Make search visible on every page, return sensible results for misspellings and synonyms, and read the search terms report monthly. Terms returning nothing are the most valuable data your shop produces.

When search stops being optional

With a small catalogue, a menu is faster than typing and search is decoration.

Somewhere above roughly fifty products that reverses. The menu becomes a hierarchy people have to guess their way through, and typing what you want becomes the quicker route.

By a few hundred products, search is how a substantial share of buyers navigate, and its quality is a direct factor in sales rather than a convenience.

Many shops cross that threshold without noticing, and keep treating search as a box in the corner.

Searchers are your best visitors

Worth knowing because it changes how much attention this deserves.

Somebody who types a product name has already decided what they want. They are further along than somebody browsing a category, and they convert at a noticeably higher rate in most shops.

Which means a poor search result is failing your most committed visitors, at the moment they were ready to buy.

It also means the loss is invisible in the usual way: they do not complain, they conclude you do not stock it, and they buy elsewhere.

Make it findable

The most basic failure and still common.

Search hidden behind an icon that opens a panel, absent from product pages, or present only on the homepage.

It should be visible on every page, as an actual box on desktop where there is room, and reachable in one tap on a phone.

Where space is tight, a magnifying glass is acceptable provided tapping it opens a field immediately with the keyboard ready, rather than navigating to a separate page.

What good results require

The second is where most small shops fail, because the site uses one vocabulary and customers use another, and nothing connects them.

The no-results page is doing damage

A page saying no products found is a dead end, and it is where customers leave.

It should never be empty. Offer the closest matches, the categories most likely to contain what they meant, a prompt to check the spelling, and a way to ask a person.

That last one is valuable beyond the individual sale. A customer who asks whether you stock something is telling you about demand, and giving them an easy route to ask converts a dead end into information.

Where you genuinely do not stock it, saying so plainly and suggesting the nearest alternative keeps the relationship, which silence does not.

The search terms report

The most underused data any shop holds.

Your platform or analytics records every term typed into your search box, and almost nobody reads it.

It is a direct, unfiltered list of what customers want, in their own words, with no sampling and no interpretation required.

Three things to look at each month.

Terms returning no results, which are either products you do not stock or products you do stock under a different name.

Terms used frequently, which should probably be categories or navigation items rather than things people have to search for.

And the vocabulary itself, compared with the words on your site.

A worked example

A garden supplier looked at their search terms for the first time after two years.

The most searched term returned nothing. It was the common name for a product they stocked and listed under its botanical name.

Four of the top twenty terms were misspellings of the same word, all returning nothing because the search required exact matching.

Two terms were for products they did not stock at all, searched often enough to suggest genuine local demand.

They added common names alongside botanical ones, enabled fuzzy matching, and ordered in one of the two missing products.

Sales through search roughly doubled in a quarter, entirely from products they already had, and the new line sold out twice in its first season.

Fixing vocabulary rather than search

An important distinction, because the instinct is to configure the search engine.

If people search for a common name and your products are listed under a technical one, the honest fix is to put the common name on the product.

Synonym mapping in the search tool solves it for the search box and does nothing for anybody arriving from a search engine, reading a category page, or scanning your catalogue.

Adding the word customers use to the product name or description fixes all of those at once.

Use synonym mapping for genuine alternatives and abbreviations. Use rewriting for cases where your vocabulary is simply not the customer's.

Search is a research tool for the whole business

Beyond the shop, these terms answer questions other departments pay for.

What to stock next, which is the demand signal in the no-results list.

What words to use on service pages and in advertising, since these are the terms your buyers actually use.

And what your customers assume you sell, which is occasionally a surprise and occasionally a business opportunity.

Ten minutes a month, and it is the cheapest customer research available to a shop of any size.

The counter-case

A shop with twenty products does not need any of this.

Search on a small catalogue is a box people rarely use, and investing in matching, synonyms, and results tuning is effort spent on a route almost nobody takes.

There is also a version of this that goes too far, where a shop installs a sophisticated search product with recommendations and personalisation on a catalogue that a decent menu would handle.

The signal to act is usage rather than catalogue size alone: if a meaningful share of visitors are using the search box, it deserves attention proportionate to that share. If almost nobody uses it, the navigation is working and the box is fine as it is.

What to do

  1. Check search is visible on every page.
  2. Search for three of your own products, misspelled.
  3. Open your no-results page and see what it offers.
  4. Pull the search terms report for the last year.
  5. Fix the vocabulary on products, not just the synonyms.
  6. Promote frequent terms into navigation.
  7. Read the report monthly from now on.

Step four usually produces at least one finding worth acting on the same week.

Matching your customers' vocabulary is covered in words your customers use.


Frequently asked questions

When does internal search start to matter?

Somewhere above about fifty products, when the menu becomes a hierarchy people guess their way through. By a few hundred, search is how a substantial share of buyers navigate.

Why do searchers matter more than browsers?

Because somebody typing a product name has already decided what they want and converts at a higher rate. A poor result fails your most committed visitor at the moment they were ready to buy.

What should a no-results page do?

Never be empty. Offer closest matches, likely categories, a spelling prompt, and an easy way to ask a person, which turns a dead end into a demand signal.

What is the search terms report?

A record of every term typed into your search box. It is a direct, unfiltered list of what customers want in their own words, and almost nobody reads it.

Should I fix search or fix my product names?

Usually the product names. Synonym mapping fixes the search box only. Putting the customer's word on the product also helps external search, category pages, and your catalogue.

Does a small shop need good search?

No. With twenty products, search is a box people rarely use. Judge by whether a meaningful share of visitors actually use it, not by catalogue size alone.

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.

Never read your search terms?

Pull the last year. The top term returning nothing is usually something you already stock under another name.

Start a Conversation