What Google Ads Management Costs, and What You Get for It

The four ways Google Ads agencies bill, what each one quietly rewards, what is included, and the questions to ask before you sign a contract.

By Gavin Sevastian · Updated August 4, 2026 · 9 min read

There is no independent, published benchmark for what Google Ads management should cost. Not a good one, not even a bad one. Every fee range you will find in a search result was published by a company that sells the service. That does not make the numbers wrong, but it does mean they were written by someone with an interest in where you anchor before the first call.

This page is published by a one-person Google Ads business based in Windsor that sells exactly this service. So everything below is either sourced to Google's own policy pages and public datasets you can open yourself, or it is clearly labelled as my opinion. Weigh it accordingly.

1. What management actually costs, and why nobody can give you a straight number

The numbers you will find, and who published them

Search this question and you get a remarkably consistent set of models: a percentage of ad spend, a flat monthly retainer, a one-time setup fee, or a sliding scale where the percentage falls as the budget grows.

The same bands repeat across pricing pages, but the pages are generally published by firms selling management and rarely point to a survey underneath. Corroboration is not verification. Repetition tells you what the industry advertises, which is genuinely useful, but it is not a measurement of what anyone pays.

So read those bands as a description of how the market markets itself. Not as a benchmark you are failing to hit, and not as proof that a quote outside the range is unreasonable.

What actually moves the price on your account

Five things move a quote more than anything else. How many campaigns and service lines you run. Whether conversion tracking has to be built from scratch or already works. Whether landing pages are in scope or you already have them. How many locations you cover. And how fast you expect an answer when you text on a Saturday.

Your budget matters too, though less than most people assume, and it matters differently depending on the billing model (covered next). If you are still working out what to spend, start there rather than with fees.

The sixth factor is the one nobody puts on a pricing page: whether the person who quotes you is the person who touches the account.

2. Percentage of spend and flat fee: the two models you will actually be offered

Percentage of ad spend

You pay the manager a share of what you spend with Google, usually with a monthly floor. It is the most common model in paid search and it is easy to explain, which is a large part of why it survives.

Here is what it pays for, stated plainly and without conspiracy. The manager's revenue rises when your budget rises. They earn nothing extra for finding an efficiency that lets you get the same jobs for less money, and their income is barely affected by whether the spend actually worked. That is not an accusation about any individual. It is arithmetic, and it applies to honest operators too.

It also has real virtues worth naming. It scales down for a small advertiser who could not carry a serious flat retainer. It caps itself automatically when you pause for a slow season, which matters in a trade with a genuine off-season. And a contractor who is nervous about committing to a fixed monthly number often finds it the easiest first step.

The practical test is simple. Ask what happens the month you want to cut spend in half. A good answer treats that as a normal conversation. A bad answer talks you out of it before asking why. If you have not settled on a number yet, work out what your own budget should be before you let a fee structure decide it for you.

Flat monthly fee

You pay a fixed amount regardless of what you spend. Revenue is locked, so the only variable the manager controls is how many hours they put in.

That cuts both ways, and both sides deserve saying. It rewards doing less once an account settles into a rhythm, because month eleven pays the same as month one whether or not anything happened. It also penalises them for the client whose account genuinely needs unusual attention, which is exactly the client who most needs it.

The virtue is that budget advice comes out uncontaminated. Telling you to spend less costs a flat-fee manager nothing, so when they say your budget is fine where it is, you can believe them. Same when they say a campaign should be paused. That is worth more than it sounds, because "spend more" is the single most common recommendation in this industry and it is very hard to evaluate when the person making it gets paid for it.

The buyer's risk under a flat fee is the account that goes quiet. No news, no reports that say anything, no changes. Section 6 covers why that is so hard to spot from the outside.

Base plus percentage, and other hybrids

A floor plus a share of spend above it, or a base plus a bonus tied to results. This is probably the most common structure in practice, and it is usually presented as the best of both.

It is more honest to say it softens both extremes without escaping either. A hybrid still pays more when your budget grows, just less steeply. It still pays the same in a quiet month, just at a lower base. Ask what each component rewards separately, rather than treating the blend as if the incentives cancelled out. They do not cancel. They average.

3. Hourly and performance-based: the two you should look at harder

Hourly

You pay for time, tracked and invoiced. It is transparent about inputs and completely silent about outputs.

The structural problem is that it charges you for inexperience. The operator who solves your problem in one hour bills a sixth of the one who takes six, so the model pays for slowness and taxes competence. The quieter cost is that you start rationing your own questions, and the questions you swallow are usually the ones worth asking.

