SHOPPING ADS

Performance Max is spending your budget on brand searches instead of new customers — here is how to fix the feed and the structure

PMax replaced Smart Shopping in 2022 and brought new problems with it. Budgets drift toward brand keyword traffic, YouTube, and returning customers. The Shopping placements that drove new customer acquisition under Smart Shopping are underserved. The fix is not a higher budget. It is the right feed, the right asset group structure, and a clear separation of brand and non-brand traffic.

This is for you if

Who this is for

You have Shopping or PMax campaigns running. Revenue is coming in but the ROAS is inconsistent and you cannot tell whether PMax is finding new customers or just converting people who already knew your brand. Your agency has told you PMax is performing. You want a second opinion on what it is actually spending on.

You ran Smart Shopping successfully for two or three years. Google forced the migration to Performance Max in 2022. Since then, new customer acquisition from Shopping has been harder to attribute and your blended ROAS looks fine but your new customer ratio has dropped. You suspect PMax is harvesting existing demand rather than creating new demand.

You set up PMax through the Google Ads interface or through Shopify's Google channel integration. There is one campaign, one asset group, and the feed is pulling directly from Shopify's default product data. It is running. But you have never audited the titles, never segmented by margin, and never checked what queries it is actually triggering. You want to know whether the setup is working or just spending.

What's broken

What is broken

PMax asset groups not segmented by product margin or category

When a single PMax campaign contains all products with a single ROAS target, Google's algorithm allocates budget toward whatever converts most easily. That is usually lower-priced products, products purchased by repeat customers, and products with the most historical conversion data. High-margin hero SKUs that need new customer acquisition compete for budget against low-margin products converting existing customers. The campaign reports a blended ROAS that looks acceptable, but the margin-weighted return is significantly lower.

Product feed titles using brand names and internal SKU codes

Australian shoppers searching Google Shopping use descriptive product terms: "women's linen midi dress Australia," "espresso machine under $500," "timber bedside table natural." They do not search by brand name or SKU. When feed titles are built around the brand's internal product naming ("Aria Dress — SKU4492 — Sage"), Shopping impressions are limited to branded searches and the campaign misses the high-intent category searches that drive new customer acquisition.

No BNPL promotion in Shopping feed

Afterpay, Zip, and Laybuy are standard payment options for Australian e-commerce. For products priced above $50, adding BNPL messaging to Shopping ads through Google's promotion feed (or annotation) improves click-through rate by making the effective cost visible at the ad level. Most Australian brands running Shopping ads have not configured this. The result is that competitors who have configured BNPL annotations appear more attractive in Shopping results for the same product at the same price.

Shopping campaigns cannibalising brand search

Performance Max does not have a native brand exclusion that operates the same way as a negative keyword in a Search campaign. Without a brand exclusion list applied to the PMax campaign, Google will show Shopping ads for searches that include your brand name. Those brand-intent searches would convert through your Brand Search campaign at near-zero CPA. Instead, PMax is claiming credit for conversions it did not generate, inflating its reported ROAS, and consuming budget that should be allocated to non-brand acquisition.

What we engineer

What we do

PMax campaign audit and spend analysis

We pull spend data by asset group, channel, and product group to identify where PMax is actually allocating budget. We flag brand cannibalisation, low-margin product overspend, and YouTube/Display allocation that is not producing measurable Shopping revenue.

Product feed audit and title rewrite

We audit every feed title against actual Australian search query data. Titles using brand names, SKUs, or internal naming conventions are rewritten to match descriptive category searches. We use Google Keyword Planner and Search Console data to inform title structure for every major product category.

Asset group segmentation by margin and category

We restructure your PMax campaign with asset groups segmented by product category and margin tier. ROAS targets are set per asset group to reflect the margin available in each segment, not a single blended target across all products.

