Electronics is the one category where Google Shopping punishes a sloppy feed immediately. Jewellery or apparel can get away with a vague title and a missing identifier. A phone case can't. Someone searching "WH-1000XM5 price" has a model number in their head, Google has a product catalogue keyed to GTINs, and if your listing doesn't carry that identifier it either gets disapproved or quietly loses the auction to a reseller who did the boring work. So before anyone touches bid strategy, fix the feed: GTIN, brand, MPN, condition, and a title that leads with the model number. That single pass is worth more than a month of campaign tinkering.
What follows is how we set up Google Shopping for electronics catalogues on Shopify, including the parts that usually get skipped.
GTINs are not optional, and most electronics stores get this wrong
Google matches your product to a known item using the GTIN. For branded electronics — a Sony headphone, a Lenovo laptop, an Anker charger — that identifier exists and Google expects it. Leave it blank and set identifier_exists: no and you're telling Google the product is custom-made. It isn't. The listing gets limited reach, or it gets pulled.
On Shopify the mapping is straightforward: the variant Barcode field becomes the GTIN, Vendor becomes the brand, and the SKU can serve as MPN. The problem is that most catalogues we inherit have barcodes filled in for maybe 40% of variants, usually whichever ones were imported from a supplier CSV. Everything added manually by the merchandising team is blank.
Fix it with a bulk export, fill the column, re-import. Tedious, one afternoon of work, and it moves more impressions than anything else you'll do that week.
The awkward case: white-label electronics. If you're importing TWS earbuds or power banks from a contract manufacturer and selling them under your own name, you genuinely own the brand and you need to buy your own barcode prefix — GS1 India issues them. Until you do, you're capped. Google treats an unidentified branded-looking product with suspicion, and the misrepresentation review in this category is harsher than in most.
Write titles the way buyers type model numbers
A shopper searching for electronics usually knows more than the shopper searching for a kurta. They type model numbers, storage sizes and wattage. Your title should front-load exactly that.
The order that works:
- Brand, then model number, then product type, then the one spec that varies, then colour. "Samsung Galaxy S24 5G Smartphone 256GB Onyx Black" beats "Galaxy S24 — Flagship Performance, Stunning Display".
- Drop marketing adjectives entirely. They occupy the 70-odd characters Google actually shows and they match nothing.
- Keep the title under 150 characters, but assume only the first 70 get read.
If your Shopify product titles are written for the storefront and you don't want to change them, use a supplemental feed or a feed app to override title for Google alone. The on-site title and the feed title do not have to be the same string, and for electronics they usually shouldn't be.
One more thing: ALL CAPS anywhere in the title triggers an editorial disapproval. So does promotional text burned into the main image — "20% OFF" stickers, watermarks, a logo in the corner. Electronics suppliers love sending images like that. Strip them.
Price is the ranking factor you cannot write your way around
In Google Shopping for electronics, price comparison happens in the results page itself. The same model, from six sellers, side by side. If you're ₹1,400 above the cheapest listing on an identical SKU, you will get impressions and almost no clicks, and your campaign will look like a targeting problem when it's a pricing problem.
Merchant Center has a price competitiveness report. Use it weekly. It tells you the benchmark price for each product, which is the number that actually decides your CTR.
Here's the arithmetic that matters more than the report. Say you sell TWS earbuds at ₹2,499 with a 22% gross margin — ₹550 a unit. Your Shopping CPC is ₹9 and the product page converts at 1.8%. That means 1 ÷ 0.018 ≈ 56 clicks per sale, so 56 × ₹9 = ₹504 of ad spend per order. You've made ₹46. Then a Razorpay fee of roughly 2% takes ₹50, and you're underwater before a single COD return.
That's the normal state of affairs for low-ticket accessories, and it's why so many electronics stores conclude Shopping "doesn't work". The levers are: raise AOV with a bundle so the ₹504 is spread over ₹4,000 of cart value, push the mix toward higher-margin own-brand SKUs, or stop advertising the ₹2,499 SKU and advertise the ₹7,999 one. Not "optimise the campaign".
Custom labels should map to margin, not to category
Google gives you five custom label slots. Most feeds use them to repeat the product type, which is useless because Google already has the product type.
Use them to encode things only you know:
- custom_label_0 — margin band. High, mid, thin. This is the one you'll segment campaigns on.
- custom_label_1 — stock depth. You don't want Performance Max spending on 3 units of a discontinued model.
- custom_label_2 — price tier, in rupee brackets. Under ₹2,000 behaves nothing like over ₹30,000, and a single campaign averages the two into a target ROAS that fits neither.
- custom_label_3 — COD eligibility, if you restrict it by value or pincode.
Shopify's native Google channel handles the basics well. Custom labels, per-variant shipping weights and product highlights usually need a dedicated feed app or a supplemental feed uploaded to Merchant Center from a Google Sheet. The Sheet route is free and perfectly good if your catalogue is a few hundred SKUs and your margins don't change daily.
