Ecommerce keywords are the searches people type into Google when they're about to buy a product you sell, or when they're deciding which one to buy. They aren't the words you use in the warehouse. They're the customer's words, and in the UK they rarely include the word "buy".
The guides on the first page of Google UK for "ecommerce keyword research" explain the same four keyword families and suggest adding "buy", "price" or "online" to the product name. Google's own AI Overview for that search uses "buy Royal Canin dog food online" as its example of a transactional query. We measured how many searches these modifiers actually get on three products sold online in the UK, using Google Ads search volume data, and compared them with the same modifiers in the US. The result changes a piece of advice you'll find almost everywhere.
How UK shoppers search before they buy: data on three products
We picked three products with different price points and seasons: the air fryer (small appliance, peaks from November to February), running shoes (sportswear, with sizes and genders) and the sofa bed (furniture, a considered purchase). For each one we took the base search and the variants with the modifiers that ecommerce SEO guides call "transactional" or "commercial".
| Search | Air fryer | Running shoes | Sofa bed |
|---|---|---|---|
| Product only | 201,000 | 74,000 | 246,000 |
| + best | 18,100 | 9,900 | 6,600 |
| + sale | 4,400 | 3,600 | 12,100 |
| + deals | 1,300 | 260 | 210 |
| cheap + product | 6,600 | 2,900 | 8,100 |
| + amazon | 2,400 | 260 | 2,400 |
| + price | 1,600 | 30 | 140 |
| + reviews | 880 | 390 | 170 |
| + online | 40 | 90 | 90 |
| buy + product | 590 | 320 | 720 |
| purchase + product | 590 | 320 | 720 |
Three things stand out. First: "buy air fryer" gets 590 searches a month against 201,000 for the base search, about three in every thousand, and "purchase air fryer" returns exactly the same figures. Second: "air fryer online" gets even fewer, 40 searches, because anyone looking for a product on Google takes it for granted that they'll find it for sale online. Third: the modifiers that carry weight are "best", with 18,100 searches for air fryers and 9,900 for running shoes, and the price words: "sale" gets 12,100 searches for sofa beds and "cheap" 8,100. "Best" is a comparison search, not an immediate purchase, and it leads to a different page from a product page.
A technical detail worth knowing. "Best air fryer" and "air fryer best" return identical numbers, month by month: 27,100 in November 2025, 12,100 in July 2026. Google's tool treats them as the same search, and one page covers both. The same goes for "buy" and "purchase" in the table above. Writing two pages for two word orders of the same search is the quickest way to make them compete with each other.
Hardly anyone types "buy", in the UK or in the US
The advice about "buy" comes from American guides, and they're full of examples like "buy nike air max size 10". We set the UK figures against the US market, with the air fryer as the reference product.
| Modifier | UK (air fryer) | Share of base search | United States (air fryer) | Share of base search |
|---|---|---|---|---|
| Base search | 201,000 | 100% | 550,000 | 100% |
| best | 18,100 | 9.0% | 74,000 | 13.5% |
| deals + sale | 5,700 | 2.8% | 6,300 | 1.1% |
| amazon | 2,400 | 1.2% | 3,600 | 0.65% |
| price | 1,600 | 0.80% | 3,600 | 0.65% |
| cheap | 6,600 | 3.3% | 9,900 | 1.8% |
| buy | 590 | 0.29% | 1,300 | 0.24% |
| online | 40 | 0.02% | 140 | 0.03% |
In the UK "buy air fryer" is 0.29% of the base search, close to the US figure of 0.24%. Both are small. When we ran the same test on the Italian market, "comprare friggitrice ad aria" (buy air fryer) came to 0.012%, about 24 times lower. Where the UK really differs from the US is price. British shoppers type "sale" or "deals" two and a half times more often than Americans in proportion (2.8% against 1.1%), "cheap" almost twice as often (3.3% against 1.8%), and they add Amazon's name about twice as often (1.2% against 0.65%).
