GOOGLE SHOPPING KEYWORDS: WHY THERE ARE NONE

The single most common misconception I correct with new e-commerce clients is the search for google shopping keywords. They open a Shopping campaign expecting the same keyword box they know from Search, with match types and keyword-level bids, and it isn't there. That's not a bug or a hidden setting. Shopping simply doesn't work that way, and once you understand why, everything about how you optimise it changes. This article is about that mental shift, drawn from years of running both Search and Shopping side by side on live accounts.

In Search, you tell Google which queries you want by writing a keyword list. In Shopping, you don't tell Google anything with keywords, you show it your products, and it decides which searches those products deserve to appear against. The lever moved from a keyword field to your product data. That one architectural difference explains why feed quality obsesses experienced Shopping managers and why "add more keywords" is the wrong instinct for a Shopping campaign entirely.

I'll be direct throughout about what actually governs your query eligibility, what levers you really have, and why the title of a product does more work than any keyword ever could here. If you've been treating Shopping like Search with pictures, this is the correction that unlocks it.

Why google shopping keywords don't exist

Let's be precise, because the wording matters. There is no positive keyword targeting in a Shopping campaign. No keyword list, no exact/phrase/broad match types, no keyword-level bids. Google is explicit about this in its own product documentation: Google's About Shopping ads documentation states plainly that Shopping ads use your Merchant Center product data, not keywords, to decide how and where to show. Read that once and the confusion evaporates: the input you're looking for was never in the campaign, it's in the feed.

The reason Google designed it this way is retail-centric management. A merchant with 40,000 SKUs can't realistically write and maintain a keyword list per product; it would be unmanageable and instantly stale. So Google inverted the model: instead of you predicting every query, it reads the structured data you already maintain for each product, title, type, category, identifiers, and matches that against real searches at query time. For anyone coming from Search, this is the hardest habit to break, because the instinct to "add the keyword" has nowhere to go. Understanding what Google Shopping is at a structural level is what finally makes the absence of keywords feel deliberate rather than missing.

The practical consequence is that all the energy you'd spend on keyword lists in Search moves onto your product data in Shopping. That's not less work, it's different work, and it's the work that decides whether your products surface for the searches that convert. This is exactly why so much Shopping effort goes into optimising the product feed rather than tuning a campaign: the feed is the only place the query-matching signals live.

How google shopping works without keywords: what governs the queries

If keywords don't govern eligibility, what does? Your product data, read as a whole. When someone searches, Google matches that query against the text and structured attributes in your feed and decides which of your products are relevant enough to enter the auction. In practice, your feed is your keyword list, just expressed as descriptive data instead of a targeting field. The clearer and more query-shaped that data is, the more often, and more accurately, your products surface.

The heaviest signal is the product title, which I'll come to in its own section because it deserves it. After that, the structured attributes carry the load: product type, Google product category, brand, GTIN and the other identifiers. These do two jobs. They disambiguate, telling Google a "mouse" is a computer peripheral and not a rodent or a Disney character, and they categorise, placing your item in the right competitive set. Missing or vague attributes quietly narrow the queries you're eligible for, which is why I treat completeness as non-negotiable, not cosmetic.

The third factor is Google's own matching judgement, which you don't control directly but you do influence through data quality. This is the part people find frustrating because there's no dial to turn, only inputs to improve. That's also the reason the honest answer to "how do I add keywords to a Shopping campaign" is unsatisfying: you don't add them to target, you shape the data Google reads. Once you internalise that the query set is an output of your feed rather than something you dictate, you stop looking for the missing keyword box and start fixing the data that actually moves eligibility.

The product title: the closest thing to a keyword

If Shopping has a keyword, it's the product title. It's the single richest text signal Google reads about an item, and in my experience rewriting titles is consistently the highest-return change you can make to which queries a product becomes eligible for. When I inherit an underperforming feed, titles are the first place I look, because generic or brand-first titles leave enormous relevance on the table.

The principle is to front-load the terms buyers actually type and mirror real search language: product type, key attributes like colour, size, material, and brand, ordered by how people search rather than how your PIM happens to store them. A title like "Nike Air Zoom Pegasus 40 Men's Running Shoes Black Size 10" tells Google exactly which searches the product belongs in, far more than "Pegasus 40" alone. This is the closest analogue to keyword targeting Shopping offers, and it works through accurate description, not bidding, so precision beats stuffing every time. Strong Google Shopping product titles do the work a keyword list would do in Search, without any of the maintenance overhead.

What I explicitly avoid is keyword-stuffing titles with unrelated terms to chase volume. It reads badly to shoppers, can trip policy, and Google's matching is good enough that irrelevant padding rarely helps and often hurts click quality. The goal is a title that's both human-readable and query-complete: everything a buyer needs to recognise the product, phrased the way they'd search for it. Get that right across the catalogue and you've done the equivalent of an entire keyword strategy, one product at a time.

Negative keywords: the one keyword-like lever you keep

There is exactly one place keywords still appear in Shopping, and it's subtractive. In Standard Shopping campaigns you can add negative keywords to stop your products showing for searches you don't want, irrelevant terms, wrong intent, or queries that historically don't convert. Negatives don't target; they exclude. They refine the pool Google already matched your products into, rather than choosing that pool. That distinction is the whole reason they coexist with a keyword-less targeting model.

On the accounts I manage, disciplined negatives are often the fastest way to clean up wasted Shopping spend. You pull the search terms report, find the queries burning budget with no return, and exclude them. It's the closest Shopping gets to the query control you'd have in Search, and it pairs naturally with the negative keyword workflow for Shopping campaigns, which is where I'd point anyone who wants the operational detail. The catch: Performance Max gives you far less of this, usually only account-level negatives via support, so if granular query exclusion matters to you, that's a real argument for Standard Shopping.

Treat negatives as ongoing hygiene, not a one-off. Search behaviour drifts, new irrelevant queries appear, and a negative list left untouched slowly loses its edge. A short weekly pass through the search terms report keeps the excluded set current and your matched traffic clean, which matters more in Shopping than Search precisely because you can't proactively pick the queries in the first place.

My workflow when there are no keywords to add

Put together, here's how I actually run query optimisation on a Shopping campaign given there's no keyword field. First, I use keyword research not to build a list to upload, but to learn how buyers phrase their searches, then I bake that language into titles, product types and descriptions. The research still matters; it just lands in the feed instead of a campaign setting.

Second, I attack the titles: rewritten to front-load buyer language, completed with the attributes that disambiguate, ordered by search intent. Third, I make sure the structured attributes are complete and accurate, product type, Google product category, brand, GTIN, because these decide the competitive set each product enters. Fourth, I run the search terms report to see which real queries triggered which products, and I use it two ways: to add negatives that trim waste, and to spot where a title needs rewording to capture better intent. That report is my feedback loop; it tells me exactly how my feed is being matched.

If I had to compress google shopping keywords into one line, it's this: stop looking for the keyword box and start treating your feed as the keyword layer. Understanding how google shopping works without keywords, product data matched to queries, means your leverage is descriptive accuracy, not a targeting list. Write titles the way buyers search, complete your attributes, exclude the wrong queries with negatives, and let the search terms report guide the next edit. Do that and you get the query control Search users chase with keyword lists, earned instead through data. That, not a hunt for a missing field, is the craft of Shopping.