Performance Max Assets: A Practical Guide for UK Advertisers

You've launched a new Performance Max campaign, uploaded a Merchant Center feed and added a few headlines, then watched Google assemble ads across placements you can't inspect in the same way as a traditional Search campaign. A week later, one asset is labelled weak, another has barely served, and the instinct is to delete both.

That reaction is understandable, but often premature. Performance Max assets are inputs to an automated system, not isolated adverts with a simple pass or fail result. The practical challenge for UK SMEs is to provide enough useful creative variety for Google to explore, while retaining enough structure and patience to make sensible decisions.

What Performance Max Assets Actually Do

A UK retailer can upload a product feed, add headlines, descriptions, images and videos, then let Google assemble ads across Search, YouTube, Display, Discover, Gmail and Maps. The system adapts each combination to the placement and user context. Google's UK Performance Max overview describes this process as automatically mixing and matching creative assets to support conversion goals.

Take an outdoor furniture retailer. The feed can supply the product name, price, availability and main image. An asset group can add messages about delivery, garden use, materials and seasonal demand. A headline may perform well in Search, while an image suits Display and a short video fills a YouTube placement. The retailer does not need a separate campaign for every possible combination.

The trade-off is control. You provide the raw material, but Google chooses which assets to combine, where to show them and which users to prioritise. Repetitive, contradictory or narrow inputs reduce the quality of those choices. Well-organised inputs give automation more relevant combinations to test, without forcing the account into a large manually duplicated structure.

Asset groups give the system a theme

An asset group combines text, images, videos and other inputs around a defined theme. Ecommerce groups might follow product categories, ranges, promotions or landing pages. Lead generation groups can centre on a service, location or customer need.

The assets should make sense together. A group for commercial vans should not combine business finance copy with lifestyle imagery aimed at private car buyers. Clear themes give advertisers a practical control point, even though Google assembles the final ad.

Every Performance Max campaign needs at least one asset group and can contain up to 100 asset groups, as confirmed in the Google Ads API documentation for asset groups. One group can support an initial launch. Separate groups help distinguish products, propositions and landing pages when those differences affect the message or conversion path.

Practical rule: Supply coherent material and give the system room to test. Do not delete an asset because its early signal looks weak unless it has enough impressions, time and comparable context to justify that decision.

Asset reporting is directional, not a verdict on creative quality. A low-rated asset may have received limited exposure, appeared in combinations that were not its strongest fit or still be contributing value that the report does not show clearly. Review performance alongside volume, conversion data and the role of the wider asset group before making changes.

For a practical discussion of the control trade-off, see PPC Geeks' guide to controlling automated Performance Max campaigns. Performance Max is a system that turns a structured pool of inputs into channel-specific combinations, not a single ad format with one fixed result.

Performance Max Asset Requirements and Recommended Minimums

A new Performance Max campaign can launch with thin creative coverage, but limited inputs also limit the combinations Google can test. For a stronger starting point, Google's UK guidance recommends 11 or more headlines, 4 or more descriptions, 20 images across orientations and 5 videos for an asset group. These targets give automation more relevant text, visual formats and messages to match with eligible inventory. The same guidance recommends allowing 2–3 weeks before replacing low-performing assets, giving the system time to collect serving data. Google's UK asset guidance

Treat those figures as a practical setup guide, not a promise of performance. Requirements and limits vary by asset type, so check the current interface and technical rules before publishing.

The available ranges include 3–15 headlines, 2–5 long headlines, 1–5 descriptions and up to 20 images. Logos and YouTube videos are optional assets. Use the technical requirements as a publishing check, while keeping the strategic decision focused on message coverage and relevance.

Build variety, not repeated wording

Eleven headlines that repeat one benefit give Google less useful choice than eleven distinct angles. A retailer might cover product benefits, category terms, delivery, trust, use cases and calls to action. A lead-generation business could test expertise, service scope, local relevance, urgency and a defined customer problem.

Images need the same discipline. Provide suitable horizontal, square and portrait formats instead of repeatedly resizing one image. Product detail, in-use and contextual lifestyle images can support different placements and messages.

Video also affects advertiser control. If you do not provide video assets, Google may create videos from existing images, text or Merchant Center listings. That can support a launch, but supplied video gives you more control over pacing, framing, branding and the opening visual.

Use Ad Strength as a diagnostic

Ad Strength highlights missing formats or limited coverage. It does not measure profitable sales. A stronger rating cannot guarantee performance, while a weaker rating does not prove that every asset should be removed.

