the Advantage+ Shopping readiness checklist: 23 things Meta scores you on .

Advantage+ doesn't just need a catalog; it needs a complete one. we mapped every signal Meta uses to decide whether your products qualify for the high-performing inventory — and which ones quietly tank your delivery.

Advantage+ Shopping campaigns are Meta’s automated equivalent of Performance Max — a single campaign that decides on its own which products to show, to whom, on which placements. when it works, it works. when it doesn’t, the answer is almost always “your catalog isn’t ready” and Meta’s UI gives you very little to act on.

What’s actually happening underneath: Meta scores every product in your catalog on roughly two dozen signals. products that score above a threshold get pulled into the Advantage+ inventory pool; products below it get suppressed or shown only to weak audience segments. nobody at Meta will give you the threshold. but the signals are knowable.

Here’s the complete list, ordered by impact.

Tier 1 — required, or you’re not in the auction

1. id stable across feed updates. if your product IDs change between feed pulls (Shopify variant_id resets, SKU schema changes), Advantage+ treats your products as new daily. learning resets. you never escape the “evaluation” period.
2. title, description, image_link, price, availability, brand, condition. these are the table-stakes. Meta validates them; missing any one disqualifies the product.
3. availability accuracy within 4 hours. Meta penalizes catalog freshness misses by suppressing entire product sets, not just the OOS items. one stale “in_stock” on a sold-out product can drag the others.

Tier 2 — high-impact for Advantage+ specifically

4. Multiple images per product (minimum 3, ideally 6+). Advantage+ assembles dynamic creatives by mixing images. one image gives Meta nothing to work with; six gives it a chance to test layout combinations.
5. Lifestyle imagery in additional_image_link. the primary image_link should be a clean product shot. lifestyle shots — model, in-context, environmental — go in additional. Advantage+ needs both for placement diversity.
6. product_type with depth. 4–5 levels of taxonomy (Apparel > Outerwear > Fleece > Quarter-Zip > Men's) gives Meta clean grouping for asset rollups. flat categorization gets binned with everyone else.
7. google_product_category. yes, on Meta. Meta still uses Google’s taxonomy as a normalization layer.
8. Variant pooling via item_group_id. same as PMAX — Meta needs to know which products are siblings. without it, learnings don’t pool across colors/sizes and most variants never get enough impressions to learn from.
9. Custom labels (0–4) populated. used by Meta as audience-segment hints. typical mapping: margin tier, best-seller flag, seasonality, target gender if not in attributes, inventory depth.
10. Aggregate review rating ≥ 3.5 stars. products below this threshold get Advantage+ deprioritization across all placements.
11. Review count ≥ 5. catalog ads with no review density underperform; Meta accounts for this in scoring.

Tier 3 — meaningful, often missed

12. color, size, material, pattern for apparel. required for Apparel category to be eligible for Advantage+ Shopping at all in some regions.
13. age_group, gender. directly influences which audience segments see your products.
14. shipping per product. Meta uses declared shipping in its profitability model for value-based bidding. blank or globally-defaulted shipping costs you the granular signal.
15. Pricing within reasonable bounds vs your category. Meta scores price competitiveness internally. an outlier-priced product (3× category median) gets quietly suppressed unless it’s clearly luxury-positioned by other signals.
16. Catalog match rate ≥ 90%. at the catalog level, your conversion events need to map cleanly to catalog products. below 80% Meta starts deprioritizing the entire catalog. (we wrote about how this gets broken silently.)

Tier 4 — edge cases that matter for specific stores

17. product_highlight bullets. 3–10 short benefit statements. shown in some catalog placements; underused.
18. sale_price_effective_date. set scheduled sales here, don’t toggle live. Advantage+ pre-trains on upcoming sales when you give it lead time.
19. multipack / is_bundle. for stores selling bundles or multi-unit SKUs. without these, bundles look overpriced in unit-price comparisons.
20. identifier_exists: false for private label. explicit beats blank. tells Meta you’re a brand, not a feed gap.
21. Schema.org Product markup on PDPs. Meta crawls landing pages to validate prices and availability against the feed. mismatch triggers warnings; sustained mismatch suppresses delivery.
22. Open Graph product tags. separate from feed; matters for the share-graph distribution Advantage+ pulls from.
23. Catalog-level audience signal: matched purchase events for ≥ 50 products in last 7 days. if your catalog is technically valid but only a handful of products see conversion volume, Advantage+ has no learning surface. send purchase events with content_ids that match the catalog (see #16).

How to audit yours

1. open Commerce Manager → Catalog → Diagnostics. fix every red flag and most yellows.
2. check Catalog → Events → “Content matched to catalog.” needs to be 90%+ for Advantage+ to perform.
3. spot-check 20 random products. for each, count how many Tier 1 + Tier 2 fields are populated. if the average is below 18 of 20 fields, you’re shipping a partial catalog and Advantage+ is rationing your delivery.
4. look at your campaign’s “Top performing products” report. if 80% of spend is on 5% of SKUs, that’s not optimization — it’s the algorithm hiding from products it doesn’t trust enough to bid on.

How Maximo handles this

We score every product in your catalog against this 23-point checklist on every feed run, surface per-product gaps, and let you fix them in batches with rules. the system also enforces that the SKU IDs in your conversion events match the catalog (#16) — which is half the catalog match-rate problem fixed by construction.

The honest framing: Advantage+ doesn’t need every box ticked perfectly. it needs you to clear a threshold across enough products that it has an inventory pool to bid into. most stores fall just short, on the same handful of attributes, for the same reason — nobody told them which ones mattered.