Short version: In-house versus agency is not a loyalty question, it is a math question about three variables: how much you spend, how complex your account is, and whether you can realistically hire and keep a senior PPC person. In-house wins at large spend where a full-time specialist pays for themselves and the work needs daily contact with product and inventory. Outside help wins when spend does not justify a senior salary but the account still needs senior skill. Most brands that get this right end up with a hybrid: someone inside who owns strategy and someone outside who does the specialist work. This guide walks through the real cost of each option and a decision framework by monthly spend.
The real question behind the question
When a founder asks "should we bring PPC in-house", the question underneath is usually one of two things. Either the agency relationship has gone stale and in-house feels like taking back control, or spend has grown to the point where the fees look like a salary and hiring feels like the obvious upgrade. Both instincts deserve a fair hearing, and both can lead to expensive mistakes if the actual math never gets done.
The honest framing: this is a resourcing decision, not a philosophy. The account does not care who logs into it. It cares whether the person doing the weekly work understands product feeds, verifies conversion data, and has enough hours to do search term reviews and budget reallocation every week. Every structure can deliver that, and every structure can fail to. So the question becomes: which structure is most likely to put a competent, attentive person in your account at a cost your margins can carry?
The true cost of in-house
The salary is the visible number, and it is only the start. A genuinely competent ecommerce PPC manager, someone who has run Shopping and Performance Max against margin targets rather than just search campaigns against lead counts, commands a proper senior salary in most markets. On top of that sit benefits and employment costs, typically another 20 to 30 percent, plus tools that an agency would have bundled: reporting software, feed management, competitive intelligence, ongoing training as the platform changes underneath everyone every quarter.
Then there is the cost nobody budgets: the ramp. A new hire, even a strong one, needs three to six months to learn your catalog, your margins, your seasonality, and your customers before their decisions get sharp. During that ramp you are paying full price for partial output, and if the hire turns out to be wrong, you restart the clock along with the recruiting cost.
The structural risk is concentration. One in-house person is a single point of failure. When they leave, and PPC people move often because demand for them is constant, the account knowledge walks out with them. There is no colleague to hand over to, no bench, and often no documentation, because documenting was always the task that could wait. Many brands come to an agency precisely at this moment, mid-panic, with an account nobody left inside understands.
The most expensive version of in-house is the one that looks cheapest: hiring a junior to manage serious spend. A junior salary against a six-figure annual ad budget feels like a saving, but a junior learning on your money makes junior mistakes at full market prices. Wasted spend in an inexpertly run account commonly runs 20 to 30 percent, and on real budgets that dwarfs the salary difference between a junior and a senior. It is the most expensive saving available in this channel.
The true cost of an agency
Agency costs deserve the same honesty. The fee is the visible part. On top of it sits onboarding time: a good agency needs several weeks to learn your catalog and margins before their work gets sharp, which is a smaller version of the in-house ramp, not an escape from it.
Then there is shared attention. Your account is one of many, and how many matters enormously. An account manager carrying thirty accounts gives yours minutes per week; one carrying eight gives it hours. Quality across the industry is genuinely variable, and bad agencies exist in numbers: agencies that set up an account once and let it coast, or that report on clicks because the revenue story is unflattering. None of that is a reason to avoid outside help, but it is a reason to select carefully. We wrote a full guide on what an ecommerce PPC agency actually does and how to judge one, and the questions in it, who works on the account day to day, what would you change in thirty days, what happens if we leave, filter out most of the bad ones before any contract is signed.
The fair comparison, then, is not salary versus fee. It is total in-house cost including ramp, tools, and key-person risk, versus total agency cost including fee, onboarding, and the diligence required to pick a good one.
Where each model wins
Where in-house wins
At very large spend, the math flips cleanly. Once monthly budgets are big enough that a full-time senior person costs less than the equivalent management fee, and the account generates enough daily work to fill their calendar, in-house is simply cheaper per hour of expert attention. The threshold varies by market, but the logic does not.
In-house also wins when the work is inseparable from operations. A brand where PPC decisions depend on daily inventory positions, purchasing calls, and product launches benefits from someone who sits in those meetings, sees the stock report every morning, and can pause a campaign the moment a bestseller sells out. An outside partner hears about the stockout in the weekly call; an insider saw it at 9 am.
And some businesses have data that cannot leave. Regulated categories, brands with genuinely sensitive margin structures, companies whose competitive edge is the numbers themselves. If your data governance rules out external access, the decision has been made for you.
Where outside help wins
Below the spend level that justifies a senior salary, outside help wins on arithmetic alone. A specialist fee buys senior-level skill for a fraction of senior-level payroll, and the alternative, a junior hire, costs less on paper and more in the account. This is where most growing ecommerce brands actually sit.
Outside help also wins on breadth. A specialist who runs many accounts sees platform changes, bidding quirks, and category patterns across a portfolio, weeks before a solo in-house person would meet them in their single account. And it wins on variable workload: seasonal brands that need heavy hours in Q4 and light hours in February are paying a full salary for a part-time need if they hire, while a retainer can flex or the scope can change.
For agencies reading this from the other side of the table, the same build-versus-buy logic applies to your own delivery, and the answer there is usually white-label PPC rather than a paid-search hire.