Performance Max Strategy: Add One Real Video Before Touching Bids
Run Performance Max as a goal driven engine: feed it accurate conversion tracking, rich creative including video, and strong first party audience signals, then use campaign structure, negatives, and focused experiments to steer it toward customers you actually want. Two moves to make this week: upload at least one real video asset, and layer in a Customer Match list before you touch your bids.
TL;DR:
- Below roughly 30 to 50 monthly conversions, Smart Bidding may lack stability; consolidate campaigns unless product lines need separate budgets or bid strategies.
- Google recommends 20 text assets, seven images, and at least one video; advertiser supplied videos average 12% more conversions than auto generated versions.
- Treat audience signals as learning hints, not targeting controls, and exclude branded searches and existing customers when acquisition, rather than retention, is the goal.
- Use Maximize Conversions when conversion values are similar; choose value based bidding when values vary, and loosen targets if delivery suddenly drops.
- Run asset experiments for four to six weeks, Google’s recommended window, and apply winners only after the full test rather than reacting to early results.
Table of Contents
- How Performance Max actually works under the hood
- Structuring campaigns and asset groups without sabotaging your own data
- Asset requirements that actually move Ad Strength
- Audience signals, search themes, and the exclusions that protect your margin
- Picking the right bidding strategy for the goal you actually have
- The weekly report checklist that actually tells you something
- Running asset experiments without breaking what’s working
- How we measure Performance Max without guessing
- Three ways advertisers burn Performance Max budget for nothing
- Managed Performance Max, run by people who actually watch the dashboard
- FAQ
- Sources
How Performance Max actually works under the hood
Performance Max is a goal based campaign type that spends across Search, Display, YouTube, Discover, Gmail, and Maps from a single setup, letting Google’s bidding systems decide channel mix in real time. You do not pick placements. You feed inputs and the system allocates.
Four inputs matter more than the rest: your conversion actions (what counts as success), your asset quality (what the system has to work with), your landing pages (where traffic lands), and your audience signals (who to look for first). Get the conversion setup wrong and everything downstream is optimized against the wrong target.
This is why accurate conversion values are the real steering wheel. Google’s own guidance is blunt about it: set up accurate tracking, allow one to six weeks for learning, and match your bidding strategy to your actual objective. Skip that step and you are letting a very capable system optimize for the wrong outcome, confidently.
Structuring campaigns and asset groups without sabotaging your own data
The biggest structural mistake advertisers make with Performance Max is treating it like a dozen tiny Search campaigns. Smart Bidding needs volume to learn, and splitting budget into too many thin asset groups starves all of them.
The better heuristic: consolidate when you don’t have the conversion volume to justify separation, and segment only when you need to force spend toward an underexposed product line or a genuinely distinct customer intent. A regional retailer with one core offer should run one tight campaign. A multi category ecommerce brand with wildly different margins and audiences has a real reason to split.
When you do segment, scope asset groups around intent or product category, not arbitrary internal org charts:
- Name asset groups by what a customer is looking for, not by your internal team structure.
- Keep the number of asset groups per campaign low enough that each one can accumulate meaningful conversion data.
- Set budget at the campaign level and let the system redistribute within it rather than micromanaging each group’s spend.
- Reserve a separate campaign only when a product line needs its own budget ceiling or a genuinely different bid strategy.
Search Engine Land’s reporting backs this up: campaigns with fewer than roughly 30 to 50 conversions a month struggle with Smart Bidding stability, which means your instinct to “organize things neatly” is often the thing quietly killing performance.
Asset requirements that actually move Ad Strength
Performance Max is only as good as what you hand it, and most advertisers hand it scraps. The minimum viable upload is not the right target.
Google recommends at least 20 text assets (15 headlines, 5 descriptions), 7 image assets (3 landscape, 3 square, 1 portrait), and at least one video. That’s the floor, not the ceiling.
Statistic: Advertisers who upload their own video assets instead of relying on Google’s auto-generated video see an average of 12% more conversions. That’s a meaningful lift for something most teams skip because video feels like extra work.
- Hit “Excellent” Ad Strength before you obsess over anything else; Google reports an average conversion lift of roughly 6% when moving to that tier.
- Refresh creative every 4 to 6 weeks, in step with your experiment cadence, not on a whim.
- Pull any asset group that spends for 2 to 3 weeks with zero conversions and rework it using practical conversion rate optimization tips rather than letting it bleed budget.
Audience signals, search themes, and the exclusions that protect your margin
Audience signals are not targeting, they’re a hint. Google is explicit that signals seed the algorithm and speed up learning, but the system will explore beyond them once it finds patterns worth chasing. Treat them as a head start, not a fence.
- Load your strongest first party data first: Customer Match lists from your CRM, then in-market segments, then custom segments built from site behavior.
- Use search themes sparingly, mainly when launching a new product or seeding a category with no conversion history yet; they give the AI a starting vocabulary when it has nothing else to learn from.
- Add brand exclusions so Performance Max isn’t quietly taking credit for branded searches your SEO and direct traffic were already winning.
- Layer in audience exclusions for existing customers when the campaign goal is acquisition, not retention.
- Apply a modest new customer bid adjustment, a small placeholder uplift as recent reporting describes it, to nudge the system toward incremental customers without distorting your reported ROAS.
Skipping brand exclusions is one of the fastest ways to make a Performance Max campaign look like a hero when it’s really just harvesting demand you already had.