Where hourly genuinely fits is defined, one-off work with a clear end: an account audit, a build, a second opinion on something specific. Bounded scope, bounded time, no incentive to stretch. For ongoing management, where the work is open-ended by nature, it is the weakest of the four.

Performance-based, pay per lead, or a share of revenue

This is the model that sounds most aligned, and it is the hardest to run cleanly. The reason is not bad faith on either side. It is that somebody has to measure the thing you are paying on, and in paid search that measurement is genuinely contestable.

What counts as a lead? Does a hang-up count? A price shopper outside your service area? What attribution window applies, and who credits the job that closed six weeks later from a phone call that was never tracked? Every one of those is a legitimate disagreement between reasonable people, and every one of them decides an invoice.

The second issue is directional. Paying per countable event pushes toward volume of countable events, which is not the same as quality of booked jobs. You will want to know what a lead is actually worth in your trade before you agree to buy them one at a time.

And anyone taking real downside risk will price that risk in and will want control over the funnel to protect it: the page, the phone number, the follow-up. That is rational of them. It is just a different set of frictions, not the removal of risk.

What the buyers who have used it longest are doing

The ANA's Trends in Agency Compensation study, which surveys client-side marketers at large national brands and therefore says nothing directly about contractor-scale fees, found use of performance incentives fell to 41 percent in its 2022 edition, down from 48 percent in 2016 and 61 percent in 2013.

Take that as evidence about direction of travel between models, from buyers with in-house teams and measurement budgets far beyond anything a contractor has. If they cannot tell whether it worked, be sceptical of a pitch that says it obviously will.

4. The management fee is the small number

Three numbers, not one

A contractor comparing a $600 quote against a $1,400 quote is optimising the smallest of three numbers, and usually the one that matters least.

Total outlay is fee plus ad spend. For most contractor accounts, spend is the larger of the two by a wide margin. Effective cost is fee plus spend plus whatever share of that spend went to clicks that were never going to become a job. Nobody quotes the third number, nobody puts it on an invoice, and for most accounts it is bigger than the gap between any two fees you are choosing between.

That is the whole reframe. You are not buying management. You are buying discipline over the larger line.

What waste looks like in accounts that have been measured

WordStream published a study of 251,236 reports run by 15,666 Google Ads accounts through its free grader between January and November 2025. About 29 percent of those accounts recorded zero conversions over a ninety-day window while still averaging over twelve thousand impressions a month. A quarter had never added a single negative keyword.

Read it with the caveats attached, because they matter. This is a self-selected sample of accounts whose owners went looking for a grader, which is not a random draw. "Wasted spend" is the vendor's own definition. The data is US. And the gap they report between accounts with negative keywords and accounts without is a correlation, not proof that adding negatives caused it, since those accounts differ in a dozen other ways too.

What survives all of that is still the useful part: a meaningful share of live accounts are spending money and producing nothing, and they look busy the entire time.

Cheap management is not a cheap channel

The fee is bounded. You know it before you sign it and it does not surprise you. The waste a manager fails to prevent has no ceiling and never appears on a statement labelled as waste.

Compare fees against the larger line they are meant to protect. A lower retainer can still be the more expensive choice when the difference in management quality creates more wasted spend than the fee saved. That is why scope and evidence matter more than a neat percentage. The mistakes that produce most of that waste are well documented and mostly unglamorous.

5. What is usually included, and what usually is not

What almost every engagement includes

Nearly every quote covers the same core: building or restructuring the campaigns, keyword and negative keyword work, writing the ads, managing bids and budgets, and sending some form of monthly report.

There is one published floor underneath that. In its guide for advertisers working with third parties, Google states that at a minimum you have the right to know the number of clicks, impressions, and the total cost of your Google ads.

Notice what that is. It is a reporting entitlement, not a scope of work. Google says nothing about what a manager has to do for you, only about what they have to show you. The distance between those two things is the entire subject of this section.

Where two quotes differ most

Two quotes at the same price can describe genuinely different jobs. These are the places they diverge, in rough order of how much money rides on each.

Landing pages: built or not, and if built, who owns them and what happens to them when the relationship ends. Conversion tracking: implemented and verified, or assumed to be working because someone installed a tag once. Call tracking, and whether anyone ever listens to the calls or only counts them. If the ads are running and the phone still is not ringing, that cluster is usually where the answer is hiding.

Then: how often ad copy gets refreshed. Whether search terms are reviewed on a schedule or only during launch. Reporting frequency, and whether a person interprets it or a dashboard emails itself to you. How fast you get an answer when you text on a Saturday. And whether the outcome of a lead, booked or junk, ever makes it back into the account.

