Most quotes for Performance Max arrive as a single number and a fee model. Fifteen per cent of spend. RM 2,500 a month. A base fee plus a slice above RM 10,000. The number is easy to compare, so that is what business owners compare.
It is the wrong comparison. Two agencies can quote the same figure and sell completely different jobs, because Performance Max hides where the work actually is. There are no keywords to write and no placements to pick. Google decides where your ads run, and you influence that decision through inputs — your product feed, your creative assets, your conversion data, your exclusions.
So the fee question is really a scope question. What does this person touch every month, and what happens to my account when nobody touches it? A cheap fee attached to a thin scope is not cheap. It is a slow leak with an invoice attached.
This guide sets out what Performance Max management in Malaysia should actually cover. It walks through how the common fee models behave at different spend levels, where the crossover points sit, and what to ask before you sign. It sits under our Performance Max and Shopping consulting work at IZILI Digital Marketing.
Before the fee models, here is a walkthrough of what optimising these campaigns involves in 2026.
The RIGHT Way to Optimise Your Performance Max Campaigns in 2026
Source video: Define Digital Academy on YouTube
PART 1 · DIAGNOSE
What You Are Actually Paying a Performance Max Manager For
IN BRIEFYou are paying for input quality, not for bid management. Performance Max sets its own placements and bids. So Performance Max management in Malaysia earns its fee on four inputs instead: a clean feed, fresh assets, honest conversion data, and tight exclusions — the things that actually steer the campaign.
In a Search campaign, you can see the labour. Keywords added, negatives applied, ad copy tested, bids adjusted. The account shows its workings, and a client can scroll the change history and feel the money.
Performance Max removes almost all of that surface. What remains is a smaller set of controls that matter more, because each one feeds the system rather than nudging it.
- The product or service data. For retail, Google’s Merchant Center product data specification sets which attributes are required — miss them and the product simply cannot serve.
- The creative assets. Google recommends filling every asset group with up to 15 headlines, 5 descriptions, 20 images and 5 videos, per its asset group best practices. Thin asset groups narrow where you can appear.
- The conversion signal. The campaign optimises towards whatever you told it to value. Tell it to value a page view and it will buy page views beautifully.
- The exclusions. Brand terms, existing customers, placements and account-level negatives decide how much budget goes to demand you already own.
Retailers feel this most sharply, because the feed carries the campaign in the same way it does for Google Shopping ads in Malaysia. Lead-generation accounts feel it through conversion quality instead — the form fills that never answer the phone.
Not sure whether your quote covers the work that matters?
We map the scope against your account before you commit to anyone. See how our Blueprint sessions run
BENCHMARK BRIEFING 1 OF 4
How Do Agencies Charge for Performance Max Management?
IN BRIEFFive fee models cover almost every quote you will see: flat retainer, percentage of spend, hybrid base plus percentage, hourly or project, and performance-based. None is inherently fairer. Each simply moves risk between you and the agency, which is why the model should follow your budget shape.
The table below models what each structure would cost at a monthly media spend of RM 10,000, so the shapes can be compared side by side rather than argued about in the abstract.
| Fee model | How it is charged | Modelled fee (RM) | Where it fits |
|---|---|---|---|
| Flat retainer | Fixed monthly fee, spend-independent |
2,500 |
Large catalogues, steady budgets |
| Percentage of spend | A set share of media spend |
1,500 |
Smaller or seasonal budgets |
| Hybrid base + share | Fixed base plus a smaller share above it |
2,000 |
Budgets that move month to month |
| Hourly or project | Billed by hours or a fixed deliverable |
1,800 |
Rebuilds, audits, feed clean-ups |
| Performance-based | Fee tied to revenue or qualified leads |
2,200 |
Mature, trusted tracking only |
Illustrative model by IZILI Digital Marketing, based on published fee conventions and Google Ads documentation, 2026. Licence.
Notice how narrow the spread is. At one spend level, every model lands within roughly RM 1,000 of the others. The model is not where your money is won or lost — the scope is.
PART 2 · DIAGNOSE
Where Performance Max Fees Quietly Go Wrong
IN BRIEFFees rarely fail because the number was too high. They fail because the scope excluded whatever later broke — usually feed maintenance or creative refresh — and those exclusions only surface once performance slides. Ask what is out of scope before you ask what is in the package.
Three exclusions cause most of the disputes we see in Malaysian accounts, and all three are invisible in month one.
- Creative production sits outside the fee. The manager can rotate assets but cannot make new ones, so the same five images run for eight months and reach quietly narrows.
- Feed work is billed separately. The catalogue drifts, disapprovals accumulate, and each fix becomes a small quote nobody wants to approve.
- Tracking is treated as the client’s problem. Nobody owns the conversion definition, so the campaign optimises towards a signal no one has audited since setup.
There is a second, subtler failure. A percentage fee rewards spending more, and Performance Max is unusually good at absorbing extra budget without complaint. If your manager’s income rises with your spend, agree in advance what evidence would justify a budget increase — and what would trigger a cut.
BENCHMARK BRIEFING 2 OF 4
Which Parts of the Job Take the Most Time Each Month?