The price finding has a match on the behaviour side too. In Jungle Scout's 2025 report, based on a survey of almost 1,000 US consumers carried out in December 2024, 63% named "price and discounts" as the factor that weighs most on their Amazon purchases (source: Jungle Scout, Consumer Trends 2025). It's a US sample, not a British one, but the UK search volumes point the same way.
Brands and retailers are worth more than modifiers
The first table hides the biggest part of demand. Variants with the name of a brand or a retailer almost always beat the classic modifiers, and in the UK the gap is very large.
| Search | Monthly searches in the UK | Compared with the base search |
|---|---|---|
| ninja air fryer | 165,000 | 82.1% |
| tower air fryer | 22,200 | 11.0% |
| argos air fryer | 18,100 | 9.0% |
| currys air fryer | 8,100 | 4.0% |
| lidl air fryer | 4,400 | 2.2% |
| nike running shoes | 33,100 | 44.7% |
| asics running shoes | 14,800 | 20.0% |
| ikea sofa bed | 40,500 | 16.5% |
| dunelm sofa bed | 9,900 | 4.0% |
For a multi-brand shop these are the most valuable keywords in the catalogue, and they belong on brand pages and on product pages with the model name in the title. "Ninja air fryer" alone gets 165,000 searches a month, and in November 2025 it reached 301,000, more than the generic "air fryer" (246,000) in the same month. For a business that sells only its own brand these are searches to know about, not to chase: "ikea sofa bed" wants Ikea, and no amount of optimisation will persuade Google to put another site first.
Lidl deserves a line of its own. "Lidl air fryer" went from 1,900 searches in June 2026 to 8,100 in August 2026, while the generic search only moved from 135,000 to 165,000. A retailer keyword can have its own seasonality, separate from the product's.
The four keyword families and the right page for each
The classic split by search intent is still useful, as long as you tie it to a page type. With the data above, this is how it works for a UK online shop.
- Product, brand and model ("ninja air fryer", "asics running shoes"): these go on product pages and brand pages. People searching for them already know what they want.
- Category with an attribute ("double sofa bed", "dual basket air fryer"): these go on category pages or filtered subcategories. This is where an online shop picks up the largest share of its traffic.
- Comparison ("best air fryer", "best running shoes"): these want a buying guide or a ranked list. For "best air fryer" the first page of Google UK shows tested reviews from BBC Good Food, The Guardian, Which?, Mumsnet and delicious. magazine, videos, Amazon's best sellers list and a large block of products. A category page with 40 items in a grid struggles to get in there.
- Informational ("air fryer recipes", 22,200 searches; "how to choose running shoes", 260; "how much electricity does an air fryer use", 170): these go on the blog or in a guides section, with internal links to the categories.
The fifth family that many guides mention, the "transactional" one with "buy" and "purchase", barely registers in our data: 590 searches for the air fryer against 201,000. The base product search is already transactional. For "air fryer" the first page of Google UK is made of shops and brand stores: Amazon, Currys, B&M, Tefal, Cuisinart and SharkNinja, plus a product carousel.
The ecommerce long tail: attributes, sizes, genders
In Ahrefs' US database, keywords with fewer than 10 searches a month make up almost 93% of the total, 2.3 billion queries, as Tim Soulo wrote in his guide to long-tail keywords, updated in May 2026. Ahrefs doesn't publish the same figure for the UK. The principle shows up in our numbers too, with one difference: in ecommerce the long tail isn't made of long phrases, it's made of attributes.
| Attribute added | Example | Monthly searches |
|---|---|---|
| Use | trail running shoes | 18,100 |
| Size | double sofa bed | 14,800 |
| Gender | mens running shoes | 12,100 |
| Gender | womens running shoes | 8,100 |
| Technical need | stability running shoes | 1,600 |
| Technical need | running shoes for overpronation | 1,300 |
| Feature | dual basket air fryer | 720 |
| Feature | cushioned running shoes | 590 |
"Mens running shoes" and "womens running shoes" together get 20,200 searches a month, more than a quarter of the 74,000 for "running shoes" on its own, and "trail running shoes" adds another 18,100. In practice, the filtered category pages carry demand on the scale of the generic one, and if the men's/women's filter on your site creates URLs that can't be indexed, you're giving up some of the biggest searches in the group. This is where choosing keywords becomes a decision about site architecture: which filters deserve a page of their own, with a dedicated URL, title and copy, and which stay as filters and nothing more.
