Google Shopping management: what the real work looks like

Search "google shopping management" and every agency page says the same three things: feed optimization, campaign management, ongoing monitoring. Sounds complete. It also tells you nothing, because a freelancer who logs in twice a month can claim all three with a straight face.

The gap between real Shopping management and campaign babysitting is enormous, and it is invisible from the outside. Both send you a monthly report. Both have a dashboard. One of them is rebuilding your product feed against real search data every week. The other is watching numbers move and occasionally nudging a budget.

We manage Google Shopping for 200+ ecom brands across 12+ niches. This is what the job actually contains, what it costs, and how to tell within one phone call whether you are buying management or babysitting.

The five jobs inside real Google Shopping management

Google Shopping is not one system. It is a product feed, a Merchant Center account, and a set of campaigns stacked on top of each other. Managing only the top layer is like tuning a car engine while the wheels are flat.

1. Feed engineering. Your product titles, descriptions, categories, and attributes decide which searches your products match. Google reads titles algorithmically: "Cool Vintage Dad Hat" is invisible, "Black Cotton Vintage Trucker Cap" wins the queries. Real management pulls your search-term reports, finds the queries actually driving sales, and rewrites titles around them. The complete feed optimization guide walks through the full system.

2. Merchant Center compliance. Disapprovals kill products silently. Shipping policy mismatches, missing GTINs, wrong categories, variant pricing collisions - each one takes a product out of the auction. Real management catches these daily and clears them fast. Our autofix engine runs six built-in fix strategies and clears 80% of issues without a human touching them; most disapprovals clear inside 24 hours. If your current provider cannot describe how they fix disapprovals, they do not have a process for it. And if Merchant Center itself is the mess - suspensions, policy warnings, half the catalogue dead - the Merchant Center agency breakdown covers when that deserves dedicated attention.

3. Campaign structure. Which SKUs go into Performance Max, which run through standard Shopping, and how the catalogue splits into campaigns. PMax under-bids long-tail queries because its bid model favours volume. Standard Shopping catches the high-margin long tail PMax ignores. Most agencies abandoned standard Shopping when PMax launched. Running both, with rules about which SKUs go where, typically lifts Shopping revenue 25-40% over PMax-only setups in catalogue-heavy niches.

4. Bid strategy. Not "set target ROAS to 4x and walk away." A blended target treats your 60%-margin hero product and your 20%-margin accessory identically. Real management sets targets per product tier, using custom labels that sort SKUs into buckets like Champions, Sleepers, and Wasters so Smart Bidding can actually read the difference.

5. Monitoring. Daily, not monthly. A disapproval wave on a Friday should be fixed by Monday, not discovered in the next report. A product that starts burning spend with no sales should be caught within days, not after it has quietly eaten a month of budget.

Notice what is missing from most service pages: jobs one and two. That is not an accident. Feed and Merchant Center work is slow, unglamorous, and hard to fake. Budget nudges are easy.

Feed engineering is where the result comes from

Here is the uncomfortable ratio: about 80% of the Shopping ROAS lift comes from feed engineering, not campaign settings. We have rebuilt 200+ Merchant Centers on exactly that split.

Why the feed matters this much is simple mechanics. Every Shopping auction starts with a matching step: Google reads your feed and decides which searches your product is eligible for. A vague title matches broad, low-intent searches - you pay for clicks from people who were never going to buy. A specific, query-informed title matches the searches where people are ready to spend. Same product, same budget, completely different outcome.

That means the matching step is decided before any bid is placed. No bid strategy can fix a feed that matches the wrong queries. This is why "management" that only touches campaign settings hits a ceiling fast: it is optimising the second step of a two-step process while ignoring the first.

The practical work looks like this: pull 90 days of search terms, find the queries that actually convert, rewrite the top 80 SKU titles around them, deploy custom labels so bids can follow margin, and fix the attributes that trigger disapprovals. First 80 SKUs inside the first week. Full catalogue inside 30 days for a typical 500-SKU store.

What Google Shopping management costs

Two separate costs, and it is worth keeping them separate in your head. Your ad spend goes to Google - the agency never touches it. The management fee is what you pay the agency on top.

Fee models vary: flat retainers, hourly rates, or a percentage of spend. ZenoX charges a tiered percentage of monthly Google Ads spend, and the rate drops as you scale:

• First €10k of monthly spend: 10%
• €10-30k: 9%
• €30-80k: 8%
• €80-150k: 7%
• Above €150k: 6%

Each rate applies only to the spend inside its own bracket, not to the total. So as the account grows, your effective rate falls automatically. No setup fee, no lock-in, and the feed engineering and Merchant Center work is included - not billed as a mysterious "feed optimization add-on." The full math is on the pricing page.

Why a percentage of spend instead of a flat retainer? Alignment. A flat fee pays the agency the same whether your account doubles or flatlines. A tiered percentage only grows when your spend grows - and spend only grows sustainably when the revenue behind it does. If the account stops performing and spend gets cut, the fee drops too. That keeps everyone honest.

