Loading…
Your Full-Service Digital Agency & AI Strategy Partner
Real strategists. Real AI tools. Real growth. — 1Digital® since 2012
Workspace by 1Digital® — the agency platform we built. Coming to select agencies. Join the early-access list →
PMax management built for tROAS — asset group structure, audience signals, search themes, brand exclusion lists, and the PMax-vs-Standard-Shopping tradeoffs other agencies skip. We treat PMax as one campaign type inside a structured Google Ads account, not as a black box you hand a budget to and hope.
TL;DR
1Digital®manages Performance Max campaigns for 400+ brands across eCommerce, retail, and lead-gen. PMax is now the dominant Google Ads campaign type for Shopping inventory and the default recommendation for new spend — but the platform's opacity makes it easy to lose visibility into what's working. We engineer PMax accounts with asset group discipline, audience-signal steering, search-theme inputs, brand exclusion lists, conversion value rules, and PMax-vs-Standard-Shopping holdouts so you keep control over the levers that matter. Pair this page with Google Ads management for the full-account view and Google Shopping feed management for Merchant Center feed work.
Performance Max (PMax) is Google's automated, cross-inventory campaign type. A single PMax campaign competes for impressions across Search, Shopping, YouTube, Display, Discover, Gmail, and Maps — with Google's machine learning deciding which surface, which audience, and which creative combination serves each impression. You provide goals, budget, conversion data, creative assets, audience signals, and search themes; PMax decides the rest.
This opacity is the central management challenge. Search-term-level reporting is partial. Audience-level reporting is partial. Placement-level reporting is partial. The levers you do control — asset groups, audience signals, search themes, brand exclusions, conversion value rules, and creative — are the only levers that move the campaign. Pulling those levers competently is the whole job.
Asset groups are the only structural unit you control inside a PMax campaign — the PMax equivalent of ad groups in Search. They organize creative, audience signals, search themes, and listing-group product filters. Structure decisions made here cascade into every reporting and optimization surface PMax exposes.
For most eCommerce accounts, asset groups are organized by product category, brand, or seasonal theme — each with creative and signals aligned to that theme. A camping brand might run separate asset groups for Tents, Sleeping Bags, Cookware, and Apparel; a beauty brand might split by Skincare / Makeup / Hair / Fragrance with audience signals reflecting each category's shopper intent.
For B2B or lead-gen PMax campaigns, asset groups can mirror customer cohorts — Enterprise prospects vs SMB prospects, or Vertical A vs Vertical B — with audience signals and creative tuned to each cohort.
Each asset group's listing group filters control which Merchant Center products it can promote. We split high-margin SKUs into their own asset groups with separate tROAS targets, isolate brand vs non-brand product feeds, and segment new arrivals or BFCM hero products for budget pacing control.
Audience signals are not targeting constraints — they're steering inputs. They tell PMax's machine learning where to start looking, then the system expands beyond the signal as it finds conversion inventory. Good signals shorten the learning period; weak or missing signals leave PMax burning budget on inventory it has to discover from scratch.
Search themes (introduced 2023, expanded 2024-2025) let you tell PMax which queries the campaign should be eligible to compete for. They're directional — not exact-match targeting — and Google will still serve adjacent queries it identifies as related. We treat search themes as steering inputs, not keyword replacements:
Brand exclusion lists are the single most important PMax setting that ships off by default and is most often left unconfigured. Without them, PMax will aggressively serve against your own brand search queries — cannibalizing Search brand campaign impressions at a much higher CPA, inflating PMax ROAS with revenue Search would have captured at a fraction of the cost, and obscuring incrementality measurement.
We configure account-level brand exclusion lists matched against your Search brand campaign keyword sets, plus competitor brand exclusion lists where competitor conquesting isn't the intent. For accounts where competitor conquesting is the intent, those exclusions are intentionally omitted and the incrementality math is documented.
PMax replaced Smart Shopping in 2022 and now dominates Shopping inventory by default — but Standard Shopping still exists, still has product-feed reporting PMax doesn't, and still has its place in mature accounts. The right mix is account-specific.
We typically run PMax as the primary Shopping vehicle with Standard Shopping holdouts on specific high-volume SKUs or product lines where reporting transparency justifies the structural complexity. The holdout serves as a measurement counterfactual — if PMax outperforms on shared SKUs, the structure is validated; if not, more inventory shifts to Standard.
Creative is the second-highest-leverage PMax lever after audience signals. PMax mixes-and-matches text headlines, descriptions, images, videos, logos, and (for retail) product feeds across every served placement. Asset variation directly drives which combinations PMax discovers and scales.
PMax bidding is only as smart as the conversion value it sees. We tune three layers:
Adjust conversion value at bid time based on audience, location, or device — boosting bids for high-LTV customer cohorts, low-return-rate geographies, or platforms with above-average AOV. Without value rules, PMax treats a $100 conversion from a one-time buyer the same as a $100 conversion from a known repeat purchaser; with rules, it doesn't.