The rule of thumb we use: an attribute deserves an indexable page if it has measurable search volume and the catalogue has enough products to fill it. A "running shoes for overpronation" subcategory with two items in it is a thin page, even if the keyword gets 1,300 searches.
Seasonality: Black Friday and the sales
An annual average tells you little when you need to decide when to publish or update a page. The same data, month by month, tell a different story.
| Search | Peak month | Lowest month | Ratio |
|---|---|---|---|
| air fryer black friday | 9,900 (November 2025) | 10 (June and August 2026) | 990 to 1 |
| sofa bed black friday | 4,400 (November 2025) | 10 (June and August 2026) | 440 to 1 |
| air fryer | 246,000 (November 2025 - February 2026) | 135,000 (June-July 2026) | 1.8 to 1 |
| sofa bed sale | 18,100 (January 2026) | 6,600 (July 2026) | 2.7 to 1 |
| running shoes sale | 5,400 (January 2026) | 2,400 (June 2026) | 2.3 to 1 |
| running shoes | 90,500 (January and April 2026) | 60,500 (October 2025, June-August 2026) | 1.5 to 1 |
In the UK "sale" has one season, from December to January, with Boxing Day and the January sales: "air fryer sale" peaked at 12,100 in December 2025, "sofa bed sale" and "running shoes sale" in January 2026. There's no summer peak to match. "Black Friday" has a single peak, concentrated in November. A page built for these searches needs to be published and linked from the menu weeks before the peak, because Google needs time to crawl it. To see the curves for your own sector, Google Trends and the monthly history in Keyword Planner are enough.
How to do keyword research for your online shop
This is the method we use, in order. It starts from the data you already have and only moves to paid tools when they're needed.
- Start with Search Console. The Performance report shows the queries your site already gets impressions for. Filter by category pages and look for searches with plenty of impressions and an average position between 8 and 20: they're the closest to bringing clicks.
- List your products and their attributes. Gender, size, material, colour, brand, model, use. Each attribute becomes a variant to measure.
- Measure UK volumes. Google Ads Keyword Planner is free; all you need is a Google Ads account. For competitor analysis you need tools such as Semrush or Ahrefs: we compared them in our guide to keyword research tools.
- Check the results page for each keyword before you assign it. If Google shows only articles for "best running shoes", that keyword doesn't belong on a category page. This check is what separates a keyword list from a plan.
- Group. Searches that lead to the same results page go on the same page. Keyword clustering stops two pages of the catalogue from taking rankings off each other.
- Check Amazon and the marketplaces. Amazon's search suggestions show the attributes customers actually type, often before they appear in SEO tools. We cover this in our guide to Amazon SEO.
- Assign one main keyword per page and note the page type next to it: category, product, brand, guide.
Low volumes need a different filter from the one in most guides. A keyword with 30 searches a month, like "running shoes price", isn't worth the work for a big retailer. For a small niche shop, a search of that size in its own niche can be half the market. Keyword difficulty matters at least as much as volume.
Where to put keywords on your pages
Google's ecommerce documentation, updated on 10 December 2025, focuses more on structure than on copy. The guide "Help Google understand your ecommerce site structure" asks you to link the menu to categories, categories to subcategories and subcategories to every product, and to link your best-selling products often from the homepage and from editorial content. In keyword terms: the most important search should sit on the page that gets the most internal links.