IN BRIEFFeed and asset work absorb close to half the monthly effort on a retail Performance Max account, and reporting absorbs the least. That ratio is the fastest test of a proposal: if the hours are weighted towards reporting, you are buying observation rather than intervention.
The model below spreads a 24-hour monthly workload across the six workstreams that keep an account healthy, and names what breaks when each is skipped.
| Workstream | Share of effort | Hours | What breaks if skipped |
|---|---|---|---|
| Feed and Merchant Center health | 6 | Disapprovals spread; products stop serving | |
| Asset production and refresh | 5 | Ad strength falls; eligible reach narrows | |
| Conversion tracking and data quality | 4 | The system optimises to the wrong action | |
| Search themes and exclusions | 3 | Budget leaks into brand and junk queries | |
| Budget, bidding and target reviews | 3 | Spend drifts away from the target return | |
| Reporting and decisions | 3 | Drift goes unnoticed for a full quarter |
Illustrative model by IZILI Digital Marketing, based on Google Ads and Merchant Center documentation, 2026. Licence.
Google itself advises waiting two to three weeks before judging and replacing weak assets. That single guideline sets the rhythm of the job: fortnightly attention, monthly decisions, quarterly rebuilds.
PART 3 · DESIGN
What Belongs Inside the Scope, and What Does Not
IN BRIEFPut everything that changes weekly inside the retainer and everything that changes yearly outside it. Feed maintenance, asset rotation and exclusions belong in scope. Brand photography, website rebuilds and choosing your content language are projects, and pricing them as retainer work inflates the fee for everyone.
A clean scope has three tiers, and confusion between them is what produces both padded retainers and surprise invoices.
- Continuous work — inside the fee. Feed error monitoring, disapproval fixes, asset rotation, exclusion lists, budget and target reviews, and the monthly decision meeting.
- Periodic work — inside the fee, scheduled. Quarterly asset group rebuilds, seasonal campaign structures, and an annual conversion tracking audit.
- Project work — outside the fee, quoted. New video production, a landing page build, a catalogue restructure, or a migration between Merchant Center accounts.
Malaysian advertisers running multilingual assets should decide the language question at scope stage, not later. Producing headlines in English and Bahasa Malaysia doubles the creative load, and a retainer priced for one language will not quietly stretch to cover two.
BENCHMARK BRIEFING 3 OF 4
What Do You Effectively Pay Per Ringgit of Spend?
IN BRIEFA flat retainer of RM 2,500 costs 83% of a RM 3,000 budget and 5% of a RM 50,000 one. The crossover against a 15% fee sits near RM 16,700 in monthly spend. That single number should decide how you structure Performance Max management in Malaysia.
The model below tracks the effective management cost as a share of media spend across five budget tiers, holding the three common structures constant.
| Monthly spend | Flat retainer, as % of spend | Flat | 15% model | Hybrid |
|---|---|---|---|---|
| RM 3,000 | 83% | 15% | 48% | |
| RM 6,000 | 42% | 15% | 28% | |
| RM 12,000 | 21% | 15% | 18% | |
| RM 25,000 | 10% | 15% | 13% | |
| RM 50,000 | 5% | 15% | 10% |
Illustrative model by IZILI Digital Marketing: flat RM 2,500; 15% of spend; hybrid RM 1,200 plus 8%. Licence.
The lesson runs both ways. Small advertisers who insist on a flat retainer are paying a management premium that swallows the campaign. Large advertisers on a percentage fee are paying more every time they scale, for work that does not scale at the same rate.
PART 4 · DESIGN
Choosing a Fee Model Without Overpaying
IN BRIEFPick the model that matches how your budget behaves. Steady spend above roughly RM 17,000 favours a flat retainer, and spend under RM 10,000 favours a percentage. Anything in between usually lands on a hybrid, which is where most Google Ads engagements settle.
DECISION BOX · WHICH FEE MODEL TO ACCEPT
| Option | Cost predictability | Best spend range | Main risk |
|---|---|---|---|
| Flat retainer | High — fixed | Above RM 17k/mo | Crushing at low spend |
| Percentage of spend | Low — moves monthly | Under RM 10k/mo | Rewards spending more |
| Hybrid base + share | Medium | RM 10k–25k/mo | Two levers to negotiate |
| Performance-based | Low — outcome-linked | Any, with clean data | Arguments over attribution |
Verdict: Choose a percentage fee while spend is below RM 10,000 a month and still finding its level. Move to a flat retainer once spend is stable above roughly RM 17,000, where a percentage starts charging you for scale rather than for work. Only accept a performance-based fee if you would defend your conversion data in front of an auditor.
Want the fee maths run against your own numbers?
Bring your spend, margin and catalogue size, and we will model all three structures with you. Meet the consulting team
BENCHMARK BRIEFING 4 OF 4
Is Malaysia’s Online Market Big Enough to Carry the Fee?
IN BRIEFYes, and the consumer side is growing fastest. Malaysian business-to-consumer e-commerce grew 11.3% in 2024 against 7.6% for business-to-business, which is why a management fee is easier to justify for retail and lead-generation accounts than for slow-moving wholesale catalogues.