Any one of those can swing the real value of an engagement further than the fee gap you were comparing. If you already have someone and you are trying to work out whether the work is actually happening, that is a separate question worth its own answer.

What is almost never included unless you ask

Other ad platforms are out of scope by default. So is anything on your actual website as opposed to a landing page, along with CRM setup, review generation, and following up on leads once they land in your inbox.

None of that is a slight, and a provider who excludes it is not cutting corners. It is scoping. The failure mode is assuming it was included, finding out in month three that it never was, and treating a scoping gap as a betrayal. Ask, get the answer in writing, and price accordingly.

6. What the manager does that you cannot see

The work that repeats every week

Search terms drift. The phrases people type change with the season, with the weather, and with whatever a competitor started advertising last Tuesday. Google's matching decides which of those phrases your ads are allowed to appear for, and that set is never the same two months running.

Competitors move too. They raise bids, rewrite their ads, enter your city, leave it again. None of that announces itself inside your account. It arrives as a slow change in what you pay and who you reach.

So the recurring work is not one difficult task. It is a set of ordinary ones that never finish, and the account only stays current because somebody keeps looking. That is also part of why results take the shape they do in the first few months: early on you are mostly finding out what the account is really matching.

The work the platform forces on you

Google ships changes to bidding, campaign types, and default settings on its own schedule, and it does not ask first. Some of those changes are genuine improvements. Others quietly widen where your money goes.

The defaults are the part worth understanding. They are set in Google's commercial interest, which is a reasonable thing for Google to do and a poor thing for you to accept unexamined. Overriding the ones that do not suit a local contractor is baseline competence, not an upsell.

It is also completely invisible to you. Nobody sends a report saying a default was caught before it cost anything. Which is exactly why it is the first thing to slip when a manager is carrying too many accounts for the fee they are charging.

Why a neglected account looks fine for months

Here is the uncomfortable part. A well-run account and an abandoned one produce the same experience for a long time. Ads appear when you search for yourself. The dashboard shows numbers. The invoice arrives on time.

A conversion tag that broke in April reports no problem. A call path that stopped connecting does not send an alert. An account drifting into searches that were never going to hire you just looks like an account with a lot of impressions.

All three present identically, as silence. Which is why checking whether the work is happening has to be something you do on purpose, rather than a complaint you wait to have.

7. The rules Google puts on anyone who manages your account

The management fee has to be disclosed in writing, and on every invoice

Google's transparency requirements for third parties are short and specific. If you charge a management fee separate from the cost of Google Ads, in Google's words, "you must clearly inform customers. At a minimum, inform new customers in writing before each first purchase and disclose the fee on all customer invoices."

Read that as a floor rather than an ideal. Being told the fee on a call does not satisfy it. Seeing it once in a proposal and never again on an invoice does not satisfy it either.

So if you have been paying for a year and have never seen your management fee written as its own line, that is not paperwork someone forgot. It is a published requirement that is not being met, and it is a completely fair thing to raise without apology.

The spend they report has to be Google's number

The same policy page carries the sentence I would put in front of every contractor buying this service: "When sharing Google advertising cost data with customers, report the exact amount charged by Google, exclusive of any fees that you charge."

Think about what that rules out. A single blended figure on a report, where fee and ad spend arrive as one number, makes your fee impossible to audit. It also makes a markup on your ad spend indistinguishable from the ad spend itself. You cannot tell those apart from the outside, which is precisely why the rule exists.

There is a second requirement worth knowing about. Google states that if 80 percent or more of a third party's customers spend less than $1,000 USD a month, that third party has to share Google's advertiser disclosure notice with all its customers, linked in a clearly visible location on its website. Google names some examples of where, such as a homepage footer, but the binding part is that it is visible, not which page it sits on. Plenty of contractor accounts sit under that line, so the rule catches a lot of the providers quoting you. This business treats itself as inside it.

One account per advertiser, and your Customer ID on request

Two more, both from Google's third-party policies. Google requires a separate account for each end-advertiser being managed, so several businesses sharing one account is outside the rules rather than merely untidy. And Google requires that you be given the Customer ID for your account when you ask for it.

That last one is a single-line email you can send this afternoon: what is the ten-digit Customer ID on my Google Ads account? There is no legitimate reason for that to be difficult, and how fast the answer comes back tells you something on its own.

