Are you doing SEO or Ads?
Get a free consultation with us
Stop wasting money on ads that are not working.
Get advertising that is intentional and measured properly.
- Reach your ideal target audience.
- Maximise ad spend efficiency.

Should I hand my whole budget to Performance Max is the question I have been asked more than any other in paid search, and the answer has not changed in three years. Performance Max is neither a silver bullet nor a black box. It is a campaign type that reaches inventory your keyword campaigns never touched, and it performs exactly as well as the goals and assets you feed it. Loose goals produce cheap conversions that look like success until someone checks whether the enquiries were worth answering. I have spent 13 years managing paid search for Australian businesses. The accounts that get a return from Performance Max made three decisions before they turned it on: they fixed conversion tracking, they wrote down which conversions represent real value, and they kept their keyword campaigns running alongside it. It is a single campaign that can serve across every Google surface, driven by goals rather than keywords. Google's own description is the clearest place to start. Google Ads Help on Performance Max states that "Performance Max is a goal-based campaign type that allows you to access all Google Ads inventory, including YouTube, Display, Search, Discover, Gmail, and Maps, from a single campaign." That is the whole idea in one sentence. You supply assets and a goal, and the system decides which combination runs where. It replaced Smart Shopping and Local campaigns and absorbed their function, so a retail account that used to run Smart Shopping has effectively been migrated into it. What stayed the same is more important than what changed. Google still combines your inputs with its own systems. Google Ads Help notes that the assets and settings you provide, such as images, data feeds and bidding goals, are combined with Google AI to serve across all available inventory and optimised automatically towards your goals, to help drive the best conversions and value. Read that sentence carefully, because it is doing a lot of work. Your inputs set the ceiling. The system cannot buy better conversions than the goal you handed it describes. The campaign optimises towards whatever you count as a conversion. If you count page views, or count every form submission as equally valuable, it will buy the cheapest version of that action, and the cheapest version is rarely the one that pays your invoices. Decide before launch which events represent real value. A booked consultation, a completed quote request and a phone call over sixty seconds are different things, and they should not share a value. If you cannot separate them in tracking, fix that first. A campaign pointed at a loose goal does not fail loudly, it succeeds at the wrong thing. Every campaign needs an asset group, and the minimum is only a starting point. Google's developer documentation on Performance Max asset requirements states that "Each AssetGroup in a Performance Max campaign requires an initial minimum set of assets, which can be existing or new." The same page notes that retail Performance Max campaigns linked to a product feed generate the minimum required assets automatically, and that uploading additional assets is recommended for maximum reach. Treat the minimum as the start line rather than the target. Bring several headlines, descriptions, images and video, and keep each asset group tight enough that a viewer could describe what it is selling in four words. Mixed themes in one asset group are the most common cause of an account that spends steadily and converts nothing. Spend can land on any of the surfaces listed above, and the split is not something you set directly. There is no line in the interface that says put 60 per cent here and 40 per cent there. The system allocates towards whichever combination it believes will hit your goal at the lowest cost. What you can influence is the assets you supply and the goal you choose. Provide video and the campaign can use YouTube. Provide a feed and Shopping becomes available. Give it nothing but a headline and two images and you have quietly restricted where it can run. In a Search campaign, every click has a query attached and you can see exactly what someone typed. In Performance Max, much of the reporting arrives grouped or unlabelled. That is not the platform hiding from you, it is the nature of a campaign that runs across surfaces where there is no query in the same sense. This is why comparisons between the two campaign types go wrong. Judging Performance Max on the same metrics you use for Search, such as average cost per click or impression share, tells you very little. Judge it on what it produced, and judge that against the cost of producing it another way. Search themes let you signal the topics and queries the campaign should treat as relevant, which is the closest thing to keyword targeting inside the campaign type. Brand exclusions stop the campaign bidding against your own brand traffic. Account level negative keywords keep it away from queries you know are worthless. Those three controls are worth setting before you scale spend, not after. They are also the controls most accounts have never touched, which is usually why a Performance Max campaign appears to be competing with the Search campaigns sitting beside it. Final URL expansion lets the system choose a different landing page from the one you nominated. Google describes the behaviour precisely. Per Google Ads Help, "If you turn on Final URL expansion, Google may replace your Final URL with a more relevant landing page based on the user's search query, and generate a dynamic headline and description to match your landing page content." That can be useful on a large site with many strong pages, and it can send traffic to a page you would never have chosen deliberately. If you sell services that should not be sold interchangeably, decide this consciously and review the landing page report rather than leaving the default in place. Google protects keyword control in a specific and documented way. Google Ads Help states that Search campaigns containing an exact match keyword are given priority to serve over Performance Max when a query hits an exact match keyword. In practice, this means your exact match terms are still yours to own. That protection is why I never recommend pausing the Search campaigns that already convert. They hold the highest intent queries at the lowest risk, and they give you a stable baseline to compare the new campaign against. The protection is not absolute. If your Search campaign is limited by budget and cannot serve, the query can still be picked up elsewhere. This is the honest version of the common complaint that Performance Max is stealing brand searches. The fix is not to pause anything. Make sure your best converting terms are covered by exact match keywords in a Search campaign with enough budget to actually serve, then let the two run side by side. If you want that structure checked before you increase spend, a Google Ads audit before you scale spend is the cheaper order of operations. Conversion count on its own is the least useful number in the account. A