Performance Max works for UAE stores that have a clean product feed, a real conversion signal and at least AED 8,000 to AED 10,000 a month to feed it. It wastes budget when it is launched with no conversion history, when brand terms are not excluded, or when someone edits it every few days. Run standard Shopping first, then move to PMax once you know which products actually sell.
The short version
- PMax needs conversion volume. Below roughly AED 8,000 a month it never learns.
- Exclude your brand terms or PMax will spend your budget buying customers you already had.
- Feed-only PMax is the safest way in. No images, no video, Shopping placements only.
- Six weeks before you judge it. Two weeks before you touch it.
Performance Max is either the best campaign in a UAE account or the most expensive way to buy traffic you already owned. There is not much middle ground, and which one you get is decided in the first fortnight.
Here is how we run it, and the specific ways we have watched it go wrong.
What Performance Max actually is
One campaign type that serves across Search, Shopping, YouTube, Display, Discover, Gmail and Maps, with Google deciding the split. You hand over a product feed, some assets and a goal, and Google allocates the budget across all of it.
That is genuinely powerful when the machine has good signal. It is genuinely dangerous when it does not, because a campaign with nothing to optimise towards will find the cheapest clicks available and report them as a win.
When PMax prints money
| Condition | Why it matters |
|---|---|
| Clean product feed | Titles, images, GTINs, prices and availability correct. PMax is a feed campaign wearing a costume. A bad feed caps everything else. |
| 30+ conversions a month already | The algorithm needs volume. Without it, PMax is guessing with your money. |
| AED 8,000 a month or more | Below that, the budget spreads across too many placements to learn any of them. |
| Accurate conversion values | Purchase value passing in AED, not a flat value per order. Value-based bidding with no values is just bidding. |
| A product range | PMax needs something to choose between. Single-product stores usually do better on Search plus Shopping. |
Hit all five and PMax will usually beat what you were running before. Here is one running properly.
Note the status in that screenshot: Limited by budget. That is the state you want to reach. It means the campaign can spend more profitably than you are letting it.
When it quietly wastes money
It eats your brand. This is the big one. PMax will take the cheapest conversions on offer, and nothing is cheaper than someone typing your own store name. Reported ROAS looks excellent. Incremental revenue does not move. Add brand terms as account level negatives, or run brand Search separately and keep PMax off it.
It drifts into Display. If you upload a full asset group with images and video before you have proven the Shopping side, PMax will find cheap display impressions and spend there. Start feed-only: no images, no headlines, no video. Shopping placements only. Add assets later, deliberately.
It gets edited to death. Every meaningful change restarts learning. We have inherited accounts where the budget was adjusted twice a week for two months. The campaign never left the learning phase once.
It hides which products are working. PMax reporting is thin by design. Use the listing group report and the search terms insights, and check them weekly, or you will not notice that 80 percent of spend went to three products with the worst margins in the catalogue.
The setup we use on a new UAE account
Week 0. Fix the feed first. Nothing else matters if titles are wrong. Then verify conversion tracking end to end, with purchase value and AED currency passing correctly.
Week 1. Launch brand Search as its own campaign, cheap, exact match. Launch feed-only PMax with brand excluded. If the store also sells in a physical shop, get offline conversions wired in now rather than later, because PMax optimises towards whatever you feed it.
Weeks 2 to 3. Hands off. Watch, do not touch. Note which products get impressions.
Weeks 4 to 6. First real read. Add an asset group with images and headlines only if the Shopping side is working. Start excluding the products that consume budget and return nothing.
Week 7 onwards. Scale by raising the target return slowly, or by raising budget while holding target. Not both in the same week. What that spend should be returning is covered in what AED 20,000 a month should get you.
On a fashion account we launched PMax with a full asset group on day one because the creative was ready and it felt wasteful not to use it.
Three weeks later, most of the spend was on Display and YouTube, cost per conversion had roughly doubled, and the Shopping side had never been given a fair run. We rebuilt it feed-only and the account recovered inside a month. Now we always start feed-only, even when good creative is sitting there.
Is Performance Max working or just taking credit?
Send us the account. We will tell you what share of your PMax conversions are brand searches you already owned, and what the campaign is actually adding on top of that. It is usually the most uncomfortable and most useful thing you will read that month.
Get a free PMax reviewSee how we run Google Shopping and PMax for UAE stores.
Frequently asked questions
Is Performance Max good for small UAE stores?
It works if you have a clean product feed, a real conversion signal and at least AED 8,000 to AED 10,000 a month to give it. Below that the campaign never gathers enough conversions to optimise and it will quietly spend most of the budget on cheap display placements.
Should I run Performance Max or standard Shopping?
Run standard Shopping first if you are new, because you can see exactly which products and search terms produce sales. Move to Performance Max once you know what works and you want the extra reach. Going straight to PMax with no history is how people lose a quarter.
Why is Performance Max eating my brand searches?
Because it will happily take the cheapest conversions available, and your own brand name is the cheapest one there is. Add your brand terms as account level negatives on the PMax campaign, or run a separate brand Search campaign and exclude the brand from PMax.
What is a feed-only Performance Max campaign?
A PMax campaign with no images, headlines or videos attached, so Google can only serve Shopping placements. It behaves much more like classic Shopping and it is a good way to get PMax reach without letting it wander into display and YouTube before you are ready.
How long before Performance Max settles?
Give it six weeks minimum, and do not touch it for the first two. Every meaningful edit restarts the learning. Most of the failures we inherit are campaigns that were adjusted every three days.