Google Shopping works for fragrance only once the feed is right, and fragrance feeds break in a specific way: size variants. A 30ml, 50ml and 100ml of the same scent submitted as three unrelated products will compete against each other and split your data three ways. Fix variants first, then titles, then bidding.
The short version
- Size variants are the number one cause of wasted spend in fragrance feeds.
- Titles should read the way people search: brand, scent name, concentration, size.
- Separate brand searches from generic ones or brand traffic will flatter everything.
- Shopping suits stockists of known brands far more than new houses nobody searches for.
Is Google Shopping right for a fragrance brand at all?
It depends entirely on whether people search for what you sell. If you stock designer and niche bottles that already have demand, Shopping is often your cheapest revenue. If you are a new house nobody has heard of, Shopping has almost nothing to capture and Meta should come first, as we argued in what it costs to advertise a perfume brand in the UAE.
That is the single biggest strategic decision in this category and most brands get it backwards, starting with Shopping because it feels more measurable and then wondering why volume is so low.
Why do size variants break fragrance feeds?
Because the same scent in three sizes looks like three products to Google unless you tell it otherwise. Without a shared item group ID, they compete in the same auctions, split conversion data between them, and each one learns more slowly than the single product would have.
It also produces a poor shopper experience, with three near identical listings from the same store. Set the item group ID, pick the size you want shown as the default, and let the others sit underneath it.
Concentration matters too. Eau de parfum and eau de toilette of the same scent are genuinely different products at different prices, and should not be grouped as variants of each other.
What should fragrance product titles look like?
Structured the way people actually search. Brand first, then the scent name, then concentration, then size. Someone hunting a specific bottle types most of those words, and a title matching that order gets matched more reliably.
Avoid marketing language in titles. Luxurious, captivating and signature are wasted characters in a Shopping title. The description is where that belongs, and even there it earns less than a clear list of what the bottle is.
How should bidding be structured?
Split brand searches from everything else. Someone searching your store name or an exact bottle you stock converts at a completely different rate from someone browsing the category, and blending them makes both look average.
If you are a stockist, expect a large share of easy revenue from people searching for specific designer bottles. That is fine, but do not let it flatter the account. Look at the non brand segment separately to see whether Shopping is genuinely acquiring customers or just intercepting people who were coming anyway.
On one UAE store we run, which sells a large catalogue with heavy variant complexity rather than fragrance, the last 90 days returned AED 213,502 in conversion value on AED 39,372 of spend. That is a 5.1x, and most of the work behind it was feed structure rather than bidding. Feeds are where Shopping accounts are won.
| Feed problem | What it causes | Fix |
|---|---|---|
| Sizes submitted as separate products | Self competition, split conversion data | Shared item group ID |
| EDP and EDT grouped as variants | Wrong price shown, poor match quality | Separate product groups |
| Marketing words in titles | Weaker query matching | Brand, scent, concentration, size |
| Brand and generic blended in one campaign | Brand traffic hides poor generic performance | Split campaigns and read them separately |
| Missing GTIN on designer bottles | Reduced eligibility and reach | Populate GTIN wherever it exists |
A mistake we made: we inherited a fragrance feed where every size was its own product and celebrated an early efficiency gain after consolidating them. What we had not noticed was that the consolidation defaulted to showing the 100ml, the most expensive option, on every listing. Click through collapsed within a week. Consolidating variants is right, but which variant you surface is a separate decision and we had let the feed make it for us.
If your Shopping account is busy but not profitable, the feed is usually where the answer is. Book a growth call or see how we run Google Shopping for UAE stores.
Frequently asked questions
Usually not at first. Shopping captures existing demand, and a new fragrance house has very little. If you stock known designer or niche bottles people already search for, Shopping can be your cheapest revenue. If nobody searches your name, start on Meta.
Give every size of the same scent a shared item group ID so Google treats them as variants rather than separate products. Without it they compete against each other and split conversion data, which slows learning across all of them.
No. They are genuinely different products at different price points, and grouping them causes the wrong price to surface and weakens match quality. Group by size within a concentration, not across concentrations.
Brand, scent name, concentration and size, in that order, because that is close to how people type the query. Leave marketing adjectives out of the title entirely, since they take up characters without improving matching.
Because they convert at very different rates. Blended together, strong brand performance hides weak generic performance, and you cannot tell whether Shopping is acquiring new customers or simply intercepting people who were already heading to your store.