Somewhere between £1m and £10m turnover, most businesses hit the same fork. Marketing is happening, but nobody senior owns it. The founder is doing it badly in the gaps, or a capable junior is producing activity nobody has connected to a plan. So the question gets asked: hire a marketing director, appoint an agency, or go fractional?
It is usually framed as a shopping decision. It isn’t. It’s a diagnosis, and the four options solve different problems. Pick the wrong one and the mistake takes around nine months to surface, by which point you have lost a year of compounding.
We sell fractional marketing leadership, so treat what follows with appropriate suspicion. There is a section near the end on when we are the wrong answer.
Four questions, in this order
Answer these before you look at a single proposal. Together they point at a model more reliably than any comparison table.
- Do you know what to do, or do you need someone to decide? If you have a clear plan and need it built and run, you are buying execution. If you cannot say in one sentence which customers you are chasing and why they should pick you, you are buying judgement. Buying execution when you need judgement is the most expensive error at this size.
- Do you have anyone to execute? Someone to write the email, brief the designer, load the campaign, chase the list. Not necessarily senior. If the answer is nobody, a strategist gives you excellent plans and no output.
- Is the work continuous or a project? A rebrand, a website, a market entry, a bid for a big contract: projects with an end. Demand generation, content and brand maintenance are continuous. Continuous work wants an owner. Projects want a supplier.
- How much of your own time can you honestly give it? Agencies need briefing and decisions. Freelancers need direction. Even a permanent hire needs a year of your attention. If the truthful answer is two hours a month, that eliminates more options than anything else.
The four options, honestly
Permanent hire
Good for: continuous work, institutional knowledge, and businesses with enough marketing to fill a full week for years. Nobody will care about your business the way an employee does.
Cost shape: fixed, and larger than the salary. Employer National Insurance changed in April 2025 and is unchanged for 2026 to 2027: the secondary Class 1 rate is 15%, up from 13.8%, and the secondary threshold fell to £5,000 a year from £9,100, per HMRC’s rates and thresholds for employers. The Employment Allowance rose to £10,500 from £5,000, which softens it for smaller payrolls but is a fixed annual sum, not a per-head one. Add pension, equipment and recruitment fees: the all-in number is well north of the headline.
Speed to value: slowest. Three months to hire if you are quick, three more before they are useful, six to twelve before they are effective.
When it goes wrong: the part people underweight, and the risk profile is changing. Under the Employment Rights Act 2025, the government’s implementation timeline sets out that from 1 October 2026 the Employment Tribunal claim time limit rises from three months to six, and from January 2027 the unfair dismissal qualifying period falls to six months, alongside fire-and-rehire protections and uncapped compensatory awards. None of that makes hiring a bad idea. It does mean a 2026 decision carries a different shape of risk than a 2023 one. We are not employment lawyers and this is not legal advice; that conversation belongs with an employment solicitor before you write the offer.
Failure mode: hiring one person to be both strategist and doer. You get either a strategist frustrated by admin, or a capable executor promoted past their experience who quietly does no strategy.
Fractional or interim leader
Good for: businesses needing senior judgement, an owner for the plan, and someone to direct existing staff or suppliers, but without five days a week of director-level work.
Cost shape: fixed monthly, variable by choice, cancellable. UK advertised rates give a benchmark. McCracken Marketing publishes advisory engagements at £1,500 to £3,000 a month, embedded one day a week at £2,000 to £3,500, and two days a week at £4,000 to £6,000, aimed at businesses turning over £1m to £25m. Communications Edge advertises £3,000 to £7,000 a month or £750 to £1,250 a day, aimed at £500k to £20m turnover. Consistent enough to plan against.
Speed to value: fastest of the leadership options. Weeks, because you are buying someone who has done it before.
When it goes wrong: you give notice, usually one to three months, and it ends. That asymmetry is most of the value.
Failure mode: they become an expensive adviser producing decks that nobody actions, because there was no execution capacity underneath. See question two.
Agency
Good for: specialist depth, scale, and work you cannot sensibly employ for: production, media buying, technical SEO, design. A good agency has seen your problem across forty other clients.
Cost shape: variable and scalable, which suits volatile trading. You can turn it up and down as you cannot with a salary.
Speed to value: fast on output, slower on understanding. Producing within weeks; understanding your market in months, if ever.
