SME marketing plans often have one of two flaws. Either it is a list of actions for the year, with no direction, that gets dropped by April. Or it is a beautifully presented five-year vision that nobody knows how to turn into tasks for next month. An SME marketing plan that holds up has to do both, at different levels of detail.
When I build a marketing plan, I always go about it the same way: a diagnosis, a few priority workstreams, then a plan at six months, one year, three years and five years, with a monthly committee to keep it on track. Here I describe that method, what goes into each horizon, and the tools I first built for my content machines that help me keep the plan alive. I am not claiming it is the only good way to do it.
01Why an SME marketing plan needs several horizons
In France, an SME as defined by Insee (the national statistics office) employs fewer than 250 people and turns over no more than 50 million euros. In practice, marketing in these companies often rests on one or two people, sometimes on the managing director. So the plan has to be short, and it has to help you say no.
There is a second, deeper reason. Marketing actions do not all work at the same pace. In a short video, Les Binet explains, drawing on his research with the IPA, that sales activation produces strong effects that fade quickly, while brand building produces slower effects that accumulate. He cites a balance often observed at around 60% of budget for brand and 40% for activation, and adds that this balance shifts with category and context.
I would not copy that ratio as it stands into an SME. But the underlying idea is right: if the plan only looks at the next six months, it never funds what pays off in three years.
Search is a good illustration. Google says so itself: some changes take effect in a few hours, others can take several months. An SEO workstream judged after six weeks will almost always be abandoned too early.
02Start with a short diagnosis
I never start by writing the plan. I start by looking at what exists, for a few days, not a few months. The diagnosis does not need to be exhaustive. It needs to let you choose.
What I look at
- The figures you have: where enquiries and sales come from, what Search Console and Analytics show, what each paid channel costs.
- What customers say: the questions that come up in meetings, the reasons people buy or walk away, when the sales team writes them down.
- What is already running: the website, content, campaigns, tools, and who actually looks after them.
- What is blocking: often something very concrete, like a website nobody can update without an outside supplier.
- The company's framework: its sales objectives, its budget, what the business plan already says.
On that last point, the marketing plan does not float on its own. The UK government's guide to writing a business plan sums it up well: the document covers objectives, strategies, sales, marketing and financial forecasts. The marketing plan details one part of it. If it contradicts the sales objectives, the marketing plan is the one that is wrong.
03Pick a few workstreams, not twenty
The diagnosis always produces too many ideas. The hardest part of the job is keeping only a handful. I prefer to stop at five or six workstreams: a small team cannot carry many more at once.
A workstream, for me, is an expected result, not an activity. "Post on LinkedIn" is not a workstream. "Get qualified meeting requests from LinkedIn" is. Each workstream has an owner, a metric and a first deadline.
To sort them, I ask three questions of every idea. Does it move us towards a written sales objective? Do we have the means to do it properly, not by halves? Does it unblock something else? The ones that unblock go first. On one of my content production projects, the bottleneck was not the writing but the publishing, done by hand. Automating publication before producing more was the right call, even though the initial urge was to write more.
The ideas you set aside do not vanish. I keep them on a "not now" list, with the reason. It stops the same debates from reopening every month.
04Six months, one year, three years, five years: what each horizon holds
The principle is simple: the closer the horizon, the more detail. The further away, the shorter it gets.
Six months: what we do
This is the only truly operational horizon. Each workstream is broken down into actions, with a person and a date. It holds the committed budget, the suppliers, and the metrics you will look at every month. If an action has no name next to it, it is not in the plan.
One year: what we aim for
The year's objectives, with numbers where possible, and the overall budget per workstream. No detailed actions beyond six months: you do not yet know what you will have learned.
Three years: where we want to be
A few sentences. The markets or audiences you want to have opened up, the position you want to hold, the foundations you need in place, such as a website you control or a body of content that brings in traffic without advertising.
Five years: the direction
One or two sentences, consistent with the company's strategy. Their main job is to check that the six-month workstreams are not heading the opposite way.
An example of the format, for an SME selling B2B:
- 6 months: the workstreams, their actions, names, dates and committed budget, reviewed monthly.
- 1 year: objectives with numbers and budget per workstream, reviewed every six months.
- 3 years: audiences, positioning and foundations, reviewed once a year.
- 5 years: the direction in two sentences, reviewed once a year.
05Keeping the plan going once it is written
A plan filed away in a folder is useless. What keeps it alive is a review rhythm and one single place where you can see the state of each workstream. Here I borrow two rules from my content machines, the ones that write and publish the articles on this blog and for Moonworkers, my own company.
A versioned calendar as the single source of truth
In my content machines, every topic has a line in a file, with a status: to do, in progress, published, on hold. "On hold" means a human decision is needed, and the machine does not touch it. I run the workstreams of a marketing plan exactly the same way. One line per workstream, a status, a name. Whatever is "on hold" becomes the agenda for the next meeting with the leadership team. I describe this logic in more detail in my note on the content machine with human review.
A monthly committee fed by a report that sends itself
The plan is reviewed once a month, in a committee with the leadership team. The figures arrive before the meeting, through an automatically sent report, so the session is spent deciding rather than discovering charts. I set out the agenda I suggest in my note on the monthly marketing committee, and how the report is built in the one on automated SEO reporting.
On metrics, I keep a piece of advice that even HubSpot gives, a software vendor with every reason to want you measuring a lot: look at real outcomes rather than vanity metrics such as impressions. An enquiry, a sign-up or a sale rather than a view count.
06What a five-year plan will not do
I would rather be clear about the limits. A five-year plan predicts nothing. In an SME, the offer, the team and the market often change well before then. The long horizons are there to keep a direction, not to commit budget.
The method also has a cost. The diagnosis and the writing take time, and the monthly committee takes more every month. For a micro-business with two or three marketing actions under way, it is probably too much: one page with the half-year's priorities and a quarterly check-in will do.
Finally, a plan does not replace delivery. A carefully written plan fails fast if nobody has the time to carry it. If the diagnosis shows nobody is there to do the work, that is the first workstream, ahead of all the others.
07Where to start this week
If you are starting from scratch, I would begin with three simple things. Take half a day to list what is already running and where your customers really come from. Write the five or six workstreams that seem most important on a single page, with a name next to each. And set the date of the first monthly committee with the leadership team, before the plan is even finished.
The rest gets refined as you go. A six-month plan reviewed every month is worth more than a perfect five-year plan nobody rereads.
This is the kind of work I do on short engagements, from diagnosis to plan to getting the committee started. If you would like to talk it through, my strategy and project direction services describe how I work.