Agency vs in-house marketing: who owns the strategy

Agency vs in-house marketing is the choice of who executes a company’s marketing. An agency is an outside firm serving several clients in parallel with ready-made disciplines, from paid media to design. An in-house team is made up of your own employees, who work only for your company and learn its product and buyers every day.

TL;DR. Before choosing who executes, a company must know who decides what the market should believe, which evidence backs the promise and in what order the budget is spent. Execution can be bought or hired, while strategy and the memory of past decisions stay inside. Without an owner for those, both options inherit the same confusion under a different invoice.

We start from a belief we apply to every company we work with. Value is lost most often in the gap between what a company actually does and what the market understands about it, and nobody who takes direction from someone else can close that gap on their own.

So this guide places the decision on the layer where it is really made, before any comparison of vendors, rates or team charts.

What is the difference between in-house and agency marketing?

The practical difference is dedication versus reach. An in-house team works only for you, knows the product from the inside and talks to sales every day. An agency brings several disciplines at once, capacity you can scale up or down and exposure to other markets, while splitting its attention across clients and learning your company from the briefs it receives.

Large advertisers in the United States have largely settled the dilemma already. The Association of National Advertisers found in its study “The Continued Rise of the In-House Agency”, published in May 2023, that 82% of its members now run an in-house agency, up from 78% in 2018, 58% in 2013 and 42% in 2008.

The same study shows that 92% of those respondents still work with outside agencies as well. On average, 61% of their work was done in-house in 2023, against 58% in 2018, and the rest went out for two main reasons, lack of bandwidth inside or capabilities the company simply does not have.

The lesson for a founder is less dramatic than the debate suggests. Mature marketers do not choose between the two models, they combine them and move work back and forth according to capacity. The data describes execution only, and the decision about direction does not move with capacity, because it determines what both sides end up executing.

Respondents named cost efficiency as the single primary benefit of an in-house agency, by a wide margin. Better knowledge of the brands, institutional knowledge and dedicated staff followed. Institutional knowledge matters most for this argument, because it is precisely what leaks away every time execution changes hands.

What does an in-house agency actually cover?

ANA defines an in-house agency as a department, group or person carrying out responsibilities that an external advertising or marketing communications agency would typically perform. The definition excludes internal public relations and says nothing about strategy, which confirms that the whole comparison is about who executes.

That detail is easy to miss, and it shapes everything that follows. Whether you call it an in-house agency, an internal creative team or a marketing department, the label describes production capacity. It does not tell you who decides what that capacity should produce, or why.

For a company with more than 3 million euros in revenue, the in-house version often starts as one person. That person inherits a website, a social calendar and a budget, and does not always inherit a written positioning, because in a founder-led company the positioning may still live only in the founder’s head.

What does “strategy” actually mean in this comparison?

Marketing strategy is the decision about the place you want in the buyer’s mind and the evidence that earns it, made before any campaign runs. Execution is everything you do to carry that decision into the market, meaning the site, content, ads, social media and email. Mixing the two up is why “agency or in-house” feels harder than it is.

Michael Porter drew the line exactly where most founders get lost, in the article that defined the discipline, “What Is Strategy?”, published in Harvard Business Review in 1996.

“The essence of strategy is choosing what not to do.” (Michael E. Porter, Harvard Business Review, 1996)

In the same passage Porter explains that without trade-offs there would be no need for choice, any good idea would be quickly imitated, and performance would depend wholly on operational effectiveness. Applied to our topic, the sentence has a direct consequence for how you hire.

Neither an agency nor an in-house hire can decide on your behalf what the company refuses to do. They can inform the decision, bring market data and propose better options. The call on what to leave out, so that there is room for what matters, belongs to the founder or to the executive the company depends on.

Our test for every positioning statement separates the two quickly. If a competitor could sign the same sentence word for word, the sentence describes the category and leaves the company unnamed. Execution cannot pass that test by definition, since any good operator can reproduce a campaign, a reel format or a landing page once they have seen it.

So the useful question is less about the price of execution and more about the person who keeps strategy stable while execution changes around it, from one vendor to the next, from one hire to the next and from one channel to the next.

Which layer of the company does the decision really sit on?

The decision sits one layer above the one where it gets discussed. Our methodology reads a company on four stacked layers, from L1, performance marketing, to L2, revenue and the commercial process, to L3, orchestration and memory, where the AI Brain lives, and up to L4, positioning and category architecture.

The agency versus in-house choice concerns almost only L1, which is also the most visible layer of the company.

L1, performance marketing. This is where cost per lead, acquisition cost, return on ad spend and conversion rate live. Nearly every agency delivers this layer, and the first marketing hire usually lands here too, because it is the most visible and the easiest to measure.

L2, revenue and the commercial process. This is where the pipeline, the length of the sales cycle and the speed of movement between stages live. A good lead lost at the third meeting is an L2 problem, however much it looks like a campaign problem.

L3, orchestration and memory. This is where company data, written procedures and the reasons behind past decisions are kept. It separates a company that learns from every campaign from one that restarts from zero whenever a vendor or an employee leaves.