8. Who owns the account, and why it decides everything else

Manager ownership depends on how the account was connected

This is a consequential detail dressed as administrative trivia. Google's page on ownership of client accounts says a manager that creates a new client account automatically becomes its owner manager. A manager that links an existing client account does not receive that status by default.

If an agency created the account from its manager account, that manager holds the extra ownership controls by default. If you created the client account and later linked the agency, it does not. Google separately says the client account retains ownership of its data, so this setting is about a manager's administrative powers, not who owns the advertiser's data.

Neither of those is sinister. It is a default. It is just a default with teeth.

What an owner can do

Google lists the owner's privileges, and the ones worth knowing are these. An owner can invite users to the client account, remove users from it, and grant or revoke administrative access on it. That is not theoretical. It means the party holding ownership can remove you from your own account.

Ownership also travels upward. Google states that it is transitive, so if your manager account is an owner, every manager above it in the ownership chain has the same status. In a white-label or subcontracted arrangement that can include another company. It is worth asking about directly if your provider mentions a partner, a network, or a parent agency.

The way out, and the condition attached to it

Google says the client account keeps its own data and that a client-account user can unlink an owner manager.

Turning off ownership is an administrative action, and Google's own instructions state that the client account must have at least one user with administrative access for it to be completed. If nobody on your side holds Admin, there is no one who can pull the lever. The exit exists, and it only works if you already hold the key.

Which makes the action item boring and important. Practical control means direct access through an email address your business controls, Admin rights on that user, and a record of the ten-digit Customer ID. Confirm those while nothing is wrong.

9. Badges, certifications, and what they actually measure

What the Google Partner badge requires

A Partner badge looks like a quality signal, and the requirements are published, so you can decide for yourself. Google's page on becoming a Google Partner lists three: a minimum optimisation score of 70 percent on the registered manager account, a 90-day ad spend of $10,000 USD across all managed accounts, and at least half of the firm's account strategists certified in Google Ads.

The spend bar is the one worth noticing. Ten thousand USD over ninety days, across every account the firm manages, works out to roughly $3,300 a month in total. A single mid-sized contractor account can carry that alone. It is a participation threshold, not a ranking.

Worth separating this from the badge on your own business. A Partner badge says something about the agency. Google Guaranteed is a screening check on your company and shows up next to your own listing, which is a different thing entirely and often more useful to a homeowner.

What optimisation score measures

Optimisation score is Google's estimate of how well an account is set to perform. A firm can maintain the score by applying or dismissing Google's recommendations. Some recommendations are sensible. Some widen targeting, broaden match types, or raise budgets.

So a 70 percent score tells you the manager account has kept up with that recommendation workflow. It does not prove that a particular account is run well for a contractor in Barrie, and it is worth knowing before you read the badge as evidence about your own account.

There is a related fact worth stating without any conspiracy attached, because Google publishes it plainly. Premier Partner status is scored partly on existing client growth, measured by year-over-year ad spend growth among current clients, and on total annual spend across managed accounts. That incentive sits upstream of whatever billing model your agency uses. A flat-fee agency faces it too.

What certification is

Google Ads certification is an online assessment of product knowledge.

That is a factual description, not a criticism. Somebody has to know the vocabulary. It just is not evidence of judgment, and it is not meant to be.

10. The questions worth asking before you sign

Money and reporting

Five questions, and each one has a weak answer worth listening for.

Is the fee stated separately from ad spend on every invoice? Will my monthly report show Google's cost on its own, unblended? Is there a setup fee, and specifically what does it buy that the monthly fee does not? What happens to the fee if I cut my budget in half for the winter? And does the fee move when spend moves, in which direction, and by how much?

The weak answers sound like this. "It is all in the dashboard." "We can walk you through it on the call." "Nobody has ever asked that." None of those are refusals, which is what makes them easy to accept. They are just answers that leave you exactly where you started.

Access and ownership

Whose name is the Google Ads account in? Am I set to Admin on it? What is my ten-digit Customer ID?

Then the exit questions, which are the ones people skip because it feels rude to ask at the start. If we part ways, what do I keep: the account itself, the conversion history, the landing pages, the tracking setup, the phone numbers? And is there a manager account above yours, meaning a partner network or a parent agency, that also holds access?

Ask all of these before you sign, not after. Every one of them is cheap to answer on day one and expensive to resolve on day two hundred.

The work, and the exit

Who actually touches my account, and how often? That is the single most useful question on this page, because the word "agency" does not tell you the team size and the person who sells is often not the person who works.

Then: what would cause you to tell me to spend less? What does month four look like compared with month one? Is there a minimum term, and what is the notice period?

