Almost every agency versus in house decision is made on a number that is wrong before the conversation starts. Somebody puts a monthly agency fee next to a salary, notices the salary is smaller, and concludes that building it internally is the cheaper route.
A salary is not a cost. It is the largest line inside a cost, and in the UK it understates the real figure by at least a sixth before you have bought a single software licence. The agency vs in house marketing question is worth answering properly, but it can only be answered once both sides of the comparison are counted the same way: total cost of output against total cost of output.
This post costs both options honestly. What an in house creative team actually costs once statutory on costs land, what agencies charge and why, where the crossover point sits at different spend levels, the volume problem that almost nobody prices in, and the situations where building in house is genuinely the right call.
Key takeaways
- A UK salary understates the cost of an employee by at least 16% on statutory on costs alone: 15% employer National Insurance above the secondary threshold, plus a minimum 3% pension contribution.
- At the 2026 UK creative average of around £44,600, one hire is roughly £52,000 a year before software, hardware, recruitment fees, holiday cover or the management time to direct them.
- On payroll alone, a three person creative team is level with a percentage of spend agency fee somewhere between £65,000 and £130,000 of monthly ad spend, depending on the rate you are quoted.
- The cost comparison is the easy half. The harder constraint is volume, because a small in house team physically cannot produce enough distinct creative positions to keep an account covered.
- In house genuinely wins on brand heavy work, single product companies, deep product knowledge, and at very large scale where a percentage fee stops making sense.
What an in house creative team actually costs
Start with the salary, then add the things that are not optional. Employer Class 1 National Insurance is charged at 15% on earnings above the secondary threshold of £96 a week (GOV.UK). Workplace pension auto enrolment requires a minimum employer contribution of 3% (GOV.UK). Neither is negotiable and neither appears in the number people compare against a retainer.
Digital Waffle puts the projected average salary for UK creative and design roles in 2026 at around £44,600 (2026 UK Creative and Design Salary Guide). Run that through the two statutory lines and the picture changes immediately.
Salary: £44,600
Employer National Insurance at 15% above £96 a week: about £5,900
Pension at the 3% minimum: about £1,300
Fully loaded before anything else: about £51,800
That is a 16% uplift on the number most people budget with, and it is the floor rather than the answer. Still to come: software licences per seat, a machine capable of editing video, recruitment fees if you use a recruiter, holiday and sick cover, onboarding time before the person is productive, and the management time somebody senior has to spend deciding what the person should make. In our experience that last one is the most consistently underestimated cost in the whole exercise, because it is paid out of the time of whoever is already busiest.
One person is not a creative team
The other quiet assumption in most in house plans is that a single hire covers the function. Performance creative is at least three jobs, and the people who are excellent at one are rarely excellent at all three.
- Strategy. Deciding which psychological positions the account needs to argue from, reading performance data, and briefing against gaps rather than against taste.
- Production. Sourcing creators, running shoots or generations, and getting raw material in reliably enough to hit a weekly cadence.
- Post. Editing, motion, statics, versioning and the unglamorous work of turning one shoot into fifteen genuinely different assets.
Three fully loaded hires at the 2026 average is about £156,000 a year, or roughly £13,000 a month, before software or hardware and before the team has shipped anything. That is the number to compare against an agency fee, not one salary.
What agencies actually charge
Three models dominate. A percentage of ad spend, typically in the 10% to 20% band. A flat monthly retainer, commonly quoted between $2,000 and $15,000 or more depending on scope. Or a hybrid, a smaller base fee plus a percentage on top (Get Ryze, 2026). We went through the trade offs of each, from both sides of the table, in our post on what Meta ads agencies actually charge.
The relevant point for this comparison is what the fee buys that a payroll line does not. There is no recruitment lag, no notice period, no holiday cover problem, no fixed commitment when spend drops, and no single point of failure when one person leaves. You are also buying pattern recognition across a portfolio of accounts rather than the pattern recognition of one account, which matters more than it sounds when a vertical shifts.
Agency vs in house marketing: where the crossover sits
Set the three person team at £13,000 a month and solve for the ad spend at which a percentage fee matches it. At 20% the crossover is around £65,000 of monthly spend. At 15% it is around £87,000. At 10% it is around £130,000.
Read that carefully, because the direction surprises people. Below those spend levels the in house team is the more expensive option, not the cheaper one. Above them the percentage fee starts to look expensive relative to fixed payroll, which is exactly why large advertisers eventually bring capability in house or renegotiate onto a flat fee.
Three adjustments push the real crossover higher than the arithmetic suggests. Software, hardware and recruitment are excluded from the payroll figure. Management time is excluded and is not free. And the comparison assumes the three person team produces the same volume of genuinely distinct creative as an agency does, which is the assumption that fails most often.
The volume problem nobody prices in
Cost is the easy half of this decision. The harder half is coverage. Meta's current retrieval system narrows billions of eligible ads to roughly a thousand auction candidates in milliseconds, which rewards a library that argues genuinely different things and quietly penalises one that is twenty versions of the same argument. Coverage is not a volume target, it is a spread of psychological positions, and we set the full argument out in our guide to Meta ads creative fatigue for mobile apps.