On the page itself, these are the places that count:
- the title tag and the H1, with the main keyword at the start and the attribute that sets it apart (mens, double, dual basket);
- the URL, short and stable, with no parameters for the categories you want to rank;
- a short category description, above or below the grid, that uses the attribute variants and links to the subcategories;
- the product name on the product page, with brand and model written the way the customer writes them, not with the internal code;
- product structured data, which Google uses for price, availability and reviews in the results: our guide to schema markup explains how to set it up.
The rest of the technical work on filters, variants and out-of-stock pages is in our guide to ecommerce SEO.
The UK context: a big online market, crowded results pages
According to the Office for National Statistics, 28.8% of retail sales in Great Britain were made online in August 2026, up from 28.4% in July, and online spending was 8.9% higher than in August 2025 (source: ONS, Retail sales, Great Britain: August 2026, published 18 September 2026). More than one pound in every four spent in shops is now spent online.
For a small business this means competing on results pages where Amazon, Argos, Currys, the brands' own stores and marketplaces like Etsy are already present. The keywords you can win are the ones with a precise attribute, where the big generalist retailer has a weak page. We saw this with Macropix, a LED screen manufacturer based in Milan and one of our clients in Italy: none of the nine keywords that bring the most enquiries contains the company name, and on the "ledwall" cluster its share of voice reached 25.55%, ahead of Amazon at 18.08%. The full case is in our guide to SEO for manufacturing.
The mistakes we see most often
- Optimising for "buy" or "online". For our three products these variants stay under 0.5% of the base search, and "online" stays under 0.15% for all three. Filling titles and copy with "buy online" brings no traffic and makes the result less appealing to click.
- One category for every comparison search. "Best air fryer" wants a guide with criteria and named models. If you don't write it, the click goes to a review site that then sends the customer to Amazon.
- Indexing filters without a rule. Indexing every combination of colour and size creates thousands of near-identical pages. Only index the attribute that has volume and enough products.
- Ignoring seasonality. A "black friday" page published in mid-November arrives when searches are already at their peak.
- Descriptions copied from the manufacturer. The same text appears on dozens of sites. A product page stands out with the information the customer looks for and the manufacturer doesn't write: real sizing, running costs, comparison with the previous model.
If you'd rather have someone do this analysis for you, it's the first step of our SEO service for online shops. And if your catalogue also runs on Google Shopping, the same keywords are what you need to set up the campaigns: we explain how in our guide to Google Ads for ecommerce.
Frequently asked questions about ecommerce keywords
The ones that combine the product with a precise attribute (gender, size, feature, model) and have measurable search volume in the UK. In our data "mens running shoes" gets 12,100 searches a month and "double sofa bed" 14,800. For a multi-brand shop, searches with a brand name are often the most valuable in the catalogue: "ninja air fryer" gets 165,000.
Rarely. In the UK "buy air fryer" gets 590 searches a month against 201,000 for the base search, and Google counts "purchase air fryer" as the same search. The product name on its own is already a buying search.
One main keyword, plus the variants that lead to the same Google results page, such as the two word orders of "best air fryer". If two searches show different results pages, they belong on different pages.
Google Ads Keyword Planner is free (you need a Google Ads account) and shows volumes and monthly trends for the UK. Search Console shows the queries your site already gets impressions for. Google Trends helps you read seasonality.
They show how present the marketplace is in the customer's mind: for air fryers "amazon" accounts for 1.2% of the base search in the UK, against 0.65% in the US. They aren't keywords to rank for, but they tell you how much it matters to be on Amazon with the same catalogue.
One last observation. Lists of "100 best ecommerce keywords", like the one published by SEO.com, are built on the US market and carry a way of searching that doesn't match the UK. British shoppers don't type "buy". They type "sale", "cheap", the name of a retailer or the name of a brand, with peaks tied to Black Friday and the January sales that no generic list can know about. Ten minutes in Keyword Planner with the products in your own catalogue are worth more than any list you can download.
