Identify the actual Grab advertising route
A restaurant promoting its GrabFood storefront and a brand buying media across the Grab ecosystem are not the same brief. Interfaces, eligibility, objectives and approvals differ. We document the advertiser type, account owner, connected outlets, tools visible now and the person authorized to approve changes. If a requested placement is unavailable, the plan says so instead of presenting it as an agency-controlled feature.
Reconcile ad spend and orders before optimizing
For each outlet we preserve the period, spend, ad-attributed orders, total orders, sales, average order value and promotion cost. We test whether advertising and organic categories overlap, and whether time zones, currencies and attribution windows align. ROAS is used only with an agreed advertising-revenue definition; total sales movement is not automatically credited to media.
Review the storefront and operation beside media
Advertising cannot compensate reliably for closed hours, unavailable bestsellers, weak images, an uncompetitive price, a narrow delivery radius, cancellations or slow preparation. The review therefore covers listing quality, menu visibility, top items, promotion mechanics, rating, availability and the operating signals the account exposes. Marketing and operating causes receive separate owners and deadlines.
Scale one controlled commercial test
We choose one hypothesis—budget, bid, offer, menu set, daypart or area—and define the control period and stop condition before making the change. The next review uses a working range rather than only one previous week. More spend is approved only after commission, discount and contribution margin are considered, so a higher order count does not hide weaker economics.