The figures below come from the Department of Statistics Malaysia. The 2023 column is derived from the published year-on-year growth rates rather than quoted directly.
| Transaction type | 2023 (RM bn) | 2024 (RM bn) | Change |
|---|---|---|---|
| Business-to-business | 817.5 | 879.6 | +7.6% |
| Business-to-consumer | 336.7 | 374.7 | +11.3% |
| Business-to-government | 30.3 | 33.8 | +11.4% |
| All transactions | 1,184.5 | 1,288.1 | +8.7% |
Aggregated by IZILI Digital Marketing from DOSM digital economy statistics, 2023–2024.
Demand is not the constraint. Shopper behaviour is not the constraint either — MCMC’s e-Commerce Consumers Survey found 78.3% of online shoppers buying on a smartphone, which is exactly the surface Performance Max serves. What decides whether the fee pays back is your own margin and catalogue quality.
PART 5 · DEPLOY
Questions to Ask Before You Sign
IN BRIEFSix questions separate a scope from a sales deck. They cover account ownership, creative production, feed responsibility, exclusion policy, reporting cadence and exit terms — the same six we work through when clients ask us to compare agencies in Kuala Lumpur.
- Who owns the Google Ads and Merchant Center accounts? If the agency owns them, you cannot take your history with you.
- How many new creative assets per quarter are included? A number, not “as needed”.
- Who fixes feed disapprovals, and within how long? Name the owner and the response window.
- What is your brand exclusion policy? A manager who refuses to exclude brand terms is buying credit for sales you already had.
- What triggers a recommendation to reduce spend? The answer tells you whether the incentive is aligned.
- What is the notice period, and what do I receive on exit? Feed files, asset files, and account access, in writing.
PART 6 · DRIVE
How to Judge the Fee After 90 Days
IN BRIEFJudge inputs at 30 days, stability at 60, and return at 90. If feed errors are down and asset groups have been rebuilt once by month three, the fee is working even when revenue is still catching up. It is the same sequence we use when testing newer ad formats like YouTube Shorts.
Set the review checkpoints at the start, so the quarterly conversation is about evidence rather than mood.
- Day 30 — inputs. Feed error count falling, conversion actions audited, brand exclusions live, first asset refresh shipped.
- Day 60 — stability. Cost per conversion settling within a predictable band; disapprovals no longer recurring on the same products.
- Day 90 — return. Contribution after media and management fee, measured against the break-even return you set before launch.
One caution on the final number. If your management fee is 15% of spend, your campaign must clear break-even plus that 15% before the engagement pays for itself. Advertisers who forget to include the fee in the break-even calculation routinely mistake a losing account for a marginal one.
FAQ
Frequently Asked Questions
1. How much should Performance Max management cost in Malaysia?
Quotes for Performance Max management in Malaysia usually cluster around a flat monthly retainer or roughly 10–20% of media spend. The right figure depends on catalogue size and how much creative work sits inside the fee. A 30-product catalogue with fresh photography needs far less monthly attention than a 3,000-product feed that changes weekly.
2. Is a percentage-of-spend fee fair for Performance Max?
It is fair while your budget is still finding its level, and less fair once it stabilises. The deciding factor is whether the work grows with the spend; feed and asset effort largely does not, so above roughly RM 17,000 a month a percentage starts charging you for scale instead of labour.
3. Does Performance Max need less management because it is automated?
No — it needs different management, not less. Automation removed bid and placement work but made input quality decisive, and Google’s own guidance to wait two to three weeks before replacing weak assets implies a continuous testing cycle rather than a set-and-forget campaign.
4. Should the ad spend sit inside the management fee?
Keep them separate in almost every case. A bundled figure hides how much is actually reaching the auction, and it depends on nothing except your own preference for simplicity — which is rarely worth losing visibility of the media-to-fee ratio.
5. How long should I commit before judging a Performance Max manager?
Ninety days is the fair minimum, with monthly checkpoints along the way. It depends on how much repair work the account needed at the start: an account with a broken feed and unaudited conversion tracking may spend its first six weeks simply getting to a clean baseline.
THE VERDICT
Buy the Scope, Not the Percentage
The fee model is the easiest part of this decision and the least important. Every structure lands within a narrow band at any given spend level, and the difference between a good engagement and a bad one is almost never the percentage.
What separates them is what someone touches each month. Feed health, creative supply, conversion data, exclusions — the four inputs the campaign actually reads. A quote that covers all four at a higher number beats a quote that covers one at a lower one, every time.
So do the work in this order. Write your three-tier scope, ask the six questions, then compare prices against the crossover maths for your own spend level. By then the choice usually makes itself, and you will know what you are buying rather than what you are being sold. If you want a second pair of eyes on the comparison, that is exactly what our Performance Max consulting sessions are for.
Holding two Performance Max quotes and unsure which is honest?
Book a free Blueprint consultation. We will diagnose what your account needs each month, design the scope and fee model around your spend level, and hand you a comparison you can take to any agency.