campaign can double its conversions while halving the value of the leads, and the dashboard will look excellent throughout. Judge on conversion value where ecommerce is involved, and on lead quality where it is not. For lead generation, that means taking a month of enquiries, marking which ones became quotes, which became customers and which were never going to, and then asking which channel produced them. Most businesses have never done that cross check. It is the single most useful hour you will spend on a paid account. Two reports carry most of the signal. Asset group reporting tells you which set of assets is being served and how it performs, which is where you find out that one group is carrying the campaign while two others quietly consume spend. The channel report gives you the rough split across the surfaces and is best read as a trend across months rather than a precise accounting. Look for the shape of the line rather than the decimals. A channel going from nothing to a third of spend in a month is a change worth understanding, regardless of what the cost per conversion says. The first failure looks like a win in every report. Costs fall, conversions rise, and the only person who notices something is wrong is the one answering the phone. It happens when the goal is set loosely, such as counting every form submission the same, so the system optimises towards the cheapest submission it can find. The tell is a mismatch between conversion count and sales conversations. If conversions are up and quotes are flat, the campaign is buying activity rather than demand. The second failure is subtler. Performance Max can serve on brand searches, including your own business name, and claim credit for people who were always going to buy. Those conversions are real, but they were not created by the campaign, and counting them as growth flatters the report and misleads the next decision. The way to test it is simple: turn on brand exclusions, watch the conversion count, and see how much of the growth survives. That comparison is uncomfortable and worth doing once a quarter. Four things, in fifteen minutes. Spend against plan for the week. Conversion count and value, separated so a cost drop and a value drop cannot hide each other. Search terms and search themes, looking for spending on anything irrelevant. And the landing page report if final URL expansion is on. Write the numbers down somewhere you can see the previous week. A single week tells you almost nothing, and four weeks in sequence tells you almost everything. Leave the goal alone until the campaign has enough data to leave its learning phase. Changing a goal resets that clock, and businesses that tinker weekly never accumulate a fair test. If the goal is accurate and the campaign underperforms, split by theme or by product category rather than by inventing a new conversion action. Where a decision needs to rest on what the spend actually returned, measuring return rather than clicks puts the comparison on a footing that survives scrutiny, and working out what a realistic monthly test budget looks like first stops the test from being decided by whatever was left in the account. The accounts I manage are run through Google Ads management that keeps a close eye on where the money goes, and the pattern that holds is patience on goals and discipline everywhere else. No, and there is no reason to treat it as either or. Performance Max works best alongside keyword campaigns that already convert, because the Search campaigns keep control of your highest intent queries while Performance Max finds demand you were not bidding on. Keep a measured share of budget on the campaign type you understand best until you have enough data to compare. Yes, and it needs to be accurate rather than merely present. The campaign optimises towards whatever conversions you count, so counting page views, or counting every form submission as equally valuable, teaches it to buy the cheapest version of that action. Decide which conversions represent real enquiries or real sales value before you turn it on. More than the minimum, in practice. Google's own documentation states that each asset group requires an initial minimum set of assets, which can be existing or new, and that uploading additional assets is recommended for maximum reach. Treat the minimum as the start line. Bring a range of headlines, descriptions, images and video, and keep them relevant to the theme of that asset group. Google's own documentation is specific here: Search campaigns containing an exact match keyword are given priority to serve over Performance Max when a query hits an exact match keyword. If your Search campaign is limited by budget, Performance Max may still serve on exact match terms occasionally. The practical fix is to make sure your best converting terms are covered by exact match keywords in a Search campaign with enough budget to serve. It is a trade off you should make deliberately rather than by default. When it is on, Google may send traffic to a more relevant landing page than the one you nominated and generate headlines and descriptions to match that page. That can help on a large site with many relevant pages, and it can misfire if you sell products or services that should not be promoted interchangeably. Review where the clicks actually landed. Give it enough data to leave the learning phase before drawing conclusions, and judge on conversion value and lead quality rather than conversion count. A campaign can look strong while filling your pipeline with enquiries you will never sell to. Check a month of enquiries by source and by quality before deciding to scale, split or pause. Are you doing SEO or Ads? Stop wasting money on ads that are not working. Get advertising that is intentional and measured properly.What Performance Max actually is
One campaign across Google's ad inventory
What it replaced and what stayed the same
What it needs from you before it can run
Conversion tracking you can trust
Assets, asset groups and product feeds
Where the money goes
Search, Shopping, YouTube, Display, Discover, Gmail and Maps
Why channel reporting looks different to Search
The controls you still hold
Search themes, brand exclusions and negative keywords
Final URL expansion and automatically created text
Why exact match Search still gets priority
What happens when a query matches exactly
When Performance Max takes the query instead
How to judge a Performance Max campaign honestly
Conversion numbers, conversion value and lead quality
Asset group and channel reporting worth reading
The failure modes that look like success
Cheap conversions from the wrong audience
Brand traffic taking credit for demand you already had
A plain English 30 day review routine
What to check weekly
When to change goals, split campaigns or wait
Performance Max questions I get asked most
Should I move my whole budget into Performance Max?
Do I need conversion tracking set up first?
How many assets do I actually need to supply?
Why does my Search campaign sometimes lose a query to Performance Max?
Is Final URL expansion safe to leave on?
How soon should I judge the results?
Get a free consultation with us





Latest Blog

WINNER

FINALIST

FINALIST

WINNER

WINNER

FINALIST

FINALIST

WINNER

Xugar Holiday Offer