L4, positioning and category architecture. This is where the criterion you are compared on gets decided. It is also the heart of our methodology, since this layer determines whether the rest of the chain works for you or against you.

Our reading of the market is that most companies operate only on L1 and L2 and expect immediate results from advertising. The agency versus in-house choice is made down there as well, so it compares two ways of covering the first layer and leaves L3 and L4 without an owner.

The building rule runs bottom up, because you cannot position a company that has no data yet about its own customers. Messaging follows the opposite rule, top down, from positioning to the commercial offer and then to the ad. Anyone who jumps straight to the ad with a bigger budget pays what we call the Confusion Tax on every impression.

Why does infrastructure come before the agency or hire decision?

Infrastructure comes first because it decides what anyone who follows will have to execute. The infrastructura-inainte-de-marketing pillar in our Codex holds that a client rarely has a marketing problem and far more often has an infrastructure gap, meaning systems, channels, positioning and a brand foundation, and that marketing built on a weak base burns money.

The operating rule behind the pillar applies directly here. Before any advertising you diagnose the channels, the sales people and the clarity of the brand, and if one of them is missing, that is the work, in that order. Only after the diagnosis does it make sense to ask who should execute.

An example from the positioning analysis we published for Cablero Steel Group, a steel cable manufacturer in Iași, Romania, shows the mechanism. The company states on its own site that it produces one linear metre of cable every 1.6 seconds, while its industrial references sat on a secondary page and in a trade magazine article from 2021.

When we analysed the site in September 2026, the top row of the home page talked about free shipping and fast delivery, which is what any reseller of the same goods promises. A new agency would have optimised ads around that message, and a new hire would have written more content about it.

The move we proposed shifted value from the product to the risk the product protects, with the line, translated from Romanian, “We don’t sell the cable. We sell what is not allowed to fail.” It also brought the existing evidence forward, a decision that could not come from whoever executes, since it means giving up the reseller message altogether.

The same pattern appears in the analysis we published for Zeelandia Romania, a producer of bakery ingredients. Our thesis was that the company holds the assets of a dominant position and speaks the language of an interchangeable supplier, while its factory, lab, certification and joint development with customers sat three levels deep in the navigation or lower.

In both cases the problem lived on L4, and any execution vendor would have amplified the existing confusion. The proposed solutions moved, raised or named what already existed, and a move of that kind is a strategic decision.

What does 2026 data say about marketing budgets?

Data from 2026 shows budgets that are close to frozen and expectations that keep rising, which is exactly the environment where the wrong execution choice costs the most.

The Gartner 2026 CMO Spend Survey, run among 401 marketing leaders in North America, the United Kingdom and Europe, found that marketing budgets rose only slightly to 7.8% of company revenue in 2026, from 7.7% in 2025.

In the same survey, 56% of CMOs said in 2026 that they lack the budget their strategy requires, and 54% reported insufficient resources. The pressure comes from being asked to deliver more with roughly the same money.

CMOs also allocate, according to Gartner, an average of 15.3% of the marketing budget to AI initiatives in 2026.

Seventy percent of them call becoming an AI leader a critical goal for 2026, and 70% also admit that their internal marketing processes are not mature enough to implement and scale AI effectively.

Ewan McIntyre, VP Analyst and Chief of Research in the Gartner Marketing practice, named the risk in the survey’s press release with a sentence that applies directly to our topic.

“The risk is that CMOs invest in AI tools faster than they build the data foundations, processes, governance and talent required to scale them.” (Ewan McIntyre, Gartner, 2026)

The same logic applies to the agency versus in-house decision, since both are execution tools placed on top of a foundation that may not exist yet.

This is where the second Codex pillar comes in, eficientizare-prin-strategie-nu-doar-optimizare. It holds that the marketing budget exists for development, meaning strategy and architecture, and that endlessly optimising the same channel leads to the same ceiling. It also holds that the right strategy lowers the real cost of acquisition and of each click more than any ad tweak.

Applied to our topic, the pillar explains why switching agencies rarely changes results. A new agency starts with energy, optimises the comfortable channel and reaches, within a few months, the same ceiling as its predecessor. A new hire walks the same road, only more slowly and at a fixed cost.

Our clearest evidence comes from the case study we published about a cardiology clinic in Bucharest, one of our clients. We rebuilt the positioning before touching a single campaign, and within seven months revenue rose 39% and bookings rose 57%, on the same ad budget. These are our own figures, published together with the method in that case study.

What stays in-house even when you work with an agency?

Three things stay inside the company whoever executes, and they are the real object of the choice. The first is the positioning decision, the second is the accumulated memory about customers and decisions, and the third is the budget treated as a decision about priorities, which nobody outside the company can make for you.

The positioning decision. You can ask an agency or a consultant to draft it, yet final approval of what you promise and why you deserve to be believed stays with you. An agency that takes this decision alone, without the founder signing off, has made a business decision on your behalf and delivered it to you as a marketing asset.