Picking the right bidding strategy for the goal you actually have
Match the bid strategy to the objective, not to whatever your last campaign used. If every conversion is worth roughly the same to you, run Maximize Conversions. If order values vary (and for most ecommerce and lead-gen accounts, they do), Maximize Conversion Value or Target ROAS gives the algorithm something more useful to chase than raw volume.
Target CPA and Target ROAS both need enough historical data to work. Industry guidance puts the reliable minimum at around 30 to 50 conversions per month; below that, you’re feeding Smart Bidding too little signal to find a pattern.
The single most common self-inflicted wound here: setting a tROAS or tCPA target so aggressive that the system simply stops spending rather than chase an unreachable number. If delivery suddenly drops, your target is probably the reason, not the market.
The weekly report checklist that actually tells you something
Most Performance Max dashboards get opened once a month, skimmed, and closed. That’s backwards. A handful of reports, checked weekly, catch problems before they compound.
- Search terms insights: review which queries triggered spend and flag categories that have no business being there.
- Asset-level performance: identify assets dragging Ad Strength down and swap them out rather than letting the whole group underperform.
- Channel performance breakdown: confirm spend is landing where your margin can support it, not just where volume is easiest to find.
- Landing page spend report: catch traffic quietly routing to pages that were never meant to convert.
Expanded reporting controls now include campaign-level search term views and negative keyword lists, which means there’s genuinely less excuse for flying blind than there was a year ago. Use final URL expansion deliberately: keep it on with a page feed guiding the system toward preferred pages, and turn it off entirely only when you need to lock delivery to a single path.
Running asset experiments without breaking what’s working
Performance Max supports structured A/B testing of assets, and most advertisers either skip it entirely or pull the plug too early to learn anything.
- Set up an asset experiment that splits traffic between your control group and a treatment group with the new creative.
- Let it run a minimum of 4 to 6 weeks, which Google’s own guidance recommends for reaching statistical significance.
- Resist the urge to check daily and declare a winner after five days; small sample sizes lie convincingly.
- Apply the winning asset set account-wide only once the result holds across the full window, then retire the loser.
Pro Tip: Lock your experiment calendar before you start. Checking results early and reacting to noise is the single fastest way to waste a perfectly good test.
How we measure Performance Max without guessing
Most agencies hand you a PDF once a month and call it reporting. We run live Looker Studio dashboards tied directly to GA4 and Google Business Profile, so when a Performance Max asset group stalls, we see it the same day, not thirty days later in a slide deck.
That speed changes what we spend time on. Instead of auditing accounts for weeks before touching anything, we put budget into creative and execution first: new video, fresh image sets, better landing pages, then iterate against real data. For teams managing Performance Max in house, our Google Ads blog covers the tactical side in more depth.
Three ways advertisers burn Performance Max budget for nothing
Weak conversion tracking is the quiet killer: if your values are wrong, the algorithm is optimizing against a target that doesn’t reflect your actual business, fix it by auditing every conversion action and its assigned value before you touch bids. Underinvesting in creative is the lazy habit: seven mediocre images and no video is not a strategy, it’s a shrug, fix it with a real creative sprint. And constant tinkering is the expensive one: adopt a two week cadence for minor changes and respect the 4 to 6 week experiment window, or you’ll never know what actually worked.
— Chris Breikss
Managed Performance Max, run by people who actually watch the dashboard
If reading all of this made you tired just thinking about the asset refresh calendar, that’s fair, it’s a lot to run alongside everything else your business needs. We manage Performance Max as part of Google Ads & LSA, paired with the creative volume it actually needs to perform.
- Google Ads & LSA: campaign structure, bidding strategy, and weekly optimization handled end to end.
- Content Factory: the headline, image, and video volume Performance Max needs to hit Excellent Ad Strength.
- Live dashboards: GA4 and Google Business Profile connected, visible anytime, not bundled into a monthly recap.
We prioritize creative and execution over process, and you’ll see the difference in how fast things move. See your marketing live and check what a real-time view of your campaigns looks like.
FAQ
Is Performance Max worth it?
For most advertisers with accurate conversion tracking and enough creative to work with, yes: Google reports an average 27% increase in conversions or value at a similar cost compared to Search or Shopping-only setups. It performs worse for accounts with thin tracking, weak creative, or very low conversion volume.
How does Performance Max work?
Performance Max runs a single goal based campaign across Search, Display, YouTube, Discover, Gmail, and Maps, letting Google’s bidding systems decide channel allocation automatically. You control the inputs: conversion goals, creative assets, landing pages, and audience signals, and the system optimizes spend against those within your budget.
Is $20 a day good for Google Ads?
There’s no single number that applies to every business or goal; what matters is whether your daily budget supports enough conversions per month for Smart Bidding to learn, with around 30 to 50 conversions monthly considered a reasonable threshold for stability. A $20 daily budget may be enough for a low-cost lead form and far too thin for an ecommerce account with a high average order value.
What are two benefits of Performance Max?
It reaches customers across six Google surfaces from a single campaign setup instead of managing channels separately, and it can lift conversions meaningfully: advertisers see an average 27% increase in conversions or value versus Search or Shopping alone, with uploaded video assets adding roughly 12% more conversions on top of auto-generated creative.
Sources
- Performance Max best practices for lead generation - Google Ads Help
- Top Performance Max optimization tips for 2026 - Search Engine Land
