Should I switch to Performance Max?
Google keeps saying “move to Performance Max” — should I? The rep suggests it, the panel nudges it, the agency is curious — and the decision belongs to your account. 🧭
PMax is not a bad product. Opened at the wrong time on the wrong ground, it becomes expensive and makes what works invisible.
Short answer: PMax is a multiplier in a ready account and a blackout in an unprepared one. Three conditions come first. ⚖️
What is the logic of PMax?
One campaign, all the inventory: search, shopping, display, video, maps. 🌐
Three conditions before switching
BU BÖLÜMÜN ÖZETİ
- Condition 1: clean conversion tracking
- Condition 2: enough conversion volume
- Condition 3: brand traffic separated
All three are non-negotiable. 🔒
Condition 1: clean conversion tracking
It must be clear which conversion turns into money. Fed a weak signal, PMax grows the wrong thing; setup sits in the tracking guide. 🎯
Related reading from the archive: Performance Max and AI · shopping feed optimisation.
Condition 2: enough conversion volume
With a low monthly conversion count the algorithm cannot learn. In that case, building volume with search comes first. 📊
Condition 3: brand traffic separated
PMax pulls brand searches in and inflates the report. Without a separate brand campaign you get a success illusion; the logic sits in the brand defence guide. 🏷️
How should the switch be made?
Not overnight, but under control. 🎚️
How do you keep it under control?
The black box is not entirely closed. 🔦
Four common mistakes
BU BÖLÜMÜN ÖZETİ
- Mistake 1: closing the search campaigns
- Mistake 2: not separating brand traffic
- Mistake 3: launching with weak assets
- Mistake 4: deciding in two weeks
All four produce the same outcome: unmeasurable spend. 🚧
Mistake 1: closing the search campaigns
Surrendering control also loses your clearest data. The search campaign is your reference point. 📉
Mistake 2: not separating brand traffic
The report looks brilliant and revenue does not move. A “PMax is great” verdict reached without separation is wrong. 🎭
Mistake 3: launching with weak assets
One headline, low-resolution images, no video. An account that gives the system little material gets little back. 🧱
Mistake 4: deciding in two weeks
Closing before learning completes throws the investment away. The decision window is four weeks minimum. ⏱️
The decision table
Three questions, one answer. 🧪
📝 Notes From the Field
In one e-commerce account PMax looked miraculous: cost per conversion was half the rest. We separated brand searches. The entire improvement came from brand traffic; new-customer numbers had not moved. PMax was not bad — the report was misleading.
📖 Short Glossary
Asset group: the set of text, image and video you give PMax. Search themes: topic hints that steer the system. Brand share: the proportion of your own brand searches entering the campaign. Learning period: the first weeks while the campaign calibrates.
⚡ Quick Summary
PMax multiplies a ready account and blacks out an unprepared one. 🧭 Three conditions: clean measurement, enough conversion volume, separated brand traffic. Switch in parallel and keep search campaigns open. Decide after four to six weeks, with brand share separated out.
🎯 Next Step
Let us measure whether the three conditions hold in your account: see the digital audit. To talk it through use the consult your expert form; the budget decision sits in the budget guide.
Frequently Asked Questions
Sık Sorulan Sorular
It manages the channels together using the assets you supply — text, images, video, feed — and distributes toward a conversion goal. The system picks the channel, not you. ⚙️
Visibility is thin: you cannot see which channel or query earned the result with the clarity classic campaigns give. The management frame sits in the PMax guide. 🕶️
E-commerce with a product feed, broad inventory, enough conversion volume and clean measurement. With those four, PMax usually pays. ✅
Low-conversion, single-service accounts with weak measurement. There the system cannot learn and spends the budget exploring. 🚧
Complete them in order: measurement, then volume, then brand separation. Only then does PMax become meaningful. 🪜
No: working search campaigns are preserved. PMax opens in parallel and is tested with part of the budget. ⚖️
Rich and accurate: multiple headlines, descriptions, quality images, video where possible and a current feed. Weak assets mean weak results. 🎨
From current performance: the target cost per conversion starts near today’s reality. An unrealistic target stalls the system. 📏
Four to six weeks minimum: a PMax closed sooner has only paid the learning cost. ⏳
Asset group performance, search themes and product-level results. Limited, but enough to produce decisions. 📋
With account-level negatives and by keeping the brand campaign separate. Those two settings let the report show the truth. ✂️
After four to six weeks, if cost per conversion sits well above target and does not improve even after brand is separated. 🔴
Cost below target with total conversions rising. Growth is gradual; the logic sits in the budget guide. 🟢
If not, the answer is no — tracking first. This item is absolute. 🎯
If not, build volume with search first. PMax feeds on volume; it does not create it. 📊
If not, separate it — otherwise the report means nothing. Three yeses make a controlled switch sensible. ✅
Let us measure: a digital audit writes down where the three conditions stand. To talk it through use the consult your expert form; the service sits on the Google Ads consulting page. 🎯
PMax uses the platform’s inventory as a whole, so it is widely recommended. The recommendation is not made in bad faith, but it does not know your account’s measurement and volume position; the decision stays with you.
It can, but without a feed and with low volume the return usually sits below a search campaign’s. In that structure, search comes first and PMax is tried afterwards if needed.
Where your search campaign is strong on a query, it generally takes priority. Even so, account-level negatives and a separate brand campaign reduce overlap and report contamination.