When it goes wrong: you leave, and the knowledge leaves with them. Anything they built and hold, from ad accounts to creative files, should be in your name from day one. Check that before you sign, not at the exit.
Failure mode: the agency sets your strategy by default, because nobody on your side is senior enough to challenge them. That is not their fault; it is what happens when a supplier is hired to fill a leadership gap. They optimise what they were asked to optimise, and it can be the wrong thing for years.
Freelancer
Good for: defined execution against a plan that already exists. A copywriter, a paid social specialist, a designer. Cheap, quick, flexible.
Cost shape: purely variable, the lowest-commitment option. Speed to value: days.
When it goes wrong: you stop booking them. Almost no downside, which is why so many businesses drift into a freelancer-only model by accident.
Failure mode: three freelancers, no coordination, and the owner acting as unpaid account manager. Costs less in cash and more in your time than anything else, which brings you back to question four.
The hybrid that usually works at this size
For most £1m to £10m businesses the answer is not one of the four but a combination: a fractional or interim director setting direction and holding the plan, one competent executive or coordinator in-house doing the daily work, and specialists bought in by the project.
Director-level judgement is expensive per hour and you need few hours of it. Execution is cheaper and you need a lot of it. Buying both in one person means overpaying for the execution or underbuying the judgement. At £8m with a real growth agenda that maths flips: employ a director properly. Below roughly £2m you may not need a permanent leader at all, just a plan twice a year and someone to run it.
One aside, because it comes up every time. There is no credible UK primary source for what SMEs spend on marketing as a percentage of revenue; the figures circulating online trace back to US surveys, vendor blogs or nothing at all. Build the budget from what the plan requires.
Read the conditions before you commit
The employment cost picture bears on the fixed-versus-variable choice. The ONS Business Insights and Conditions Survey, Wave 157, fieldwork 18 to 31 May 2026 and 9,777 responses, found 66% of businesses with 10 or more employees reported staffing costs rising in May 2026. Asked how they would adapt to future rises in employment costs, 44% said they would raise prices, 38% would absorb it in margins and 23% would reduce headcount. The FSB Small Business Index for Q4 2025, published 13 January 2026 from 1,200 responses, put confidence at −71, the lowest since 2020, with taxation cited by 64% as a cost pressure, a record high, labour costs by 56%, and 26% having cut their workforce in the quarter.
Marketing budgets, though, are not collapsing. The IPA Bellwether Report for Q2 2026, published 16 July 2026, recorded a net balance of +6.9% revising budgets upwards: 23.8% increased, 16.9% cut, 59.4% held flat. The dominant behaviour is holding steady. Rising fixed employment costs against flat marketing budgets is exactly the environment in which variable and part-time models earn their keep.
The mistakes that break each model
- Permanent hire: a director with no budget and no team, then surprise when they leave inside a year.
- Fractional or interim: buying strategy with nobody to execute it, or buying two days a week and granting no decision rights, so they cannot own anything.
- Agency: appointing one without a clear brief and letting them define success. And not owning your own ad accounts, analytics and creative files.
- Freelancer: using several with nobody coordinating, so the brand fragments and the owner becomes the bottleneck.
- All four: changing model every twelve months. Marketing compounds. Three years of a consistent, mediocre approach beats three good approaches abandoned in turn.
When we are the wrong answer
Three cases where we are not the right call. If what you need is hands on keyboards, we are not a production team: a freelancer or a junior hire will serve you better at a fraction of the cost. If your growth depends on deep specialist channel expertise at scale, such as technical performance media at serious spend, you want an agency who do only that; we would end up hiring them on your behalf and adding a layer. And if you are pre-product-market-fit, hold off. If you cannot point to a repeatable group of customers who bought and stayed, no marketing leadership will fix that. It is a product problem, and spending on marketing first buys a faster route to the same answer.
Where we do fit is the middle case: something that works, a plan needed, and execution capability that exists but is unled. That is what Marketing and Campaigns is for. Where the question is really about direction and structure rather than marketing, Strategy and Change is the more honest starting point.
What to do next
Write down your answers to the four questions, one line each, then show them to someone who knows the business and will disagree with you. The right model is usually obvious once the four sit side by side. The hard part is being truthful about question four.
If you want a second opinion on which model fits, including if the answer is that you should hire permanently and we cannot help, get in touch. Half an hour is usually enough.