If you already run Google Ads against a live store, "should I hire a Performance Max agency?" is really three questions: what does one actually do, what does it cost, and will it make you more money than running PMax yourself. Most agency pages answer the first and dodge the other two. This one walks all three — with the arithmetic the sales call skips.
What a Google Performance Max agency actually does
Performance Max is Google's cross-inventory automated campaign: one campaign serves across Search, Shopping, Display, YouTube, Gmail, Discover, and Maps, driven by your Merchant Center product feed plus asset groups. You hand Google a feed, creative, audience signals, and a target — Google decides placement and bid per auction.
That automation is exactly why the "agency" question is confusing. If Google runs the bidding, what's left for a human? Plenty, because the inputs still decide everything. A good PMax agency works the parts automation can't:
- Account structure. Deciding what runs as PMax, what stays in Standard Shopping, and what belongs in a dedicated brand Search campaign. This is portfolio construction, not an on/off switch.
- Feed quality. For retailers, Google's automation leans hard on the feed — smec, which manages campaigns for hundreds of retailers, reports that 74–97% of PMax cost goes to feed-based ads. Titles, GTINs, images, and price parity are the load-bearing input, whatever the banner creative looks like.
- Target setting. Translating your margins into a Target ROAS that scales instead of strangling volume.
- Brand protection. Keeping PMax from quietly eating conversions your brand search or organic listings were already getting.
- Diagnosis. Reading the reports when ROAS drops, instead of guessing.
If an agency's pitch is "we'll turn on PMax and let the AI work," you're paying a retainer for a setup wizard. The value is in the structure and the diagnosis. Our Google Ads economics guide breaks down how those pieces fit together for a store that's already spending.
What a Performance Max agency costs
Agencies usually price one of two ways: a percentage of your ad spend, or a flat monthly retainer. Both have a trap.
Say your store spends $4,000 a month on Google and an agency charges 15% of spend. That's $600/month, or $7,200 a year. The percentage model quietly rewards the agency for spending more of your money — their fee rises with your budget whether or not the extra spend is profitable.
A flat retainer — say $1,500/month — removes that incentive but front-loads the cost. On a $4,000 budget, that retainer is 37.5% of what you're already spending. At that point the agency has to generate a lot of extra profit just to break even on its own fee.
Here's the math that matters. Your total cost of advertising isn't the ad spend — it's spend plus the fee. If the agency on the $4,000 account charges $600 and your contribution margin is 40%, they need to add roughly $1,500 in revenue ($600 ÷ 0.40 = $1,500) just to cover their own cost before you see a dollar. That's the bar. Benchmark context helps you sanity-check the budget itself: WebFX's 2026 aggregate puts ecommerce Google Ads cost per acquisition in the $20–50 range, so you can estimate how many orders your spend should produce before layering a fee on top. For a deeper look at sizing the budget itself, see our guide to cost-effective Google Ads budget management.
When hiring a PMax agency pays off — and when it doesn't
The binding constraint for PMax isn't talent, it's conversion volume. Google's own documentation says Target ROAS on Search and Shopping needs at least 15 conversions with values in the past 30 days to bid at all, and practitioners widely cite roughly 30–50 conversions a month per campaign before the system behaves stably. Below that floor, no agency can make the automation reliable — there simply isn't enough data.
An agency tends to pay off when:
- Your account clears that volume floor and the limiter is structure, not spend.
- You're running PMax and brand search together and suspect cannibalization (more on that below).
- Your feed is a mess and nobody in-house owns it.
- You're scaling and the account has outgrown "set the target and hope."
It tends to not pay off when your budget is small enough that the fee swamps any gain, when you have almost no conversion history for the system to learn from, or when the agency won't show you the per-channel numbers. The hybrid approach most accounts should run isn't exotic anymore — Optmyzr's study of 24,702 PMax campaigns found 82% of advertisers run PMax alongside Shopping or Search, and split-budget accounts posted the strongest returns in that data. If an agency treats PMax as a replacement for everything else, they're a cycle behind.
How to vet a Performance Max agency
The old knock on PMax was that it's a black box. That's stale — since 2025, Google ships a channel performance report that breaks PMax results out by Search, Display, YouTube, Discover, Maps, Gmail, and Search partners, plus full search-terms reporting and campaign-level negatives. A competent agency uses these. An outdated one still tells you "you can't see inside PMax."
Questions that separate a good agency from a bad one
"How will you keep PMax from cannibalizing my brand traffic?" PMax chases the cheapest conversions, and branded searches are the cheapest of all — so it tends to absorb clicks you'd get anyway and report a flattering ROAS. The fix exists: account-level brand exclusions (available since 2024, covering variants and misspellings) plus a dedicated brand Search campaign. If they don't mention brand exclusions, keep looking.