Brand exclusion and negative keyword strategy

We apply brand exclusion lists to PMax campaigns to prevent Shopping ads from serving on brand-intent queries. We audit search term data from Standard Shopping and Search campaigns to build negative keyword lists that reduce wasted spend on irrelevant queries.

BNPL promotion feed setup

We configure Google's promotion feed or Shopping annotation to display Afterpay or Zip messaging on eligible products. We verify that the annotation appears correctly in Shopping results and monitor CTR changes across products where BNPL annotation is added.

Standard Shopping campaign alongside PMax

For accounts where search query visibility is essential, we run a Standard Shopping campaign in parallel with PMax. This provides access to actual search term reports, which PMax does not surface, and allows for more granular negative keyword management.

Monthly feed maintenance and performance reporting

We audit the feed monthly for errors, price mismatches, and out-of-stock issues. Performance reports cover ROAS by asset group, new customer ratio where measurable, impression share, and feed health metrics.

What changes

What changes

Before
After
Before When a single PMax campaign contains all products with a single ROAS target, Google's algorithm allocates budget toward whatever converts most easily. That is usually lower-priced products, products purchased by repeat customers, and products with the most historical conversion data. High-margin hero SKUs that need new customer acquisition compete for budget against low-margin products converting existing customers. The campaign reports a blended ROAS that looks acceptable, but the margin-weighted return is significantly lower.
After Asset group segmentation and brand exclusions redirect budget toward non-brand Shopping placements. New customer acquisition from Shopping becomes measurable and the blended ROAS reflects actual incremental revenue, not brand search harvesting.
Before Australian shoppers searching Google Shopping use descriptive product terms: "women's linen midi dress Australia," "espresso machine under $500," "timber bedside table natural." They do not search by brand name or SKU. When feed titles are built around the brand's internal product naming ("Aria Dress — SKU4492 — Sage"), Shopping impressions are limited to branded searches and the campaign misses the high-intent category searches that drive new customer acquisition.
After Impressions increase on high-intent category searches because titles are built around what Australian shoppers actually type into Google Shopping. CTR improves because product descriptions match buyer expectations before they click.
Before Afterpay, Zip, and Laybuy are standard payment options for Australian e-commerce. For products priced above $50, adding BNPL messaging to Shopping ads through Google's promotion feed (or annotation) improves click-through rate by making the effective cost visible at the ad level. Most Australian brands running Shopping ads have not configured this. The result is that competitors who have configured BNPL annotations appear more attractive in Shopping results for the same product at the same price.
After Every asset group has a ROAS target derived from the actual margin of the products in that group. You stop over-protecting low-margin products and under-investing in high-margin hero SKUs. Budget allocation reflects business priorities, not algorithmic convenience.
Before Performance Max does not have a native brand exclusion that operates the same way as a negative keyword in a Search campaign. Without a brand exclusion list applied to the PMax campaign, Google will show Shopping ads for searches that include your brand name. Those brand-intent searches would convert through your Brand Search campaign at near-zero CPA. Instead, PMax is claiming credit for conversions it did not generate, inflating its reported ROAS, and consuming budget that should be allocated to non-brand acquisition.
After BNPL annotations, accurate availability data, and structured titles place your products alongside or above competitors who are running default feeds. In Australian Shopping auctions where multiple sellers offer similar products, feed quality is a direct differentiator.
How it works

How we work

  1. 01

    Diagnostic

    We audit your current PMax and Shopping campaign structure, pull spend data by channel and product group, review the product feed against Australian search query data, and assess conversion tracking accuracy. You receive a written diagnostic documenting exactly what is broken and what the fix requires before we discuss scope.

  2. 02

    Feed and structure rebuild

    We rewrite product titles, segment asset groups by margin and category, apply brand exclusions, and set ROAS targets per segment. If a Standard Shopping campaign is needed for search query visibility, we build that alongside PMax. Feed changes are made in a staging environment and reviewed before going live.

