Prospecting introduces an offer to people who may not know the store. Retargeting addresses people who have already shown interest and remain eligible to receive ads. The right split depends on the store’s audience, margin, purchase cycle and usable signals, not a universal percentage of budget.
Define the audience’s last known action
Separate recent product viewers, cart starters, previous buyers and broader prospective customers. Confirm which audiences can actually be built from consented, permitted data and whether they are large enough for a meaningful test. A tiny list may not deliver useful evidence even if it looks precise on paper.
Match the message to the stage
A prospecting creative should explain what the product is and why it is credible. A retargeting creative can resolve a known question about fit, delivery or returns. Both should lead to a page that answers the promise and has the right product available. Exclude people where a repeated reminder is unhelpful or inappropriate.
Compare outcomes without double counting
Validate purchases against store orders before judging either route. Compare audience overlap, frequency, spend and verified orders in a defined period. Platform attribution alone cannot prove an incremental sale; use holdouts or other controlled comparisons when volume and setup allow. Flower Center’s published case supports checking measurement before changing the ad account.
Worked decision example
Illustrative decision: a new store has almost no eligible visitors but a strong product demonstration. Begin with a constrained prospecting test, then consider retargeting only after a meaningful audience exists. A mature store with many cart starters may first investigate checkout friction before paying to re-engage them.
Practical review checklist
- Are audience sources permitted and usable?
- Does each creative answer the stage-specific question?
- Can purchases be reconciled to real orders?
Where to go next
Meta Ads service · Meta purchase-event guide · Flower Center case