Most Brisbane business owners treat Google Performance Max (PMax) like a slow cooker: you throw in some images, a few lines of text, a handful of products, and hope the 'AI magic' serves up a profit.
I’m going to be blunt: if you are running a single Asset Group containing your entire product catalogue or service list, you aren't 'leveraging AI.' You’re being lazy, and Google is happily charging you for the privilege.
At Local Marketing Group, we’ve audited hundreds of accounts from the Gold Coast up to Sunshine Coast. The most common failure isn't the budget—it’s the lack of Asset Group granularity. When you give Google a giant bucket of unorganised assets, the algorithm defaults to the path of least resistance, which usually means spending your money on the easiest (not the most profitable) clicks.
The 'Catch-All' Fallacy
The industry standard advice for PMax used to be 'give the machine more data.' Agencies told you to lump everything together so the algorithm could learn faster. In 2026, we know that’s rubbish.
When you mix high-margin services with low-margin 'foot-in-the-door' offers in one Asset Group, Google’s AI doesn't know which one keeps your lights on in Milton. It just knows which one gets a click. To win, you must structure your Asset Groups to mirror your business's actual value drivers.
1. Structure by Intent, Not Just Category
Most beginners think of Asset Groups like folders on a computer: 'Shoes,' 'Socks,' 'Hats.' This is a mistake. Google doesn't just look at what the product is; it looks at who the person is.
Instead of just categorising by product type, segment your Asset Groups by audience intent. For example, if you run a solar installation business in QLD, you shouldn't have one 'Solar' asset group. You need:
The 'Savings' Group: Focused on rising electricity costs and rebates (Data-heavy, ROI-focused copy). The 'Sustainability' Group: Focused on carbon footprint and future-proofing (Values-based imagery). The 'Premium' Group: Focused on Tesla Powerwalls and high-end inverters (Luxury lifestyle creative).
By splitting these, you can provide specific 'Signals' (Search themes and customer lists) to each group. If you feed the same generic signal to everything, you're essentially measuring ROI based on a lie, because the AI is guessing which creative fits which user.
2. The 'Hero' vs. 'Zombie' Product Split
If you’re in e-commerce, this is where you’re likely losing the most money. In any given Shopify or WooCommerce store, 20% of products usually generate 80% of the revenue.
When these are all in one Asset Group, the 'Zombies' (products with zero impressions or clicks) get ignored, or worse, they eat up 'test' budget that never converts.
The Fix: Create a 'Hero' Asset Group for your top 10-20 performers. Give this group your best lifestyle photography, your most compelling video assets, and specific customer signals. Then, create a separate 'Long-tail' Asset Group for the rest. This allows you to set different ROAS targets. You can't expect a niche $15 accessory to perform with the same efficiency as a $500 flagship product.
3. Why Your Creative is Failing the Algorithm
I see this constantly: a business owner uploads three stock photos and a blurry logo and wonders why PMax isn't working.
Google is a hungry beast that eats creative. If you provide mediocre assets, Google will default to 'Text-only' formats or—God forbid—auto-generated videos that look like a 2005 PowerPoint presentation. This is a massive ad spend tax because your Click-Through Rate (CTR) will plummet, driving up your costs.
Each Asset Group needs: Specific Search Themes: Don't just use your brand name. Use high-intent phrases your customers actually utter. Differentiated Headlines: If the Asset Group is for 'Emergency Plumbing,' every headline should mention speed and 24/7 availability. Vertical Video: It's 2026. If you aren't providing 9:16 video for YouTube Shorts and Display, you're missing out on the cheapest inventory Google has to offer.
4. The Signal is the Strategy
Asset Groups are only as good as the 'Audience Signals' you attach to them. This is where most beginners (and lazy agencies) fail. They rely on Google’s 'Interest' categories like 'Home & Garden.'
That’s too broad. It's useless.
To actually move the needle, you need to feed the machine your own data. Upload your customer lists. Use 'Custom Segments' based on the specific search terms your competitors are bidding on. I’ve seen cases where adding a high-quality first-party data list to an Asset Group reduced CPA by 30% overnight. If you aren't doing this, you're just renting data and letting Google keep the insights.
5. How Many Asset Groups is Too Many?
There is a tipping point. If you have a $50/day budget and 20 Asset Groups, no single group will ever get enough data to exit the 'Learning' phase.
As a rule of thumb for Australian SMEs: Budget < $100/day: Stick to 2-3 highly focused Asset Groups. Budget $100-$500/day: You can push to 5-8 groups, segmenting by top categories or high-value services.
- Budget $500+/day: You have the luxury of granular testing—segmenting by seasonal offers, specific personas, or clearance vs. full-price lines.
The 'Invisible' Asset Group: Brand Protection
One final tip that will save you thousands: PMax loves to 'steal' brand conversions. It will bid on your own business name, show an ad to someone who was already looking for you, and claim it as a 'win.'
While you can now use Brand Exclusions at the campaign level, I prefer to see how much of a 'Lazy Tax' a client is paying before we switch it off. If your Asset Group performance looks too good to be true, check your Insights tab. If 70% of the conversions are coming from your own brand name, your Asset Group structure isn't working—your reputation is just doing the heavy lifting while Google takes the credit.
Summary of Action Items
1. Audit your current groups: Are you mixing high-margin and low-margin products? If yes, split them. 2. Check your assets: Do you have at least one video and five high-quality images per group? If not, you’re being penalised in the auction. 3. Refine your signals: Stop using generic interests. Use custom search segments and your own CRM data. 4. Monitor the 'Insights' tab: Look for which search themes are actually triggering your ads. If they don't match the assets in that group, move them.
Structuring PMax isn't a 'set and forget' task. It’s an ongoing process of categorising intent and feeding the machine the right fuel. If your current agency tells you that 'the AI handles the structure,' they’re likely just avoiding the hard work of creative production and data segmentation.
Stop letting Google's AI play guessing games with your Brisbane business. Give it the structure it needs to actually perform.
Ready to stop wasting spend on messy PMax campaigns? Contact Local Marketing Group today and let’s get your data working for you.