{{ searchResult.published_at | date:'d MMMM yyyy' }}

Loading ...
Loading ...

Enter a search term such as “mobile analytics” or browse our content using the filters above.


That’s not only a poor Scrabble score but we also couldn’t find any results matching “”.
Check your spelling or try broadening your search.


Sorry about this, there is a problem with our search at the moment.
Please try again later.

Improving your site search and navigation can have a dramatic impact on sales. Not being able to find the product you want is one of the top - and sometimes overlooked - barriers to conversion.

In this article I've attempted to summarise what I see as the most important steps to improving the customer experience. While some require specific technology, many of these ideas can be implemented on any website.

Improve your zero search result pages

It's important to keep the customer's journey going by offering them different options at each stage of their journey through your site. If they encounter a search results page where no products are returned, and no alternatives are provided, they're likely to go elsewhere.

An optimised zero-results page might display your current top selling products, provide links to key product categories, and give customers advice on how to best use your search facility. See how Sears does it here.

Dynamic navigation refinement

Get customers to the product page in the minimum number of clicks

Imagine the parallel with a high street store. If you're in mind to purchase, you don't want to spend ages trawling the aisles before finding the items you want.

So getting customers to relevant product within the smallest number of clicks is important. Review your taxonomy and page design so that you can reach any product within three clicks.

Direct customers to relevant products with dynamic refinement options

Dynamic navigation refinements are pretty much a given for the best e-retailers these days. Refinement options give the customer the opportunity to narrow down their product selection according to product attributes such as size, colour, brand and price.

Its success lies in the fact that it very much mirrors the way that customers tend to shop offline. Once you've found the product type that you want to buy (say jeans), you then tend to narrow down your product choice according to certain criteria - the brands you prefer, the budget you have to spend, the size that's going to fit you.

ASOS is a good example here. There are various search and navigation packages available that provide this feature (see below).

Invest in an "intelligent search" package

A decent search and navigation tool will do a lot of the work for you when it comes to optimising your "searchandising", offering features such as custom landing pages, drag and drop of search results, easy methods to promote particular classes of product (e.g. certain brands), as well as total control of your front-end page navigation.

Offer a variety of navigation methods

Different customers will expect to navigate in different ways. Some will expect to find product arrange according to product type (clothing, accessories, homewares...), others will expect brand listings, while some product categories lend themselves to certain navigation choices (parents often look for Toys according to their age suitability, for example).

Multichannel retailers should examine how their customers search for products in store. Do they tend to ask for a specific brand, product type, or something else?

Rank products by sales, newness, conversion

The way in which you order your product results when customers are navigating through the site can offer an excellent selling opportunity. Pushing your higher converting products above the fold is a great technique.

Newness is an important pull within the fashion sector, while you might also want to consider sales or stock levels as a suitable metric. This product sorting is a feature of most high-end search and nav products, although it can almost always be achieved another way if you don't have access to these tools.

Rank site categories using analytics data

Take a look at how Argos orders its navigation. It's certainly not alphabetical - there's something smarter at work here.

Use your analytics to drive decisions around ordering your site navigation. This shouldn't be a one-off task - search and nav needs to be regularly reviewed to ensure that it's keeping pace with your product ranges and customer behaviour.

Custom sort options

Offer custom sort options

Allow your customers to sort their search and nav results in they want that they want - whether it's by ascending/descending price, alphabetical order, by product popularity or, if you offer them, customer reviews.

Also allow manipulation of other options such as result pagination if possible.

Encourage advanced searching

Most customers will tend to search for two or three word phrases, but some will want to do more, particularly if you have a large product catalogue.

Offer Boolean search options if possible (e.g. searching on Google for "handbag" OR "clutch bag" will show you search results containing either the word "handbag" or "clutch bag" or both).

Colour searching is essential for fashion retailers - see how effective this can be by searching John Lewis for the term "red jumper". Style searching takes things on even further - ASOS has started introducing terms like "slouchy" and "over the knee" into their search keywords.

Present non-product search results along with product results

Consider the range of things that customers look for with a search box. It's not just products that they're after.

See how John Lewis deals with the search term "oxford". Not only are we seeing Oxford Shirts, but also details on the Oxford Street store. All bases are covered. You can use the same technique to link to buying guides and product advice.

Cross-sell and upsell, ideally pre-basket

Use search and nav data to drive cross-selling opportunities. There are various ways to do this: manual options involve linking specific items in your catalogue together (e.g. cameras and camera cases), while more advanced tools will do the work for you, for example by examining common combinations of products that customers buy together.

Always look for opportunities to upsell, particularly if you're in the world of branded retail.

Aim to do this type of promotion before your customers get to the basket. Taking Dr Mike Baxter's best practice ideas on board here, it's best to keep your basket uncluttered and avoid leading your customers away from their purchase journey by confusing them with alternative products.

Use customer recommendation boxes during the search and nav journey (people who bought/viewed this also bought...)

The most elementary form of social shopping, customer recommendation boxes are commonplace on e-commerce sites for a good reason, customers like to know what's popular, and crowd behaviour dictates that each sale of an item, when advertised, tends to attract another.

Search box suggestions

Consider pros/cons of search-box suggestions

Some retailers now offer a drop down list of recognised search terms or quick links that appear when the customer starts to enter a term.

This has advantages (can help speed up searches, helps reduce misspellings) as well as disadvantages (tends to limit the range of search terms).