Variants: storage and colour are not the same problem
Submit every variant as its own item, linked by item_group_id (Shopify maps the product handle here automatically). Then the question is whether each variant has its own GTIN. For phones and laptops, yes — 128GB and 256GB are different GTINs and must be submitted as such. For colours of the same accessory, often the manufacturer uses one GTIN across all colours, which is fine and legal as long as you're accurate.
Where this breaks: stores that model storage as a variant but keep a single barcode on the parent. Google sees four identical items at four different prices and flags the cheapest one as a price mismatch against your landing page. We see it on roughly a third of first-pass audits in this category.
Performance Max, Standard Shopping, or both
The honest answer for an electronics catalogue under about 500 SKUs: start with Standard Shopping. You get search term reports, you can add negatives, and you can see which products are eating budget. Performance Max will probably outperform it on raw ROAS once there's data, but it will also spend three weeks learning on your thinnest-margin SKUs while telling you very little about why.
Negatives matter unusually much here. "Repair", "service centre", "second hand", "driver download", "manual pdf", "vs" — all high-volume, all worthless to a retailer. On a laptop catalogue, blocking the repair and driver queries alone can cut wasted spend noticeably in the first fortnight.
When you do move to Performance Max, use feed-only asset groups first. No headlines, no images, just the product feed. It behaves close to Shopping, and you can compare like with like. Add assets once you know the baseline.
Refurbished stock needs its own treatment. Google has a condition value for refurbished, and using it requires a published warranty and returns policy. Mixing refurbished units into the same campaign as new ones at a lower price looks like a price mismatch from the outside and gets accounts suspended.
The click is expensive, so the landing page has to carry it
Shopping clicks in electronics are not cheap, and they land on a product page that has to answer a spec question in about four seconds. Three things decide whether that happens.
Price parity. Feed price, cart price and the price visible on page load must match to the rupee. GST-inclusive display, any app that applies a discount at cart rather than on the page, any currency-rounding rule — all of these produce mismatches. This is the most common cause of Merchant Center suspension we deal with, and it is entirely self-inflicted.
Speed. A product page carrying four review apps, a sticky bar, a size chart and a chat widget will not render its price before a user on a 4G connection loses patience. We treat anything over 2.5s LCP on mobile as a paid-media problem rather than a dev problem, because that's whose budget it's burning. If you want a starting point, our notes on Shopify speed optimization cover the usual suspects in theme and app weight.
Comparison without leaving. Half of electronics shoppers arrive on one SKU and want to compare three. If your collection pages can't filter by RAM, storage, wattage or screen size, they go back to the results page and click a competitor's listing. Spec filters are the difference between one expensive click and a session. We built FilterPro for exactly this — metafield-driven filters and search on large, spec-heavy catalogues — because stitching it together in the theme every time got old.
India: COD, GST and the Amazon floor
Three things change the maths for Indian electronics sellers.
COD inflates your apparent conversion rate and deflates your real one. If 45% of your Shopping orders are COD and 18% of those are refused at the door, your true ROAS is about 8% lower than Google reports. Feed that correction into your target ROAS, or set the target for COD-heavy SKUs higher on purpose.
GST-inclusive pricing has to be what the feed shows. If your theme displays ex-GST anywhere for B2B buyers, make sure the feed price and the default storefront price are the tax-inclusive ones.
And the floor: on mainstream branded electronics, a large marketplace will frequently be cheaper than you, visible right beside you, and willing to lose money on that SKU. You will not win that auction on price. You win it on SKUs the marketplaces don't carry well — own-brand, bundles, newly launched models, accessories with genuine spec detail, and anything where the buyer wants a real warranty conversation. Build the campaign structure around those and let the commodity SKUs sit in a low-budget catch-all.
A sane first 30 days
- Week 1: fill every barcode, fix brand values, rewrite feed titles to Brand + Model + Type + Spec. Clear all disapprovals before spending anything.
- Week 2: Standard Shopping, one campaign, split into two ad groups by price tier. Manual CPC or Maximise Clicks with a cap. Collect search terms.
- Week 3: build the negative list from real search terms. Add custom labels for margin. Pull the price competitiveness report and reprice or de-prioritise whatever is more than 5% above benchmark.
- Week 4: move the profitable segment to target ROAS. Test one feed-only Performance Max asset group against it on the high-margin label.
Most of the gain lands in weeks one and three. The campaign settings are the part everyone wants to talk about and the part that matters least.
If you want a second pair of eyes on a feed before you scale spend, we do a free store audit that covers Merchant Center diagnostics alongside the storefront. And if the product content itself is the weak point — thin spec tables, no comparison pages, nothing ranking organically for model numbers — that's a content and SEO job, and it compounds in a way paid clicks never do.