PMax's new-customer acquisition goals (bid higher for new customers, or new customers only) tie bidding to first-party Customer Match lists. We configure NCA modes based on growth-stage objectives, balanced against blended ROAS to avoid runaway acquisition spend.
Bidding strategy choice mirrors the broader Google Ads framework: tROAS for ROAS efficiency, Max Conversion Value for growth-mode spend deployment. Targets ratchet over 30-45 day windows as the learning period stabilizes, with portfolio bidding pooling budget across multiple PMax campaigns sharing a goal.
Performance Max is Google Ads’ automated, cross-inventory campaign type. A single PMax campaign serves across Search, Shopping, YouTube, Display, Discover, Gmail, and Maps — with Google’s machine learning deciding placement, audience, and creative combination per impression. You control asset groups, audience signals, search themes, brand exclusions, conversion value rules, and creative assets; PMax handles the rest.
Pair it. PMax dominates Shopping inventory by default but Search non-brand keyword campaigns continue to outperform PMax steering for high-intent commercial queries you know convert — and Standard Shopping retains search-term and product-level reporting PMax doesn’t expose. We typically run PMax as the primary Shopping vehicle with Standard Shopping holdouts on high-volume SKUs and Search non-brand campaigns for category-defining commercial queries. Brand exclusion lists are mandatory regardless.
Without brand exclusion lists, PMax will aggressively serve against your own brand search queries — cannibalizing Search brand campaign impressions at much higher CPA, inflating PMax ROAS with revenue Search would have captured at lower cost, and obscuring incrementality measurement. Brand exclusion is non-negotiable on every account we manage, with optional competitor-brand exclusion based on conquesting strategy.
Asset groups are the only structural unit inside a PMax campaign — the equivalent of ad groups in Search. They organize creative assets, audience signals, search themes, and listing-group product filters. For most eCommerce accounts, we structure asset groups by product category, brand, or seasonal theme; for B2B and lead-gen, by customer cohort. Listing group filters control which Merchant Center products each asset group can promote, enabling separate tROAS targets per margin tier or product segment.
The highest-leverage signals are first-party Customer Match lists (purchasers, high-LTV cohorts, abandoners), custom segments built from category research keywords and competitor URL patterns, GA4 website visitor audiences (cart abandoners, viewed-but-not-purchased), similar audiences seeded from Customer Match lists, and in-market audiences relevant to each asset group theme. Multiple combined signals per asset group give PMax richer training data and shorter learning periods.
Search themes are directional steering inputs, not exact-match targeting. They tell PMax which query territories you want it eligible to compete for; Google will still serve adjacent queries it identifies as related. They’re a useful nudge into commercial-intent territory PMax might otherwise underweight, but they don’t replace Search non-brand keyword campaigns for high-intent commercial queries where direct keyword targeting still wins.
Initial learning period is typically 7-14 days after launch or major structural changes. Stable performance windows emerge around day 30. We treat day 30-60 as the first calibration milestone for tROAS targets and asset group composition, with quarterly refreshes thereafter. Conversion value rules, new-customer acquisition modes, and search theme additions each trigger short re-learning windows that we plan around major launches.
Text headlines (multiple length and angle variants), descriptions, images at multiple aspect ratios (1.91:1, 1:1, 4:5), logos at square and landscape ratios, and video at multiple lengths (6s, 15s, 30s) for Shorts, in-stream, and feed inventory. Without provided video assets, Google auto-generates stock-template video; we replace those with branded creative. Creative refresh cadence is 4-6 weeks to fight ad fatigue, with asset performance ratings (Low/Good/Best) guiding rotation.
Partially. PMax exposes search term insights at the campaign level via the Insights tab (categories and themes, not full search-term reports), and full search-term reports for queries that triggered Shopping inventory specifically. Search query coverage for non-Shopping inventory is summarized at the theme level. This is one of the structural opacity tradeoffs that justifies running Standard Shopping holdouts and Search non-brand campaigns in parallel for accounts that need full search-term transparency.
PMax management is included in our Google Ads management engagements (starting at $185/hour with four-tier packages — Starter, Growth, Enterprise, Custom; 3-month initial term). Ad spend is separate and paid directly to Google. For PMax-only engagements on accounts running other PPC vehicles externally, scoped retainers are available at the same hourly rate. Merchant Center feed work (a hard prerequisite for retail PMax) is covered under /google-shopping-feed-management.
Asset group structure, audience signals, search themes, brand exclusions, value rules, and creative reviewed against documented standards.
Related
Asset groups, audience signals, search themes, brand exclusions, value rules, and creative — pulled with discipline — turn PMax into one of the most consistent ROAS-compounders in the Google Ads stack. We've run PMax (and its Smart Shopping predecessor) since launch.
