Performance Max Audience Exclusions in 2026: What Malta Advertisers Should Change

For three years the standard complaint about Performance Max was the same: it spends your money and refuses to tell you where. In 2026 Google finally started answering that complaint. The platform now offers first-party audience exclusions, budget projections inside the campaign, expanded demographic reporting, and placement reports segmented by network. None of it turns Performance Max into a manual campaign — but together these changes decide whether the channel earns its budget or quietly burns it.

For advertisers in Malta the stakes are higher than they are for a UK or US account. On an island of roughly 560,000 people, your addressable audience is small enough that Performance Max will exhaust genuinely new prospects within weeks and start re-serving people who already know you. That is the specific failure mode audience exclusions were built to fix. If you are running Google Ads in Malta, this is the most consequential set of changes to hit the platform this year.

What actually changed in Performance Max in 2026

Google grouped the updates into two buckets: steering and reporting. The steering change is a genuine new lever. The reporting changes are the ones you will use every week.

First-party audience exclusions

You can now exclude specific Customer Match lists from a Performance Max campaign. Previously you could signal that a list was valuable; you could not tell the system to stay away from it. Now, if your goal is net-new customer acquisition, you can remove existing customers from the pool entirely.

Two caveats matter. First, this is directional steering, not hard targeting — Performance Max still decides how literally to interpret your input. Second, the feature is only as good as your data. An outdated or badly segmented customer list will produce exclusions that either do nothing or block the wrong people.

Budget reporting inside the campaign

The budget report now lives directly inside the Performance Max campaign rather than at account level. It forecasts end-of-month spend and models how changing your daily budget affects projected performance. For anyone reporting monthly numbers to a board, a franchise owner, or a client, this removes a recurring guessing exercise.

Expanded audience reporting

Google added demographic and segment-level breakdowns, including age range and gender. You can finally see who the system is actually reaching, rather than inferring it from conversion totals. This is the report that most often reveals a mismatch between the customer you think you are buying and the customer Google is delivering.

Placement reporting segmented by network

Placement reports can now be split by network — Search, Shopping, Display, YouTube, Discover, Gmail and Maps — under the "When and where ads showed" tab. For brand safety this is overdue. For budget diagnosis it is the single most useful addition of the year, because it exposes the classic Performance Max pattern of Display and YouTube absorbing spend while Search and Shopping do the converting.

Why audience exclusions matter more in Malta than almost anywhere else

Market size changes the maths. A Performance Max campaign in Germany can run for months before it starts recycling the same users. In Malta, a campaign with a reasonable budget can cycle through the relevant portion of the population in a matter of weeks.

What happens next is predictable. Frequency climbs, cost per acquisition stays flat or improves slightly, and the campaign looks healthy — because it is converting people who already knew the brand and were likely to buy anyway. That is not growth. That is retargeting wearing an acquisition costume, and it is the reason so many Maltese accounts report strong Performance Max ROAS alongside stagnant revenue.

Audience exclusions give you a way to test this directly. Exclude your full purchaser list, run the campaign for a defined window, and watch what happens to cost per acquisition. If CPA rises sharply, you have learned that a meaningful share of your reported conversions were existing customers. That is uncomfortable but useful. If CPA holds, Performance Max is genuinely finding new demand and deserves more budget. This kind of structured testing is the foundation of any serious paid media strategy in Malta, and the new controls make it possible without a third-party tool.

There is a second Malta-specific consideration. Many local businesses run bilingual creative across English and Maltese, and some serve both the resident market and inbound visitors. The expanded demographic and network reporting makes it far easier to see which of those audiences a campaign is actually reaching — particularly when Discover and Gmail placements pull in users whose intent is very different from someone searching on Google.

How to set up audience exclusions without breaking your account

The temptation is to exclude everything at once. Resist it. Work through this sequence instead.

1. Audit your Customer Match lists first. Check recency, size and segmentation. A single list called "All Customers" containing five years of data is not usable. Split it into purchasers from the last 90 days, purchasers from the last 12 months, and lapsed customers — those groups behave completely differently and deserve different treatment.

2. Exclude the narrowest list you can justify. Start with recent purchasers only. Excluding lapsed customers is usually a mistake, because they are often your cheapest reactivation opportunity.

3. Hold the change for a full conversion cycle. If your sales cycle is 21 days, do not evaluate the test at day seven. Small markets produce noisy data, so give the campaign enough volume to say something meaningful.

4. Watch new customer share, not just CPA. The point of the exclusion is to change who you acquire. If new customer share does not move, the exclusion is not doing what you intended — usually a data quality problem rather than a platform problem.