"How do you set my Target ROAS?" The right answer derives it from your margins, not from a benchmark. Setting a target far above what your account has historically hit doesn't force efficiency — Google states plainly that it may limit the amount of traffic your ads get. Raising the target is a volume decision wearing an efficiency costume.
"When ROAS drops, what's your first check?" The right answer is "change history, then conversion tracking" — not "we raise the budget." CPCs have been climbing industry-wide; WordStream's 2025 data, reported by Search Engine Land, found about 87% of industries saw year-over-year CPC increases, so a drop is often the market, not your account. A good agency rules out measurement and market before touching your settings.
"Will you report true profit, or ROAS?" This is the one that matters most, and it's the one almost no agency answers well.
The number every agency pitch leaves out: your break-even ROAS
ROAS is revenue divided by ad spend. It ignores product cost, fulfillment, fees — and the agency's own retainer. A 5.0x ROAS can still lose money on thin-margin print-on-demand.
The number that decides everything is break-even ROAS, and it's pure arithmetic:
Break-even ROAS = 1 ÷ contribution margin.
Say you sell a mug for $24. Your Printify base cost plus shipping is $13, and payment processing takes about $1. That leaves $10 of contribution, a 41.7% margin ($10 ÷ $24). Your break-even ROAS is 1 ÷ 0.417 = 2.4x. Every dollar of ad revenue below a 2.4x return is losing money before you even count the agency fee.
Now layer the fee in. If the agency charges 15% of spend, that's another cost against every sale. Your real target has to clear break-even plus a profit buffer plus the fee — not the "good ROAS" number from a blog. This is why the honest test of any agency is simple: did your true per-order profit rise after their cut? Not impressions, not clicks, not a dashboard ROAS. Profit. For the keyword-level side of this math, see how we break down what Google Ads keywords actually cost.
Where an AI employee fits
Here's the gap an agency leaves open: between monthly reviews, nobody is watching the account against your true margins in real time. That's the work PodVector AI's Victor does.
Victor is an AI employee who operates your Google Ads account directly — he pauses the losing asset group, shifts budget toward the SKUs that actually clear break-even, adds negative keywords, and adjusts targets as the numbers move. Because he also sees Shopify orders, your Meta spend, and your product costs from Printify, Printful, and Gelato, he computes the true per-order profit on every sale — product cost, fulfillment, and fees included — not the dashboard ROAS your agency reports. Victor is not a dashboard you check; he does the work and delivers the reports straight to your Google Drive.
That changes the agency question. You don't need someone to babysit the automation daily if Victor already runs Google Ads against your real profit, every day. An agency — if you still want one — gets to focus on strategy, while the day-to-day execution and the profit math run continuously underneath it. When you're ready to go further on the feed side, our guide to choosing a Google Shopping ads agency covers the product-data layer that PMax leans on.
See what Victor does inside your Google Ads account.
FAQs
How much does a Google Performance Max agency cost?
It depends on the pricing model. Agencies typically charge either a percentage of your ad spend or a flat monthly retainer. The number that matters isn't the sticker price — it's whether the agency generates enough added profit to cover its own fee. On a store with a 40% contribution margin, an agency has to produce roughly $2.50 in revenue for every $1 of fee just to break even on itself.
Do I need an agency to run Performance Max, or can I do it myself?
If your account clears the conversion-volume floor — Google documents a minimum of 15 conversions in 30 days for Target ROAS, with practitioners citing 30–50 for stability — the mechanics are learnable. The real question is whether structure, feed quality, and daily profit-aware management are getting done. Those can come from an agency, from you, or from an AI employee like Victor operating the account directly.
How do I know if my PMax agency is actually working?
Judge on non-brand performance and total-business profit, not per-campaign ROAS. A PMax campaign eating branded search can show a spectacular ROAS while adding zero incremental orders. Ask for the channel performance report, the search-terms report showing branded versus non-branded queries, and a comparison of total orders to total ad spend over time.
Is Performance Max replacing Standard Shopping?
No. They now compete on equal Ad Rank footing in the same account, and the Optmyzr study found 82% of advertisers run both, with split-budget accounts performing best. The current best practice is a portfolio — PMax plus Standard Shopping plus a brand Search campaign — not a single campaign type doing everything.
What's the biggest mistake agencies make with PMax?
Letting PMax cannibalize brand traffic and then reporting the inflated ROAS as a win. The controls to prevent it — account-level brand exclusions and a dedicated brand Search campaign — have existed since 2024. An agency that doesn't use them is selling you conversions you were already getting for free.