Review the rating alongside conversion quality, value, cost, impression volume and message relevance. Weak asset signals deserve attention only when they have enough exposure and time to support a decision. Google's automation needs room to test, but advertisers still control the quality and variety of its inputs.

For implementation ideas, see this guide to Performance Max best practices. A strong setup balances breadth of inputs with clarity of theme. More assets help when they add distinct, relevant options, not when they create repetition.

Feed-Only Versus Creative Asset Campaigns

For UK ecommerce advertisers, a feed-only Performance Max campaign can look attractive. Google's UK guidance says creative assets aren't required to launch when a Merchant Center feed is used, because product data can provide the foundation for shopping inventory. That makes feed-only useful when the immediate priority is catalogue coverage, product eligibility or a clean initial launch.

The limitation is equally important. Product data alone doesn't give the campaign the same creative inputs for non-shopping placements. Additional headlines, descriptions, images and videos can help the campaign participate more broadly across Google surfaces and create opportunities to test whether creative adds incremental reach or conversion efficiency. Google's UK retail experimentation guidance specifically describes feed-only control campaigns and treatment campaigns with extra assets.

The feed is the baseline

A creative layer can't rescue unreliable product information. Before testing additional assets, check product titles, descriptions, images, availability, pricing and landing page relevance in Merchant Center. The feed should accurately represent what the customer sees after clicking.

Feed-only can be appropriate when:

• Catalogue coverage matters most: You need products eligible for shopping inventory without adding a full creative production workflow.
• Creative resources are limited: A small team can launch from solid product data and add assets later.
• The campaign needs a clean control: A feed-led version provides a useful comparison point for incremental creative testing.

Creative assets become more important when:

• Non-shopping reach matters: You want the campaign to use messages and formats beyond product listings.
• The proposition needs explanation: Services, bundles, delivery terms and product benefits rarely fit inside feed fields alone.
• You can maintain asset quality: Extra inputs are useful only when the team can keep them relevant and aligned with landing pages.

Test the increment rather than assuming it

A fair test starts with a feed-only control and a treatment that adds a defined asset set. Keep the product range, conversion action, budget logic and landing page approach as consistent as possible. The question isn't “do more assets look better?” It's whether the added creative changes outcomes in a way the business values.

Don't add headlines, images and videos in an unstructured batch if you want to understand the result. A staged approach makes the learning more useful, particularly for smaller UK accounts where conversion volume can be limited.

The distinction between shopping-led delivery and broader automation is also relevant to the comparison of Google Shopping and Performance Max for ecommerce PPC. Feed-only is not automatically inferior. It is a narrower operating model, while creative-led campaigns offer more ways for Google to match a proposition to a placement.

How Google Assembles and Tests Your Assets

Performance Max doesn't show one fixed advert to every user. It combines supplied headlines, long headlines, descriptions, images, videos and feed information into variations suited to Google's available channels. The combination might look different in Search from the version shown in a Discover placement or on YouTube.

That means asset diversity affects the system's exploration space. If every headline repeats “shop now” and every image shows the same product angle, Google can assemble combinations, but it has little strategic variation to evaluate. Distinct inputs let the system test different relationships between message, visual and placement.

Give each asset a job

A useful asset set contains deliberate roles:

• Benefit headlines explain why the product or service matters.
• Category headlines make the offer clear to people who are still browsing.
• Proof-oriented copy addresses delivery, quality, expertise or reassurance without making unsupported promises.
• Contextual images show the item in use or place the service in a recognisable setting.
• Video assets provide movement and visual storytelling for YouTube inventory.

For a home improvement business, one headline might focus on the service, another on the customer problem and another on the area served. The images should support those angles rather than introduce unrelated themes. This makes the asset group coherent while preserving useful variation.

Repetition is the quiet failure mode. Rewriting one claim with minor word changes can consume an asset slot without giving the system a new hypothesis. Stronger variation changes the angle, not just the adjective.

Match formats to real placements

An image with a wide angle may communicate a full room or product range, while a portrait version may work more naturally in vertical environments. Video should be planned for sound-off viewing, clear branding and a visible proposition, rather than treated as a slideshow created at the last moment.

Google can generate video when the advertiser doesn't supply everything. That fallback helps with coverage, but it shouldn't be mistaken for a creative strategy. Your own footage or designed motion asset can express product detail, process and brand tone with more control.

The right question is not whether Google can create an ad from the inputs. It can. Ask whether the inputs give it useful choices that remain accurate across the surfaces where your UK customers may encounter them.

Common Mistakes When Interpreting Asset Reporting

A retailer sees one asset marked weak after a recent campaign change and removes it immediately. The account then loses a message that had barely entered the auction, while Google shifts delivery towards a narrower set of combinations. That reaction is common, but the label alone does not justify it.

