Google Ads keyword strategy for ecommerce

Which keywords should an ecommerce store target in Google Search ads?

For an online store, the right keyword is not the one with the most searches, it is the one closest to a purchase. Group keywords by intent: brand searches, product and model searches, category searches and problem-led searches. Put most of the budget on the first two tiers and scale broad category terms only once conversion data exists.

This guide covers the keyword side of a Search campaign once your Google Ads account and product catalogue are ready. If you still need the account basics, start with our Google Ads setup guide for ecommerce.

Why does search intent matter so much in ecommerce?

Two people searching for the same product can be at very different stages. Someone typing "how to iron a linen shirt" wants information. Someone typing "men's white linen shirt size L" is a few clicks away from checkout. Bidding the same on both means paying for clicks that were never going to buy.

In practice it helps to think about ecommerce searches in four tiers:

Tier	Example search	Intent	Suggested approach
Brand	"yourbrand linen shirt"	Very high	Separate campaign, low-cost protection
Product and model	"men's white linen shirt"	High	Exact and phrase match, product page
Category	"linen shirts"	Medium	Phrase match, category page
Problem and use case	"shirt that keeps you cool in summer"	Low to medium	Broad match tests, collection or content page

The examples are illustrative. Build the same table for your own catalogue so every keyword has a clear tier and a clear destination page.

Where do you find keyword ideas?

Your own data comes first. On-site search logs, the names of your best sellers, questions customers send to support and the organic queries in Search Console all show how people actually describe your products. We explain how to read organic query data in our Google Search Console guide for ecommerce.

The second source is Google's Keyword Planner. According to Google, it suggests keywords related to your products or website, shows estimated monthly searches and shows the average cost of targeting a keyword. Treat these estimates as a starting point; real volume and cost only become clear once the campaign runs.

The third source, once you are live, is the search terms report. It lists the searches that actually triggered your ads and is the most reliable source of new keyword ideas.

Which match types should you use?

Google Ads has three keyword match types:

• Broad match (no symbols): your ad may show on searches related to your keyword. Google also uses signals such as the user's recent search activity, the content of your landing page and the other keywords in the ad group.
• Phrase match ("in quotes"): your ad may show on searches that include the meaning of your keyword.
• Exact match [in brackets]: your ad may show on searches with the same meaning or intent as the keyword. It gives the most control but reaches the fewest searches.

Google states that it is critical to use Smart Bidding with broad match. The logic is simple: broad match opens you to a wide range of searches, and only a bid strategy that learns from conversion data can tell which of them are worth paying for.

A practical sequence for ecommerce: start with exact and phrase match on high-intent product and model terms. Once conversion tracking is solid and the account collects steady sales, open broad match on category and problem terms in a controlled test.

How should you structure campaigns and ad groups?

Once the keyword list exists, build the structure around intent tiers:

1. Separate brand. Brand searches convert at a much higher rate, so mixing them with other terms distorts averages and lets budget drift toward brand.
2. Create campaigns by product group or category. Keep products with similar margins and price points together so a single bid target makes sense.
3. Keep ad groups tightly themed. One ad group, one product type or category, so ad copy and landing page match the search.
4. Map each keyword to the most relevant page. Model searches go to product pages, category searches go to filtered category pages.
5. Add negative keyword lists from day one. Exclude patterns with no purchase intent, such as "free", "used" or "how to", from the start.

Avoid over-fragmentation. Every campaign and ad group needs enough data for Smart Bidding to learn. An account split into hundreds of ad groups, each with a handful of clicks a month, gives neither you nor the algorithm a useful signal.

Example structure

Example: a home textiles brand could run a lean structure like this.

Campaign	Ad groups	Match types	Landing page
Brand	Brand name, brand + product	Exact, phrase	Home, category
Duvet covers	Cotton, satin, double	Exact, phrase	Category and filtered pages
Towels	Bath towels, beach towels	Exact, phrase	Category page
Discovery (test)	General home textiles	Broad + Smart Bidding	Category page

How does Quality Score affect keyword choices?

Quality Score runs from 1 to 10 and looks at three components: expected clickthrough rate, ad relevance and landing page experience. Google's help page is explicit that Quality Score is not an input in the ad auction; it is a diagnostic tool.

The diagnosis is still useful. If ad relevance is below average, the keyword probably sits in the wrong ad group or the copy does not match the query. If landing page experience is below average, the keyword needs a more relevant page. Tight ad groups and good page mapping are therefore part of keyword strategy, not an afterthought.

How should you read the search terms report every week?

Keyword strategy is never finished. The search terms report shows the real searches behind your clicks. Google notes that some terms with low query volume are omitted for privacy reasons, so the report is not a complete list, but it is more than enough to steer by.

A simple weekly routine:

1. Sort the last 7 and 30 days by cost.
2. Flag terms that spend without selling and add them as negatives where appropriate.
3. Add well-converting terms that are not yet keywords as exact or phrase match.
4. Find terms landing on the wrong product and move them to the right ad group.

What are the most common mistakes?

• Spending on very generic one-word terms because they have high volume.
• Keeping brand and non-brand in one campaign and overestimating ROAS.
• Running broad match with manual bidding or incomplete conversion tracking.
• Sending every keyword to the home page.
• Ignoring the search terms report for months.

If you would like a second pair of eyes on your keyword structure, you can request a free growth analysis from the Performetic team through our contact page.

Key takeaways

• Tier keywords by purchase intent, not by search volume.
• Start high-intent product and model terms on exact and phrase match.
• Use broad match only with Smart Bidding and reliable conversion tracking.
• Separate brand campaigns and map every keyword to its most relevant page.
• Use Quality Score as a diagnostic, not as an auction input.
• Review the search terms report weekly and add negatives and new keywords.