5. Keep a retention channel running. If you exclude existing customers from Performance Max, something else has to reach them. Email, organic social, or a dedicated remarketing campaign should absorb that job before you flip the switch.

Reading the new network reports without fooling yourself

Network-segmented placement data is genuinely valuable, but it invites a specific analytical error: assuming the channel credited with the conversion is the channel that caused it.

Performance Max attributes on a last-click-within-the-campaign basis. If YouTube introduces a prospect and Search closes them a week later, the report will hand the credit to Search and make YouTube look wasteful. Cut YouTube on that basis and Search performance often degrades a month later, for reasons that appear unrelated.

Treat the network report as a diagnostic, not a verdict. Use it to spot the obvious problems — a placement serving on content irrelevant to your business, or Display consuming 40% of budget with negligible assisted contribution — and use holdout testing or geo splits to answer causal questions. Attribution accuracy is one of the harder problems in growth strategy for Maltese businesses precisely because sample sizes are small enough that a single week of noise can look like a trend.

What this means if you run lead generation rather than ecommerce

Performance Max has historically been weaker for lead generation than for ecommerce, and the reason is feedback quality. Ecommerce sends back a purchase and a value; lead generation typically sends back a form fill, which the system optimises toward regardless of whether that lead was any good.

Audience exclusions help here, but the bigger lever is still offline conversion import. Feed qualified-lead and closed-won signals back into Google Ads and the algorithm optimises for revenue instead of form volume. Combine that with an exclusion list of existing clients and you have a campaign structure that is actually pointed at new business — which is the whole point of lead generation in Malta, where a handful of high-value clients often matters more than raw enquiry count.

The new budget report is quietly useful for service businesses too. If your capacity is fixed — a clinic with a set number of appointment slots, a contractor with two crews — projected month-end spend lets you throttle campaigns before you buy demand you cannot service.

A practical checklist for the next 30 days

If you manage a Performance Max campaign in Malta, work through this list in order:

Week one. Export your Customer Match lists and check upload recency. Rebuild segmentation if it is a single undifferentiated list. Open the network-segmented placement report and record the current spend split across Search, Shopping, Display, YouTube, Discover, Gmail and Maps.

Week two. Review the expanded demographic report. Compare the age and gender profile Google is reaching against your actual customer base. Note any gaps — this is often where the largest inefficiency hides.

Week three. Apply a recent-purchaser exclusion to one campaign only. Leave a comparable campaign untouched as a control. Document your starting CPA, new customer share and frequency.

Week four. Set up the budget report as a recurring monthly check. Review the exclusion test, but do not act on it yet unless something has gone obviously wrong — you need a full conversion cycle before the numbers mean anything.

Frequently asked questions

Are Performance Max audience exclusions available in Malta?

Yes. The feature rolls out at account level rather than by country, so Maltese advertisers get access on the same schedule as everyone else. You will need a Customer Match list that meets Google's minimum size threshold, which is the more common blocker for smaller Maltese accounts than availability.

Will excluding existing customers reduce my Performance Max ROAS?

Very likely, at least initially — and that is often the correct outcome. Reported ROAS falls because you have removed cheap conversions from people who would have bought anyway. The number that matters is new customer acquisition cost and total revenue, not campaign-level ROAS in isolation.

What is the minimum Customer Match list size for exclusions?

Google requires a matched list of at least 1,000 users before it can be applied. Many small Maltese businesses fall short of this on a 90-day purchaser list, in which case you will need to widen the window to 12 months or combine lists to reach the threshold.

Should I still run separate Search campaigns alongside Performance Max?

In most cases, yes. Brand and high-intent generic terms belong in a standard Search campaign where you control bids and match types. Performance Max should handle discovery and the long tail. In a market as small as Malta, giving up bid control on your most valuable terms is rarely worth the convenience.

How long should I wait before judging an audience exclusion test?

Allow at least one full conversion cycle plus two weeks, and a minimum of 30 conversions in each group. Maltese account volumes are typically low enough that shorter tests produce results driven by noise rather than by the change you made.

The bottom line

Performance Max in 2026 is not a different product — it is the same automated system with better instrumentation and one meaningful new control. That is enough to change how you should manage it. Audience exclusions let you test whether the channel is buying new customers or repackaging existing ones, and network-level reporting shows you where the money physically goes.

In a market the size of Malta, both questions have a short answer window before audience saturation makes the data ambiguous. Run the test now, while your account still has a clean before-and-after to compare.