Asset reporting can show conversions, cost, impressions, clicks and conversion value at campaign and asset-group level. Google's UK reporting guidance confirms the available detail. The harder decision is judging whether the evidence is strong enough to change the creative.

A weak asset signal is a reason to investigate, not an automatic deletion instruction.

Read the context around the label

Review impressions, clicks, conversions, cost and conversion value together. An asset with limited delivery has had little opportunity to prove its value. Its restricted exposure may reflect Google favouring another combination, a format that fits fewer placements, or a reporting window that covers only part of the learning period.

An asset with substantial exposure and consistently poor commercial results deserves closer scrutiny. Check the landing page, offer, audience mix and conversion tracking before blaming the copy or image. A tracking problem can make sound creative appear ineffective.

Asset-group reporting adds another complication. The group may perform well overall while individual assets look uneven because Performance Max serves combinations rather than every asset as a standalone advert. Removing one apparently weak component can change the mix and eliminate an option for a placement that is not visible in the top-line report.

Protect the learning period

Google recommends waiting 2–3 weeks before replacing low-performing assets, as noted in the UK guidance on asset requirements and optimisation cited earlier. This does not mean every asset will improve. It means immediate edits can interrupt the system before it has gathered enough serving information.

Reporting interfaces can look precise while the underlying sample remains small. UK SMEs should be cautious when conversion volume is modest or the campaign has recently changed budget, targets, landing pages or feed content. Give the account enough stable time to distinguish a weak creative signal from uneven delivery.

Replace evidence, not anxiety

A disciplined review asks:

• Has the asset received meaningful exposure?
• Does its message fit the asset-group theme?
• Does the landing page deliver the same promise?
• Is the conversion action reliable and commercially relevant?
• Would replacing it create a clearer test, or just restart uncertainty?

Use the available data to form a hypothesis, change a defined part of the asset set, and observe the result. Google's expanding reporting and experimentation direction supports that approach, but a tidy dashboard is not complete knowledge. Keep assets that may still have a useful role until the evidence shows that replacing them will improve the account.

Practical Asset Optimisation Strategies That Scale

Sustainable optimisation doesn't come from filling every available slot and replacing anything that looks weak. It comes from creating a controlled creative system that gives Google enough variety without losing the advertiser's commercial logic.

Start with the structure. Each asset group should have one clear theme, one relevant landing page direction and a coherent set of claims. Product category, service type, customer problem and promotion can each justify a separate group when the assets and destination differ.

Structure asset groups by theme

Avoid building one oversized group that mixes unrelated products and propositions. A fashion retailer could separate workwear, occasionwear and outerwear if the imagery, copy and landing pages differ. A B2B advertiser might separate payroll services from accounting support because the buying questions aren't the same.

Keep the group focused enough that a weak result tells you something useful. Excessive fragmentation can make each group harder to evaluate, so the account needs a practical balance between relevance and manageable reporting.

Test message variations

Write assets around different customer motivations, not superficial synonyms. Test convenience against quality, product benefit against use case, or local relevance against service expertise. Keep claims supportable and consistent with the landing page.

A structured test might introduce a new message family while leaving imagery stable, then introduce a visual variation later. That won't isolate every interaction, but it gives the team a clearer read than changing everything at once.

Review weekly, act deliberately

A weekly review should look beyond asset labels. Check campaign-level efficiency, asset-group contribution, channel distribution where available, search themes or queries where available, feed diagnostics and landing page behaviour. Google's reporting surface can inform the review, but it can't decide whether a creative change makes commercial sense for your business.

Senior PPC view: The best optimisation question is often “what should we learn next?” rather than “which asset can we remove today?”

Remove underperformers with a reason

Remove an asset when the evidence is stable, the message is clearly off-theme, the claim is outdated, or the creative creates a brand or compliance concern. Replace it with a deliberate alternative, then record what changed and why.

For teams scaling complex accounts, Performance Max scaling guidance can help connect asset discipline with wider campaign structure. PPC Geeks is one option for UK businesses that need support with account audits, feed optimisation, Performance Max management and reporting.

The practical sequence is structure, test, review, then remove. Automation works best when advertisers provide clear inputs and resist the urge to interfere with every early signal.

PPC Geeks helps UK SMEs and ecommerce teams audit Performance Max asset groups, improve Merchant Center feeds, and build testing plans that balance Google's automation with commercial control. Visit PPC Geeks to discuss your account and identify which asset, feed and reporting changes should come next.