If you run Google Ads in South Africa, you have almost certainly been nudged to try Performance Max. It is Google's flagship automated campaign type, and it now sits at the centre of how the platform wants advertisers to spend. The promise is simple: one campaign that reaches the entire Google network and lets AI do the heavy lifting. The reality is more nuanced, and understanding it is the difference between Performance Max being a growth engine or a budget drain.

How Performance Max works

Performance Max is a goal-based campaign: you tell Google what a conversion is worth, supply raw materials, and the AI builds and places ads to win those conversions across the whole network. Instead of choosing keywords and placements yourself, you hand Google the inputs and trust its machine learning to assemble the right ad, for the right person, in the right place, at the right bid.

The campaign draws on Google's real-time understanding of user intent, signals from your account, and the conversion data flowing back from your site or product feed. Every time someone could see an ad, the system runs an auction decision in milliseconds, deciding whether to show your ad, which combination of your assets to use, and how much to bid. This is Smart Bidding applied across every Google surface at once.

The inputs you provide are: a conversion goal and its value, a budget and bidding strategy (maximise conversions or maximise conversion value, optionally with a target CPA or ROAS), one or more asset groups, audience signals, and (for retailers) a Google Merchant Centre product feed. Everything else, the placement, the format, the audience and the bid, is decided by the AI.

Where the ads show

A single Performance Max campaign can serve ads across all of Google's advertising channels. That is the headline feature: one campaign, full network reach. The system shifts spend towards whichever surface is delivering conversions at your target cost.

• Google Search: text ads alongside organic results for relevant queries.
• Google Shopping: product listings from your Merchant Centre feed.
• YouTube: video and image ads before and during videos.
• Display network: image ads across millions of partner websites and apps.
• Gmail: promotions and social tab placements.
• Google Maps: local placements for businesses with a location.
• Discover feed: ads in the personalised Discover feed on mobile.

Because the channel mix is automatic, you cannot force the campaign to spend only on Search or only on Shopping. You can shape it through your assets, goals and exclusions, but the final distribution is the AI's call. This is the single biggest mindset shift for advertisers used to Google Shopping or Search campaigns where placement is explicit.

Performance Max serves ads across seven Google surfaces from one campaign: Search, Shopping, YouTube, Display, Gmail, Maps and Discover. The campaign uses AI to decide the channel mix, format and bid in real time, optimising towards the conversion goal and value you set. Advertisers control assets, audience signals, budget and goals, but not the final placement of each ad. Source: Juicy Designs, Google Ads management for South African clients, 2026.

Asset groups and audience signals

Asset groups and audience signals are the two levers that shape what Performance Max does with your budget. Getting both right is where the real work of running a successful campaign lives.

Asset groups

An asset group is a themed bundle of creative for one product or service line. Inside it you provide multiple headlines, long headlines, descriptions, images, logos and ideally video. Google mixes these into ads tailored to each placement: a text ad on Search, an image ad on Display, a video on YouTube. The more varied and high-quality your assets, the more combinations the system can test, and the better it performs. Thin asset groups (one image, two headlines) starve the AI and produce weak ads.

A practical structure is one asset group per distinct theme. A car dealership might run separate asset groups for new vehicles, used vehicles and servicing, each with its own creative and landing page. This keeps the messaging relevant and lets you read performance by theme.

Audience signals

Audience signals are suggestions, not hard targets. You feed the campaign your customer lists, website visitors, custom segments built from search terms and interests, and relevant demographics. The AI treats these as a strong starting hint about who converts, then expands beyond them when it finds other people who behave similarly. Good signals shorten the learning period and improve early results. Weak or missing signals leave the AI guessing for longer and waste budget while it learns.

When to use Performance Max

Performance Max works best when you have clear conversion tracking, a reasonable budget and enough conversion volume for the AI to learn from. It is not a universal answer. The right call depends on your account maturity and goals.

It is a strong fit for e-commerce retailers with a healthy product feed, for established lead-generation businesses that can feed quality data back to Google, and for advertisers who want full-network reach without managing separate campaigns for each channel. It pairs well with a core Search campaign that captures high-intent brand and product searches.

It is a poor fit for brand-new accounts with no conversion history, for businesses that cannot track conversions accurately, and for advertisers who need tight control over exactly where ads appear. If you are starting from zero, prove demand with a focused Search campaign first, gather conversion data, then layer Performance Max on top. Our PPC management team almost always sequences it this way.

Pros and cons

Performance Max trades control for reach and automation. Whether that trade is worth it depends on your data quality and how closely the campaign is managed.

Pros

• Full-network reach from a single campaign, with no manual channel splitting
• Smart Bidding optimises towards your conversion value in real time
• Finds incremental conversions across surfaces you might not run manually
• Lower day-to-day management overhead once it is set up well
• Strong for retailers, replacing Smart Shopping with broader reach

Cons

• Limited transparency: placement and search-term data are restricted
• Can cannibalise brand traffic that you would win cheaply on Search
• Heavily dependent on accurate conversion tracking; bad data means bad spend
• A learning period during which results are volatile
• Less control over exactly where and to whom ads show

Performance Max suits advertisers with accurate conversion tracking, a sufficient budget and existing conversion data; it is a poor fit for brand-new accounts with no history. Its strengths are full-network reach and automated bidding; its weaknesses are limited placement transparency and a dependency on clean conversion data. South African businesses should usually run a core Search campaign alongside it. Source: Juicy Designs, founder-led Google Ads management, 2026.

Managing it well

Performance Max is automated, not autonomous. It still needs disciplined management to perform. The campaigns that succeed are the ones where a human keeps feeding the machine good data and good creative.

• Track conversions accurately: import offline conversions and assign realistic values so the AI optimises towards profit, not just volume.
• Feed strong assets: fill every headline, description, image and video slot, and refresh assets that underperform.
• Use real audience signals: upload customer lists and build segments from your best converters rather than leaving signals empty.
• Add brand and negative exclusions: request account-level brand exclusions and negatives so Performance Max does not buy traffic you already win cheaply.
• Watch asset group performance: use the insights and asset reporting to see which themes and creative are driving results.
• Protect your Search campaign: keep high-intent searches in a dedicated Search campaign so they are not absorbed at a higher cost.

“Performance Max is only as good as the data you feed it. We see South African businesses switch it on, leave the audience signals blank and the conversion values at zero, then wonder why it burns budget. Set it up properly, feed it your real customer data, and protect your brand searches, and it becomes one of the strongest tools in the account.”

Cobus van der Westhuizen, Founder & Google Ads Certified, Juicy Designs, reviewed and verified June 2026

Common mistakes

Most Performance Max disappointments trace back to a handful of avoidable setup mistakes. Avoiding these puts you ahead of most accounts.

• Zero or inaccurate conversion values: the AI cannot optimise towards profit if every conversion looks identical.
• Empty audience signals: launching with no signals forces a long, expensive learning period.
• Thin asset groups: one image and two headlines give the system almost nothing to test.
• No brand exclusions: Performance Max quietly claims cheap brand searches and inflates reported results.
• Running it instead of, not alongside, Search: high-intent searches deserve a dedicated, controllable campaign.
• Judging it too early: turning it off during the learning period before it has had time to optimise.

Frequently asked questions