Google Shopping Ads: Setup Guide for Indian D2C

Most Shopping advice starts with bidding.

That is like tuning an engine before checking whether the car has fuel.

In search campaigns you control the ad copy. In Shopping you do not — Google reads your feed and builds the ad itself. Your title is the headline. Your image is the creative. Your price is the offer, shown next to everyone else's.

So this starts where the results actually come from.

The feed is the campaign

Your product feed goes to Google Merchant Center, and Merchant Center feeds Google Ads. Every attribute you send is either helping Google match your product to a query or quietly preventing it.

These are the fields that decide most outcomes.

Write titles for search, not for your catalogue

Your product database calls it "Classic Tee — Blk — M". Nobody searches that.

Because the title is your matching signal and your headline at once, it has to do both jobs. The pattern that works is boring and effective:

1. Brand first — if people search your brand, it belongs at the front. If they do not, put the product type there instead.
2. Then the product type — in the words a buyer would use, not your internal category name.
3. Then the defining attribute — colour, material, capacity, whatever people actually specify when they search.
4. Then size or variant — so variants do not compete with each other for the same query.
5. Front-load the important part — titles truncate. Anything after roughly the first 70 characters is insurance, not signal.
6. No promotional language — "Best seller", "Sale", "Free shipping" in a title is a disapproval waiting to happen.

Campaign structure: start simple, split by what you learn

The instinct is to build an elaborate structure on day one. Resist it — you do not yet know which products carry the account.

Start with one campaign covering everything, with product groups split by your product type field. Let it run long enough to produce data. Then split out the products that deserve their own budget: the high-margin ones, the ones converting well, and the ones eating spend without returning anything.

The reason to split is always control — a product that needs a different budget or a different target than the rest. Splitting for tidiness just spreads your data thinner, the same trap as running too many ad sets on Meta.

Shopping campaigns or Performance Max?

If you are new to Shopping, start standard. You will learn which products and which queries actually produce revenue — knowledge that makes Performance Max far more effective later, and that PMax will not give you if you start there.

The disapprovals that catch everyone

• Price mismatch — feed price differs from the landing page, usually GST or a stale sale price.
• Availability mismatch — feed says in stock, page says sold out.
• Missing GTIN on branded goods — or worse, an invented one.
• Promotional text in images — badges, price stickers, "50% off" overlays.
• Landing page issues — no visible price, no clear add-to-cart, or a redirect to a category page instead of the product.
• Missing policy pages — returns and shipping information must exist and be findable.

Check Merchant Center Diagnostics weekly. Items drop out silently, and a campaign quietly serving half your catalogue looks like a bidding problem for months.

How to judge whether it is working

By ROAS against your actual margin, at product level rather than account level.

Shopping accounts almost always contain a handful of products carrying the result and a long tail spending without returning. An account-level ROAS that looks acceptable can hide both. Segment by product, and be willing to exclude the persistent losers entirely.

What counts as a good ROAS is a margin question, not a benchmark — and in Shopping, where price is visible in the ad, margin pressure is the whole game.

For the organic side of the same catalogue — product page structure, category SEO, technical work — see the e-commerce SEO guide. The two channels share a catalogue and almost nothing else.