Google Ads · focused work
Ecommerce search campaigns
Product listings answer a shopper who already knows what to buy. Text search campaigns answer the rest: someone describing a need, asking which type suits a job, looking for a replacement part, or comparing two options. The ad is written by you, the landing page is chosen by you, and every query you pay for can be seen and controlled.
This is the right work if
- Product listings carry all of the store's paid search and nothing serves category or use case demand
- Campaigns were built from a keyword tool export with no landing page assigned to each group
- Ad groups hold dozens of loosely related keywords sharing one generic ad
- The store solves a problem people describe in words rather than search for by product name
- Search terms show queries the store cannot fulfil and no negative list is stopping them
What it is
What a search campaign build covers
This is the paid search work that sits beside product listings rather than inside them. It starts with a keyword set grouped by what the searcher is actually doing: naming a product type, describing a use, describing a problem, comparing two options, or looking for a replacement. Each group is then assigned a landing page before a single ad is written, because the page decides whether the click can convert, and the most common failure in a store's search account is a well-built ad group pointing at a page that answers a different question.
From there the build is structure and control. Ad groups stay narrow enough that the ad can repeat the language of the query. Responsive ads carry assets that state things the store can prove, such as the range it holds, where it ships and how returns work. Negative keyword lists are shared across campaigns so a query removed once stays removed. Bidding works toward conversion value with targets calculated from contribution margin. And because the same query can be eligible in a Shopping or Performance Max campaign, budget and overlap are planned rather than discovered later in a report.
How it is done
The work, in order
What changes
- Demand described in words is served instead of being left to product listings
- Every ad group points at a page that answers the query it bid on
- Shared negative lists keep spend on searches the store can actually fulfil
- Text search revenue can be read apart from brand and from product listings
Sort demand by what the searcher is doing
Split the keyword set into product type, use case, problem, comparison and replacement queries, and move anything containing the store name into its own campaign. Assign each group a landing page at this stage, and note any group whose page does not yet exist.
Keep ad groups narrow
Build groups around a single idea so the headlines can echo the words in the query and the destination stays consistent. A group that needs three different landing pages to be honest is three groups, and splitting it usually costs nothing in volume.
Write ads with substance
Use headlines that state range, delivery scope, returns window, sizes or fitments stocked, and anything else a shopper weighs. Pin assets only where accuracy requires it. Add sitelinks to the categories a searcher would want next and structured snippets listing the brands the store carries.
Cover the catalog tail without keywords
Where the range is too large to keyword by hand, use dynamic targeting against category pages or a page feed built from the catalog. Keep it in its own campaign with its own negatives so it fills gaps rather than competing with the groups you built deliberately.
Control queries and bid on value
Review the search term report weekly, add negatives as shared lists, and check where a query is also eligible in a listings campaign. Bid toward conversion value with a target set from margin, and hold the target steady long enough for the bidding to learn.
Platform notes
Shopify
Collection URLs make a clean page feed for dynamic targeting, and the collections that exist decide which use case groups have somewhere to land, so campaign planning often produces a short list of collections to create.
WooCommerce
Product category and attribute archives can both serve as landing pages, and whichever is used needs indexable, readable content on it, because a bare archive tends to convert poorly for a query that described a need.
Should search campaigns bid on terms already covered by product listings?
Sometimes, and it should be a decision rather than an accident. Overlap is worth keeping where the text ad can say something a listing cannot, such as stock depth, a fitment checker or a delivery window. Where the listing already wins the query cheaply, spend the budget on demand it cannot reach.
Which page should a search ad send to, the category or the product?
Send product name queries to the product and everything broader to the category, because a shopper describing a type wants to see the range. If the category page is only a grid with no guidance, fix the page before raising the bid, since the click is being paid for either way.
Do search campaigns make sense for a small catalog?
They can, when the products answer a described need rather than a named item. A store with twenty products in a niche category often has more use case and problem demand than product name demand, and text ads are the only format that can address it directly.
How often should the search term report be worked?
Weekly while a campaign is new, then at a rhythm that matches spend. The point is not only removing waste. The converting terms belong in ad copy, in landing page headings and sometimes in a new ad group, so the report is a source of structure as much as a cleanup task.
Related work and answers
Find the leak.
A free Growth Analysis ranks what your store should fix first, by revenue at stake.