And one disqualifier. Anyone who guarantees you a position, a ranking, or a spot at the top of Google is describing something that does not exist. Google states it directly in its guide for advertisers: it is not possible to guarantee a specific ad position, because position is decided by an auction that changes with every single search. The same page tells you to get your budget agreement in writing. Both are worth taking literally.

11. When hiring anyone is the wrong move

Your spend is too small to carry a fee

At a low enough budget, any management fee is a large share of your total outlay, and the arithmetic breaks before the question of management quality ever comes up.

Do not use a threshold somebody published. Use your own numbers. Add your realistic monthly ad budget to the fee you have been quoted, work out what share of that total is the fee, then ask how many booked jobs the ads have to produce for the whole package to make sense at your average job value and your close rate. If that job count looks implausible for your market, the answer is not a cheaper manager. It is more budget or not yet.

Plenty of contractors are genuinely in the not-yet category, and it is worth reading honestly on whether the channel suits your business at all before you spend anything on either half.

Your constraint is not lead volume

This is the most common honest disqualifier and it costs a seller nothing to say out loud.

If calls are going to voicemail during the day, if quotes take days to go out, or if too few estimates turn into work, more leads will not fix any of that. Ads amplify what is already there. Pour more inbound into a business that cannot answer the phone and you have bought yourself a larger pile of missed calls and the same revenue.

The test is straightforward. If you got twice as many leads next month, could you actually handle them well? If the honest answer is no, fix that first. It is cheaper, it is faster, and it makes every advertising dollar you eventually spend work harder.

You want to run it yourself

Entirely reasonable, particularly for a single service in a single city with a modest budget. People do it and do it well.

What it costs is attention, not intelligence. Initial setup is finite. The part that catches people out is the recurring work described in section 6, which does not care that you had a busy month. Whether that trade is worth it depends on what an hour of your own week is worth against a management fee, and only you can price that.

If you go that way, go in properly rather than half in. A grounding in how this works for a small business is worth more than a fee saved on a badly run account.

12. What this looks like in Canada specifically

US benchmark numbers do not transfer

The large public datasets on click costs, conversion rates and cost per lead are US-only and quoted in US dollars. WordStream's 2026 benchmarks, the most widely cited set, states plainly that it is drawn from US-based search campaigns with all currency values posted in USD.

They are still the best public numbers available and worth reading. Just check the country and the currency before you use one to judge your own account, because Canadian auction density, exchange rates and competitive mix all move the result. A number that looks alarming against a US median can be perfectly normal in Ontario.

Your HST number belongs on your Google payments profile

This is the most immediately actionable money item on the page and almost nobody mentions it. Google states that GST/HST applies to all Google Ads customers with a billing address in Canada who have not added a GST/HST registration number to their payments profile. Quebec has a separate QST rule that works the same way.

So if you are HST registered and your number is not on the payments profile, you are paying tax to Google on every dollar of ad spend that you may not need to be paying there. Check the profile, and take the actual tax question to your accountant rather than to me.

What "agency" means in Canada

Statistics Canada reported that the advertising and related services industry group earned $15.4 billion in operating revenue in 2024, with just under two thirds of it in Ontario. It also noted that salaries and subcontracting together made up more than half of total operating expenses, which is the honest explanation of what a management fee actually pays for. It is a labour business.

The size picture is the surprising part. Federal industry data counts 6,295 advertising agencies in Canada for 2024 and covers businesses with annual revenue from $30,000 to $5 million. That is not a criticism of the industry. It shows a wide market of businesses by revenue, but the dataset does not tell you how many people work at a particular firm.

Which moves the real question. Not big versus small, but who is going to open your account on a Tuesday morning. If you are shopping locally, the same question applies in Toronto as anywhere else.

If you take three things from this page

The management fee is the smallest of three numbers, and the one you are most likely to optimise. Total outlay is fee plus spend, and the real cost is fee plus spend plus the share of spend that was never going to produce a job. Scope, not price, is where two similar-looking quotes actually differ, so compare what each one covers before you compare what each one charges.

And the manager ownership setting was decided by how the account was created or linked, probably without anyone raising it with you. Your own direct Admin access determines whether you can keep practical control if the relationship ends, including access to conversion history you cannot rebuild.

The useful buying decision is therefore not which fee is lowest. It is whether the scope is explicit, the billing is transparent, the account stays under your practical control, and the person doing the work can explain what happens after launch. A provider who makes those answers easy to verify has already cleared a more meaningful bar than one who merely quotes the familiar rate.