used.

A marketing plan that's meant to be used looks different. It fits on one page. It forces priorities instead of collecting them. And every person working from it can recite it from memory. This is the template I would use heading into 2027 — and the order of the fields is deliberate, each one harder than the last.

What a marketing plan is actually for

Before the template: what is the plan for? I see three jobs. It must force a decision about where time and money go. It must make it possible to say no — to that campaign, that partnership, that trend. And it must make sure five different people understand the same thing when they say "prioritise the brand". If the plan can't do those three things, it's a document, not a plan.

The template: seven fields on one page

The goal. One number, one year. Not twelve KPIs in a dashboard, but a goal that forces a real choice. "Double the number of inbound demo bookings" sharpens the planning more than "raise awareness".

The situation. Where is demand, and when? Remember the 95:5 rule: at any given moment only around five per cent of your market is actively in a buying window. Describe what you do for the five — and for the ninety-five who should recognise you next year.

The positioning. One sentence: for whom, instead of what, and why should anyone care. If you can't write it without channel words — "visibility", "being where the customer is" — you haven't done the positioning work yet. You've just picked a media plan.

The channel mix. The split between brand building and activation. Binet and Field's 60/40 is a good starting point, not a law; for B2B closer to 46/54. I've built a simple budget template for exactly this step — the Channel Mix Budget Template on this site.

The seasonality. When do the category's buying windows open, and what do we run before them? Plan by the customer's calendar, not the company's. The fiscal year and the buying year rarely share dates.

The measurement. Which two or three numbers do we look at every month? Split them into leading indicators — branded search, direct visits, subscribers — and lagging ones like revenue and pipeline. Leading measures tend to fall out of plans, though they're the only ones you can act on in time.

The stop list. What we are deliberately not doing this year. This is the template's most important field, because it's the only one that makes the plan cost something. A marketing plan without a stop list is a wish list.

A filled-in example

Take a small B2B SaaS company, say in invoicing. Goal: double inbound demo bookings by the start of 2027. Situation: buying windows cluster around the start of the financial year and around system changes; the rest of the year is about being remembered. Positioning: for smaller companies that have outgrown their spreadsheets — instead of either hiring a finance person or buying an all-in-one suite, the easier way into the everyday. Channel mix: 46/54, with the brand share going into a recurring content series and industry partnerships, and activation into search and a well-tended partner programme. Seasonality: heavy before January and the autumn close, light in summer. Measurement: branded search and demo bookings every month. The stop list: no trade fairs, no new website, no trend-jacking. Three noes are three priorities.

What's become more important for 2027

One part of the plan that older templates never had: what happens when the customer never clicks? Zero-click searches already sit around 60 per cent of all Google searches, and a growing share of research happens in AI chats and agents that never see your page. So set a goal for that channel too — to be the source the AI answer cites in your category. You can't buy that slot and the measurement is still immature, but you can work for it, and in 2027 the winner is the one who's in the answer.

And the measurement in field six needs rebuilding for that: if the customer researches in a chat and buys through a direct visit, your attribution model will call it brand. That's one of the few times the model is right.

Write it once, change it often

A marketing plan isn't a novel. It shouldn't be finished, it should be current. My habit: write it in an afternoon, live with it for a week, then change the stop list more often than the goal. The goal is rarely worth touching. The stop list always is.

A final test: can you recite the plan without looking — the goal, the positioning, one of the stops? Then it's a plan. Otherwise it's still a document.