Create spelling associations to help with misspellings 

Many search tools allow you to create linguistic associations between terms and their misspellings. The best will have common misspellings already built in. Consider misspellings that might be likely with your product range e.g. complex brand names.

Check that search and nav results agree

In most cases, it's best practice to ensure that the product results obtained through using the search box and navigating/refining are the same.

For example, on a department store website, I'd expect to see the same result set whether I clicked down the navigation path Men's > Jeans and refined to black jeans, or I searched for the phrase "mens black jeans".

Some customers will always prefer either to search or navigate through your catalogue, so it's essential that you're not presenting different purchase opportunities to these two different groups of customers (unless you have more information about the customer and you're able to direct that customer segment towards a tailored set of search results).

Location, existence, method of search box and top/side/footer nav

The principles of Steve Krug's "Don't Make Me Think" apply here. Your search box should be at the top right of the screen. Why? Because that's how the majority of websites do it, so that's where your customers will expect to see it.

In terms of navigation, similar principles apply. Your visitors will expect to see your top level categories laid out either horizontally across the screen, or vertically on the left hand side of the screen. The navigation should be rendered in resizable text for the purposes of accessibility, and you should avoid Javascript or flash menus like the plague.

Footer navigation is a great place for links to static pages, such as contact details, delivery and returns information, terms and conditions and corporate information.

Example breadcrumb trail

Use clickable breadcrumbs 

Breadcrumbs are normally used in one of two ways: either to track and reflect the customer's search and navigation journey, or to signpost the customer's current location in your product hierarchy.

Note that these might not be exactly the same thing - for example if you redirect a search for "fragrances" to a landing page, the customer might find themselves in your category section "Beauty > Fragrances" without having clicked through that navigation journey.

The breadcrumb should be clickable to allow the customer to step back to a larger set of search results.

Ensure your product hierarchy speaks the customer's language

Make sure that your hierarchy and copy talks about products in the way that a customer might. For example, the phrase "small domestic appliances" rolls off the tongue of Head Office Buying teams nationwide, but customers are much more likely to be searching for a "toaster" or "iron".

Similarly, just because your office is organised such that one team is responsible for buying men's and women's jeans, you probably wouldn't want to display them at the same time in search results.

Change sort strategies in line with seasonality and promotions

Another example of the ugly term "searchandising" (the concept of optimising your search and navigation to help with the merchandising of your products) is to change the way you sort your product results according to commercial factors.

For example, if you deal with seasonal product, it might be beneficial to promote new products at the start of the season, high converting products in-season, then products where you have high stock levels at the end of the season.

Test and test again

As with any changes to the front end of your site, testing the performance of your search and navigation will yield useful information. On a regular basis you should test your search box by entering your top search queries and considering the set of results that are returned.

Put yourself in the customer's shoes, do you see what you expected to see? If not, how could the results be improved? What else might you want to see on the page?

If you're making major changes to your search and navigation, it's worth testing it out on a group of real customers. This needn't be an expensive exercise. Many customers are more than happy to offer constructive criticism by taking part in a focus group, particularly if you offer an incentive for taking part.

Another good option, particularly during the early stages of any changes, is to put some of your colleagues in front of the site and ask for their feedback on your site search and navigation.


Hopefully you'll get some food for thought from the ideas above. Have a look at your competitors, test out their search and navigation, and see how you compare. Then go forth and optimise!

Chris Moffatt

Published 31 March, 2011 by Chris Moffatt

Chris is Online Director for Kidcount and a contributor to Econsultancy. You can connect via LinkedIn or Twitter

7 more posts from this author

Comments (3)


Georg Spielmann

Hi Chris,

this is an outstanding article!
I sometimes have a hard time convincing customers not to rank alphabetically... I even believe that the custom sort option alphabetically A-Z or Z-A does not add any value, since the names come from product managers that speak a different language than customers. So the names usually don't mean anything to a customer.


over 5 years ago


Andrew Griffiths, CEO at FindWAtt.com

Great article with some often overlooked points.

One in particular (check that search & navigation results agree) stands out to me because it's (a) rarely true (b) typically an indicator of a data problem. Comparing search results to those from navigation is one of the first tests we make to see whether a retailer has taxonomy (hierarchy) and/or product data problems.

As you said, Dynamic Navigation Refinements (a.k.a faceted search, faceted navigation) are pretty much a given for the best etailers. What's not yet a given is uniformly "good" product data to drive Site Search & Navigation - even on the best etailer's sites.

"Good" product data means "Structured Product Data" where a product has the attributes to drive the facets/filters. Yet seeing these facets in the sidebar is frequently deceptive - some products may have a value for a facet but that by no means equates to every product that should have a value for a facet actually having a value. And totting up the individual facet counts for a facet and comparing them to the number of search or navigation results is no solution as most facets (except for generic ones such as brand) don’t apply to all products. When it’s critical (e.g. with high-value items) to get facet counts right we recommend assigning the facet value “Not Applicable” to products for which the facet does not apply (and setting up the site search & nav software so that “Not Applicable” doesn’t show to customers). It’s then obvious (from blank values) which products are missing values.

Of course, getting your product data right can be a lot of work and unless you’ve got only a small number of products (or perfect data from your suppliers!) you need an automated solution to pull out facet values and ensure that the totals are correct.

over 5 years ago



Interesting but pretty average article. Very much common sense and the points that aren't are too vague to be really helpful.

over 5 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Daily Pulse newsletter. Each weekday, you ll receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.