The accumulated memory. Every campaign, every test and every conversation with a real customer produces information about what works. If that information lives only in the heads of the agency team or of the employee who leaves, you bought the execution and lost the learning that came with it.

Our AI-invata-din-dialog pillar describes how we keep that memory. The pillar holds that a methodology settles into the AI from accumulated dialogue, and that an AI trained on that corpus surfaces recurring principles on its own and produces reusable intellectual property. A system that learns from dialogue keeps the reason behind each decision alongside the list of changes.

For a company the consequence is practical. Who buys, why they buy, which objection keeps coming back and which message lost must all be written down somewhere the company owns. We build that place for every client as part of the company’s AI Brain, holding decision documents, reports and the reasoning behind each choice.

The budget as a priority decision. How much goes to brand and how much to performance, to content or to paid media, is a strategic decision. An execution vendor naturally tends to recommend more of what it delivers, and an employee tends to recommend what they already do well.

Those tendencies come from where each party stands, with no bad faith involved, which is why the budget gets decided one layer above execution, by someone accountable for the whole company and for more than a single channel.

Agency, in-house or hybrid: how do they compare?

The table below puts the three options side by side on the criteria that matter for a company with more than 3 million euros in revenue. None of the columns wins universally. The wrong column is simply the one chosen before the strategy question has an answer.

This table has a limit we state openly. A hybrid model works only when the strategy partner answers for verifiable decisions, with the reasoning written down. A consultant who delivers polished slides and disappears after launch moves the problem elsewhere and leaves it unsolved.

How do you know whether your company needs an agency or an in-house team?

The answer depends on one factor, how mature your strategy is before you hire someone or sign with an agency. If positioning is clear, tested with real customers and written down, the choice becomes operational. If it is not, the first vendor or the first hire will spend their opening months building what should have existed already.

That construction cost is paid in both options, as a salary for the in-house team and as a monthly retainer for the agency. The real difference shows up only once strategy exists. From then on the in-house team wins on steady execution speed, and the agency wins on access to several disciplines without a fixed salary for each.

Four questions, asked in this order, tell you where you stand before you sign anything.

1. Can you write in one sentence why customers choose you, and does that sentence pass the competitor test?
2. Is there one place that holds the reasons behind last year’s marketing decisions, together with their results?
3. Do you know which part of the budget builds your position and which part buys immediate demand?
4. Do sales and marketing report the same number for the same quarter?

If the answer to the first two is “no”, choosing between an agency and a hire is premature. If the answer to all four is “yes”, the choice becomes a matter of capacity and cost, and you can make it calmly, because direction no longer depends on who executes.

One more sign shows up without any analysis, when the company rewrites its message, its site and its sales deck every time it changes vendor, because strategy has then been living with the vendor and was outsourced together with the work.

Where does UNRIVALS sit between an agency and an in-house team?

UNRIVALS is not a third way of buying execution, and it is not a better version of an agency. We work as a strategy and systems partner on L4 and L3, meaning the company’s positioning and the memory that keeps it stable.

Execution on L1 and L2 attaches to that architecture, whether an in-house team or specialised vendors carry it out.

The reason is the first pillar again. Execution placed on a company without foundations burns money, and a vendor who takes the brief from someone else cannot build those foundations. The diagnosis comes first, positioning comes second, and the choice of who executes comes last, once there is something worth executing.

For a company that already has a strong marketing lead, this means keeping that person and handing them a direction they no longer renegotiate with every campaign. For a company that works with an agency, it means the brief comes from written positioning, so the agency can change without strategy walking out the door with it.

We compared a part-time marketing executive with a strategy partner in detail in our guide to the fractional CMO role. And marketing architecture as infrastructure shows how the layers connect to each other, step by step.

Frequently asked questions

Can you give me an example of in-house marketing?

A typical example is a marketing department that produces social content, email and search campaigns without outside help. The ANA study from 2023 found that 65% of respondents had moved work once handled by external agencies to their in-house team over the previous three years, most often creative for social media, search and email.

Can you have both an agency and an in-house team?

Yes, and among large advertisers the combination is the norm. In the ANA study published in 2023, 92% of respondents also work with outside agencies. The condition is that strategy lives in a single place, otherwise every vendor builds its own version of the company.

What happens to strategy when you change agencies?

If strategy lived only in the agency team’s heads, it leaves with them. If it was written down as a company asset, meaning positioning, messages, customer knowledge and the reasons behind decisions, the new agency or the new hire executes on the existing foundation instead of rebuilding from scratch.

When does an in-house marketing team not make sense?

It does not make sense when execution volume cannot justify a senior full-time salary, or when the company needs several disciplines at once, from branding to performance and video, that no single hire covers well. It makes no sense either when positioning is missing, because the first person you hire will try to invent it alone.

Where do you start if you have neither an agency nor a team today?

You start with the diagnosis, and hiring comes after it. Write down why customers choose you, what evidence you have and what the budget actually buys, then decide who executes. The signs that your current message is not working are described in our guide to an unclear brand.

If you want to find out who owns strategy in your company today, and what an executor should receive from you, whether an agency or an in-house team, ask for a diagnostic.