How product sets slice catalogues

What sets actually do#

Sets filter the catalogue into campaign-ready assortments that update as the feed updates. One catalogue can serve prospecting heroes, basket recovery, clearance and gifting at once, each with its own budget and story. Meta own catalogue ads guide describes how sets attach to campaigns.

Think of sets as editorial decisions, not database queries: the filter expresses a commercial intent. Start from the Meta Ads hub for the family view, then see catalogue ads for how sets meet the auction.

Audit inherited sets before spending: imported demo filters and ancient sale slices linger in most accounts. Archive or rename anything without a living owner. Catalogue setup resets the foundations.

Slices that earn budget#

Build sets by margin, stock depth, price band and intent: margin heroes for prospecting, viewed items for recovery, new arrivals for fans, clearance for bargain hunters. Margin-first sets pay salaries while hero sets win attention. Ecommerce playbook weights the mix.

Exclude dying stock from prospecting and give it an honest clearance home with its own pricing and canvas. Prospecting dying sizes burns money and teaches shoppers distrust. Feed health flags availability traps.

Cap collection sets at 50 products and curate harder as the number grows. Smaller assortments with clear intent beat sprawling shelves. Collections UX stages the storefront they open.

Filters that stay maintainable#

Filter on stable attributes such as category, brand, price band and availability rather than fragile title text. Title-based filters break the week marketing rewrites titles. Keep custom labels for margin tiers where native fields cannot express intent. Meta feed and placement specs show which fields each surface reads.

Test each set preview after every feed change: new categories and renamed brands silently empty old filters. A two minute preview saves a two week learning reset. Seasonal sets need this discipline most.

Document every filter rule with its commercial purpose and owner. Future editors should know why a set exists before they touch it. Test notes keep the history honest.

Naming that survives scale#

Name sets for humans at 2am: intent, margin and season in plain words, such as prospecting margin heroes or clearance winter. Codes and initials collapse once teams change. Consistent names also make reporting readable without a legend.

Version seasonal sets by year and phase rather than overwriting evergreen ones. Dated variants keep history clean and comparisons fair. Dated variants plan the year.

Separate country sets where price, language or availability differ. Suffix names by market so budgets never mix currencies by accident. Multi-country feeds multiply the need.

Sets for prospecting versus retargeting#

Give prospecting broad appeal plus margin and retargeting viewed, basket and cross-sell slices, with converters excluded from prospecting. Blended sets flatter acquisition while starving it. Judge prospecting on new customer share and retargeting on recovery rate. Retargeting splits show the structure.

Cross-sell sets pair replenishment and complements with past purchases, refreshed as ranges change. Custom audiences build the exclusions that keep the split clean.

Hold out retargeting periodically to prove incrementality; retargeting always looks brilliant until tested. Conversion Lift settles the argument.

Pruning sets that decay#

Review sets monthly: item counts, click through by set, contribution by set. Empty sets, stagnant sets and sets that open but never tap get archived or rebuilt. Attention without intent is decoration. Fatigue curves show when winners turn.

Promote winning filters across ranges instead of rebuilding from scratch each season. Libraries of proven slices compound like creative libraries. Catalogue creative builds on the same habit.

Ask for a second pair of eyes when sets sprawl past twenty. Our team audits set structure with trading calendars in hand.

Frequently asked questions#