One more thing on cost: the fee is not the number to optimise. The number to optimise is what a badly managed feed wastes. Clicks from wrong-intent queries, products sitting disapproved for weeks, one blended bid target flattening your margins - that leak is usually a multiple of any management fee. The Shopping ads cost breakdown covers the budget math from the spend side.

Campaign babysitting: what most "management" actually is

Campaign babysitting is the industry's open secret. The service is sold as management, priced as management, and reported as management. What actually happens: someone checks the dashboard weekly, nudges a budget, screenshots the graphs, and writes a summary.

The frustrating part is that from the outside it looks identical. Reports arrive on time. The account manager is friendly. Numbers get explained. For months, nothing is obviously wrong - the account just slowly plateaus while the feed rots underneath.

Here is the side-by-side:

	Babysitting	Real management
Product titles	Whatever Shopify exports	Rewritten against real search terms
Disapprovals	Noticed at month end, maybe	Caught daily, most cleared in 24h
Custom labels	None	SKUs sorted by margin and velocity
Campaign types	One PMax, everything in it	PMax + standard Shopping, SKU rules for each
Bid strategy	One blended target ROAS	Targets per product tier
Change history	A few budget edits	Feed and structural work every week

The last row is the tell. Google Ads records every change made in an account, with timestamps. Babysitting cannot fake a change history.

How to tell the difference before you sign

You do not need to be a Google Ads expert to filter this. Four questions do it.

"Show me the last five title rewrites you shipped for a client, and the search-term data behind them." A real operator has these within reach and can explain the logic in plain language. A babysitter changes the subject to bidding.

"What happens when one of my products gets disapproved?" You want a specific process with a timeframe. "We monitor for that" is not a process. "Six fix strategies, most disapprovals cleared inside 24 hours" is a process.

"How do you decide which products get which bids?" If the answer does not mention custom labels, margin tiers, or splitting the catalogue, everything runs on one blended target - which means your best products subsidise your worst.

"Walk me through your last 60 days of change history on an account like mine." The single hardest question to fake. Real management produces a paper trail of feed work, label deployments, and structural changes. Babysitting produces budget nudges.

Any agency worth hiring passes these easily - the questions only feel aggressive to providers with something to hide. If you are actively comparing providers right now, the Google Shopping agency chooser's guide runs the full filter list, fee models included.

Where to go from here

If you now suspect your account is being babysat, check the change history first. It takes five minutes and it does not lie: Google Ads, Tools, Change History, filter to the last 60 days. Count the feed-level and structural changes. If you see nothing but budget edits, you have your answer.

And if you want to see what feed-first management looks like on your own store, the Google Shopping page explains how we run it - and you can get your Merchant Center pulled up live, disapprovals first, before anyone talks about a contract.

Frequently asked questions

What does Google Shopping management include?

Five jobs. Feed engineering (title rewrites against real search terms, custom labels, attribute fixes), Merchant Center compliance (clearing disapprovals, policy alignment), campaign structure (which SKUs go into PMax, which into standard Shopping), bid strategy (target ROAS set per product tier, not one blended number), and monitoring (catching disapprovals and spend leaks daily, not at month end). If a service only covers the last two, it is campaign babysitting, not management.

How much does Google Shopping management cost?

Agencies charge flat retainers, hourly rates, or a percentage of ad spend. ZenoX charges a tiered percentage of monthly Google Ads spend: 10% on the first €10k, then 9%, 8%, 7%, and 6% above €150k - each rate only applies to the spend inside its own bracket, so the effective rate drops as you scale. No setup fee, no lock-in, and feed engineering plus Merchant Center fixes are included rather than billed as extras.

What is the difference between Google Shopping management and campaign babysitting?

Babysitting touches the campaign layer only: budget nudges, a bid tweak, a monthly report. Real management works the layer underneath - the product feed and Merchant Center. Titles get rewritten against real query data, custom labels sort SKUs by performance, disapprovals get fixed the day they appear. The test: ask for the account change history from the last 60 days. Babysitting shows a handful of budget changes. Management shows feed and structural work every week.

Do I need an agency to manage Google Shopping?

Not always. If you spend a few hundred euros a month, learning the system yourself is often the better investment - the fee would eat your margin. An agency starts to make sense around €5k/month in spend, when the feed work, disapproval handling, and structural decisions have real money riding on them and a mistake costs more than the fee.

How fast does Google Shopping management show results?

Feed changes propagate through the auction in about 3-7 days after Merchant Center approves the updated feed. At ZenoX the first 80 SKU title rewrites ship inside the first week, and a typical 500-SKU catalogue is fully reworked inside 30 days. Meaningful movement usually shows in the first 2-3 weeks as Smart Bidding recalibrates to the new match patterns.

Does Google Shopping management cover Performance Max?

It should. PMax pulls its Shopping ads from the same product feed, so feed engineering lifts both. Good management also decides which SKUs run through PMax and which run through standard Shopping - PMax under-bids long-tail queries, and standard Shopping campaigns catch the high-margin long tail it ignores. Running both typically lifts Shopping revenue 25-40% over PMax-only setups in catalogue-heavy niches.