  3. 03

    BNPL and promotion setup

    We configure BNPL annotations for eligible products, verify display in Shopping results, and document which products have annotations active. Promotion feed entries are set up for sale periods or ongoing offers where applicable.

  4. 04

    Ongoing optimisation and reporting

    Monthly feed audits, weekly performance monitoring, and quarterly structural reviews keep the campaigns calibrated to your product catalogue and margin structure. Reporting covers ROAS by segment, impression share, new customer ratio, and feed error count.

Common questions

Frequently asked questions

Should I run Performance Max or a standard Shopping campaign in Australia?

Performance Max is the primary Google Shopping format in Australia and Google allocates more Shopping inventory to PMax campaigns than to Standard Shopping. Most Australian brands benefit from running PMax as their primary Shopping vehicle, with a Standard Shopping campaign running alongside it at a lower budget to surface search term data that PMax does not provide. The Standard Shopping campaign is not intended to compete with PMax on budget. Its purpose is intelligence: actual search queries, which PMax hides, that inform negative keyword strategy and title optimisation.

How do I stop PMax from spending on my brand keywords?

Google provides a brand exclusion feature within Performance Max that can be applied at the campaign level. Unlike standard negative keywords in a Search campaign, brand exclusions in PMax are applied via a brand list that you build using your brand name and common misspellings. Once applied, PMax will not trigger Shopping or Search ads on queries matching the brand exclusion list. This does not prevent your Brand Search campaign from continuing to serve on those queries. The exclusion needs to be reviewed after major product launches or brand name changes, as new brand-adjacent terms may not be covered automatically.

Can I add Afterpay messaging to Google Shopping ads?

Yes. Afterpay annotations can appear on Google Shopping ads for Australian merchants who have Afterpay configured as a payment method on their site. The annotation is enabled through Google Merchant Centre's "buy on Google" or promotions feed, depending on account configuration, and displays the instalment amount (for example, "4 x $25") alongside the product price in Shopping results. Eligibility requires that the product price falls within Afterpay's minimum and maximum transaction limits and that Afterpay is a live payment option at checkout. The annotation does not require a separate ad format or additional creative.

How do I segment product groups within a PMax campaign by margin?

Product group segmentation in PMax is done at the asset group level. You create separate asset groups within the same campaign and use listing group filters (product type, custom label, brand, or ID) to assign specific products to each asset group. Margin tiers are most easily implemented using custom labels in the product feed: you assign a custom label value (such as "high-margin" or "tier-1") to products in each margin band, then use that custom label as the listing group filter when building asset groups. Each asset group can then have its own ROAS target set via the campaign's asset group bidding settings, allowing the algorithm to bid appropriately for products with different margin profiles.

What ROAS target should I set for Google Shopping in Australia?

There is no universal ROAS target that applies across product categories or business models. The correct ROAS target for your Shopping campaign is derived from your gross margin and your acceptable cost-of-sale percentage. If a product has a 60% gross margin and you are willing to spend 20% of revenue on paid acquisition, the target ROAS is 5x. If the margin is 40% and the acceptable cost-of-sale is 25%, the target ROAS is 4x. Setting ROAS targets without reference to margin data produces campaigns that optimise for a metric rather than for profit. We calculate target ROAS by asset group as part of the campaign build, using the margin data you provide.

Our team

The people behind the work

Not a black box. Real specialists you can call, with their names on the work.

Niraj Raut

Niraj Raut

Founder — Ecommerce SEO
Keshab Joshi

Keshab Joshi

PPC Expert
Hawrry Bhattarai

Hawrry Bhattarai

Google Ads Expert
Arogya Rijal

Arogya Rijal

SaaS SEO Expert
Start here

Find out where your Shopping budget is actually going

We start with a diagnostic. We pull your campaign spend by channel and product group, review your feed against Australian search data, and check for brand cannibalisation. You get a written summary of what is costing you and what the fix involves. No obligation to proceed.