Freelancer, expert, or agency: who should actually run your ads?

Most people searching for help with paid media type two different queries within the same hour. First "Google Ads expert", then "Google Ads agency" — or the reverse. The words feel like they describe different levels of the same thing, one more skilled and one more corporate. They do not. They describe three structurally different ways of buying the same work, each with a different cost base, a different failure mode, and a different spend range where it stops being the right answer. Picking correctly matters more than picking the best of any one category.

I run an agency, so read the rest with that in mind. I have also spent years watching accounts arrive from freelancers and leave for in-house teams, and the pattern of when each arrangement works is consistent enough to write down honestly. There is a real spend range below which hiring us would be a bad decision for the client, and I would rather say so here than discover it three months into an engagement.

The three options, defined properly

The labels are used loosely in the market, so it is worth being precise about what you are actually buying in each case. The distinction that matters is not seniority — it is how many people are accountable and what happens when one of them is unavailable.

• →A freelancer is one person selling hours, usually across several clients, typically without a formal process layer or cover. Strongest when the work is well-defined and bounded: build these campaigns, fix this tracking, run this account at a steady state.
• →An individual specialist — the person the market calls an "expert" — is also one person, but selling judgment rather than hours: a diagnosis, a strategy, an opinion on whether your account architecture is wrong. Often engaged on a project or advisory basis alongside whoever executes.
• →An agency is a team with a process layer: multiple operators, a review structure, cover for holidays and illness, and usually tooling and measurement infrastructure the individual options cannot justify building. You are paying for continuity and breadth as much as for skill.

Where a freelancer is genuinely the right answer

Below roughly $15,000 a month in ad spend on a single platform, a good freelancer is usually the best value available, and an agency is usually a bad deal. The reason is arithmetic rather than snobbery. At that spend, a typical agency fee is a large share of your media budget, and the work genuinely does fit in one competent person's week. There is not enough budget to run meaningful creative testing volume, not enough data for sophisticated bidding strategies to separate signal from noise, and therefore not much for a team to do that one person cannot.

The freelancer failure mode is specific and predictable: single-point-of-failure risk. One person gets sick, takes a holiday, takes on a bigger client, or simply loses interest, and your account goes unwatched. Budget runaways and creative fatigue do not pause for annual leave. If you go this route, the mitigations are unglamorous but effective — you own the ad accounts and pixels outright, you have documented access, and you have a named backup even if you never use them.

Where an individual expert beats both

There is one scenario where a single specialist is clearly better than either alternative: you have execution capacity but no senior judgment. An in-house marketer who can implement competently, a junior team that ships work reliably, or an existing agency doing the mechanics — and nobody who can tell you whether the whole approach is wrong.

That is a diagnosis problem, not a delivery problem, and diagnosis is exactly what an individual expert sells well. A few days of a genuinely experienced operator's attention on your account architecture, conversion definitions, and budget allocation will frequently find more value than a year of competent tactical execution pointed in the wrong direction. It is also the cheapest way to find out whether your current arrangement is the problem, before you pay switching costs to move.

The failure mode here is that advice without execution decays. A brilliant strategy document that nobody implements is worth precisely nothing, and the gap between "we know what to do" and "we did it" is where most expert engagements quietly die. If you buy judgment, be honest about who is going to act on it and by when.

Where an agency starts to earn its fee

An agency becomes the right answer when the work genuinely exceeds one person — and the honest test is not spend alone but how many things are running at once. Search plus Shopping plus Performance Max operating as a coordinated system, rather than three campaign types in the same account. Creative testing at a volume that requires briefing, production coordination, and analysis every single week. Measurement that needs rebuilding rather than maintaining, which is weeks of unglamorous work before performance moves at all. In practice these conditions cluster somewhere above $30,000 a month, though the number varies by how complex the business is rather than by the budget itself.

The second thing an agency sells is continuity, which is undersold because it is invisible when it works. Multiple people who know the account, cover during absence, a review structure that catches the thing the primary operator missed, and institutional memory that survives one person resigning. If your paid media is a material share of revenue, the cost of nobody watching for two weeks is usually larger than the fee difference between the options.

The agency failure mode is the one worth screening hardest for, because it is common and it is well hidden by the sales process: you buy a team and receive one junior operator carrying too many accounts. The senior people in the pitch were the sales layer. The mitigations are simple questions, asked before signing — who runs the account day to day, how many accounts do they hold, and who do I speak to when performance drops on a Friday. The full list is in questions to ask before hiring an ads agency.

What each option actually costs, and why

The price gap between these options is mostly a cost-base difference, not a margin difference. A freelancer carries almost no fixed cost, so the fee is close to the labor. An agency carries a bench, a process layer, tooling, and the cost of cover — which is why the same nominal hour is more expensive and why the comparison is not apples to apples. Whether that overhead is worth paying depends entirely on whether you need what it buys. The market fee structures, and how each behaves as you scale, are surveyed neutrally in how much a Google Ads agency costs and PPC management pricing models compared.

A decision sequence that usually works

1. Establish what is actually wrong before shopping for who fixes it. An independent read of the account tells you whether you have a diagnosis problem or a delivery problem, and those have different answers.
2. If spend is under about $15,000 a month on one platform, start with a strong freelancer and own everything — accounts, pixels, data, and documented access.
3. If you have execution capacity but no senior judgment, buy an individual expert's diagnosis first. It is the cheapest way to find out whether your current setup is the problem.
4. If multiple channels, real creative volume, and measurement rebuilding are all in scope at once, an agency is the honest answer — but screen for delivery seniority, not pitch seniority.
5. Whatever you choose, define in writing how success will be measured before any contract starts. Every one of these arrangements fails the same way when nobody agreed what the scoreboard was.

If step one is where you are, that is what our free audit is for. It is an independent read of your Google Ads or Meta account, and it is deliberately useful whether or not the conclusion is that you should hire an agency at all — including us. For the adjacent decision about building capability internally, in-house vs agency covers the same ground from the other direction.