Utsavify: D2C Store Launch with Google + Meta Ads, Tracking & Cart Recovery
Took a festive-gifting D2C brand from catalogue to live ad campaigns on both Google and Meta — a Google-Sheet-driven storefront with 53 products, fully verified conversion tracking, automated abandoned-cart recovery, and a same-day Merchant Center suspension reversal along the way.
The Challenge
A new festive-gifting brand needed to be selling online before the Raksha Bandhan season — with no storefront, no product data pipeline, no analytics, and no ad presence. The catalogue changed weekly (prices, stock, new bundles), the team was non-technical, and every day of delay ate into a hard seasonal deadline. They needed a store the owners could manage from a spreadsheet, and ad campaigns that could be trusted with real budget from day one.
What We Built
We built the storefront so the entire catalogue — 53 products, prices, images, descriptions, SEO metadata — lives in one Google Sheet the owners edit directly; the site and the Google Merchant Center feed both render from that single source of truth, so feed-vs-site mismatches are impossible. On top of that: GA4 + Meta Pixel wired through a config-driven analytics module with the full purchase funnel (view, add-to-cart, checkout, purchase) verified end-to-end via a real test order; a Google Performance Max campaign and a Meta Advantage+ Shopping campaign launched in the same week; a two-step checkout that captures email early; and an n8n abandoned-cart recovery sequence (1-hour and 20-hour emails) that stops the moment the customer completes their order.
How It Works
The foundation decision was making Google Sheets the database. The owners already lived in spreadsheets, so instead of teaching them a CMS, we made the storefront render everything — product cards, detail pages, pricing logic, even meta titles and descriptions — from one Sheet. A price edit appears on the live site and in the Merchant Center feed together, which eliminated the feed-mismatch class of problems entirely and meant zero ongoing developer dependency for catalogue changes.
Before a rupee went to ads, we made tracking provable. A shared analytics module fires every funnel event to GA4 and Meta Pixel from one place, IDs are environment-driven, and page views are wired to the router (the classic single-page-app tracking killer). Then we placed a real ₹149 cash-on-delivery order and traced it through GA4 Realtime, Meta Events Manager, and — after linking and importing — Google Ads' conversion column. Only then did campaigns launch.
The Merchant Center feed is generated from the same Sheet as the site, served as a live XML endpoint. Products cleared review, and we launched Performance Max with fully-loaded asset groups: 20 images, 15 headlines, a 10-second AI-generated product video, 16 search themes, and a GA4 cart-abandoners audience as the signal. In parallel, a Meta Advantage+ Sales campaign went live optimised on the Purchase pixel event.
Mid-flight, Google suspended the Merchant Center for 'Misrepresentation' — the trust-signal audit every new store risks. We executed the full recovery checklist in hours: policy pages at clean crawlable URLs, legal-entity disclosure in the footer matching the GST registration, business information alignment, and Google payments identity verification with the GST certificate and proprietor PAN. The account was reinstated the same evening, and the playbook became one of our most-read blog posts.
Checkout was rebuilt as two steps — contact first (email required), then address and payment — so an abandoned checkout still captures a reachable customer. An n8n workflow watches for abandoned checkouts and runs a two-touch email recovery sequence at 1 hour and 20 hours, cancelling instantly on purchase. The cart itself persists locally for seven days, so returning visitors find their box exactly as they left it.
Post-launch, the campaign metrics validated the setup: 7–11% CTR on Performance Max at under ₹0.50 CPC — strong signals for a brand-new domain in a seasonal category — with both platforms feeding verified conversion data back into their learning systems. The owners run the store from a spreadsheet; the automation runs everything else.