Now put a clock on it. In our experience gaming creative fatigues in one to two weeks at scale, health and fitness in three to four, and finance or software in five to eight. A team that needs to keep two dozen distinct positions alive against a three week fatigue window is not doing occasional creative work, it is running a production line. Small in house teams tend to solve this by making more versions of whatever last worked, which reads as volume in a reporting deck and as one ad to the audience.
The tell is an account where frequency sits above 3.0 in a seven day window on a handful of ads while the rest of the library barely delivers. That is not a budget problem or a bidding problem. It is a coverage problem produced by a capacity ceiling, and hiring a fourth editor does not fix it if the strategy layer is the bottleneck.
The hybrid model, and why it usually wins
Most of the arrangements we see working are not either or. The in house team owns the things that require living inside the product: brand, product marketing, App Store assets, lifecycle, the launch work that needs to be right rather than merely tested. The agency owns performance creative volume and the coverage the account needs to keep buying installs cheaply.
That split works because the two jobs have different success conditions. Brand work is judged on consistency. Performance creative is judged on variance, and specifically on how much genuinely different territory the library covers. Asking one small team to optimise for both usually produces creative that is on brand and commercially inert.
The hybrid fails in one predictable way: when nobody owns creative direction. If the agency briefs against performance data and the in house team briefs against brand instinct with no agreed decision maker, you get two libraries, two opinions, and a quarter spent arguing about which one is being held back by the other.
When in house genuinely wins
We are an agency, so treat this section as the one worth reading twice. There are four situations where building internally is the better decision and we would say so on a first call.
Brand heavy categories where the creative is the product and consistency matters more than coverage. Single product companies with one audience and one message, where the number of distinct positions worth running is genuinely small. Products with enough technical depth that briefing an outsider costs more than doing it yourself, which is real in developer tools and some fintech. And large scale, where spend is high enough that a percentage fee buys more headcount than it delivers.
There is a fifth, less comfortable one. If the reason for the decision is that a previous agency underdelivered, changing the employment model will not fix it. The failure is usually in the creative process rather than in who was employing the people, and it will follow you in house. Our post on how to hire a performance creative agency covers what to check for so it does not happen twice, and what a performance creative agency actually is is worth reading first if the category itself is new to you.
How to run the comparison for your own account
Four steps, in order, and none of them start with a retainer quote.
- Count how many distinct creative positions your account needs live at once. Not assets, positions. Group your current library by the argument it makes and see how many groups you actually have.
- Divide by your fatigue window. That gives you a required monthly output of genuinely new work, which is the real capacity question.
- Price the headcount that produces it, fully loaded. Salary plus 16% statutory, plus software, hardware, recruitment and cover, plus the management time to direct it.
- Compare that against your quoted fee at your actual spend. Not against the industry average, and not against one salary.
Run those four and the answer is usually obvious, and it is not always us.
Frequently asked questions
Is an in house marketing team cheaper than an agency?
Usually not at the spend levels most mobile apps operate at, because the comparison is almost always made against salary rather than against the fully loaded cost of employment. Once employer National Insurance, pension contributions, software, equipment, recruitment and management time are added, a salary is roughly a sixth to a quarter understated. The honest comparison is total cost of output against total cost of output, not a retainer against a payslip.
What does an in house creative team actually cost in the UK?
Take the salary and add at least 16% before anything else. Employer Class 1 National Insurance runs at 15% on earnings above the secondary threshold and workplace pension auto enrolment requires a minimum 3% employer contribution. On a salary at the 2026 UK creative average of around £44,600 that is roughly £52,000 a year for one person, before software licences, hardware, recruitment fees, holiday cover and the management time somebody has to spend directing them.
How much do performance creative agencies charge?
The three common models are a percentage of ad spend, typically 10% to 20%, a flat monthly retainer, or a hybrid of a smaller base fee plus a percentage. Per asset pricing exists but tends to reward volume over judgement. The right question is not which model is cheapest in the abstract, it is which model keeps the agency's incentives pointed at the same outcome as yours at your spend level.
At what ad spend does in house become cheaper than an agency?
On payroll alone, a three person creative team costs roughly the same as a percentage of spend agency fee somewhere between £65,000 and £130,000 of monthly ad spend, depending on whether you are quoted 20% or 10%. That crossover moves higher once the costs that never appear on a salary line are included, and higher again if the in house team cannot produce enough distinct creative to keep the account covered.
Can you run an in house team and an agency at the same time?
That hybrid is the most common arrangement we see working, and it usually splits along a clear line. The in house team owns brand, product marketing and the assets that need deep product knowledge, while the agency owns performance creative volume and the psychological coverage the account needs to keep buying cheaply. It fails when both sides are asked to do the same job with no owner of the creative direction.
Working out which side of the line you are on
The agency vs in house marketing decision is not really a cost decision, even though it is always framed as one. It is a capacity decision with a cost attached. The question is how much genuinely distinct creative your account needs every month to stay covered, and then which structure produces that volume at a defensible price.
We are a performance creative agency for mobile apps, and the first thing we do on a new account is map the positions it is not covering and work out what output it would take to cover them. If you want that run against your library before you decide either way, apply to work with us. We take a small number of mobile app clients per quarter.