e-consultancy’s Online Retail 2004 survey has some interesting findings

Written by: Jonathan Briggs

November 23, 2004 [3287 views]

New media research company e-consultancy commissioned Sales Logiq to look at the current state of major UK retailers such as Comet, John Lewis and Argos. At the end of September 2004 they published a report that is available to subscribers to the e-consultancy site. It makes interesting reading and I thought I would summarise a few of the key points in order to provide an additional perspective to my students recommending improvement to similar sites.

Use of the home page
One interesting aspect of the study was an evaluation of the amount of screen space on each home page given over to a number of different types of content: navigation, support and promotion. For example John Lewis was reported to use 7.5% of home page area for navigation while Argos used 34%, Comet used 2% for support versus Amazon’s 17% and Dialaphone used 64% for Promotion against John Lewis’ 89%.. Navigation was defined as links, search and index, support included services, product guides, basket, FAQ, gift finder, T&Cs etc and promotion included branding, product promotions, offers, catalogue order etc. While you might argue with the exact taxonomy it seems to be useful way of studying retailers and might be a good metric for other investigations.

The report argues that on many sites, the different uses of space are poorly delineated (compared for example with newspaper layouts) and suggests clearer differentiation, simplification and better use of space (including white space).

Link trails
All stores want to help their customers find what they need but few do it well. This report suggests that we need to identify three different types of users and then design paths or trails through the site to meet their requirements. The three different types of users are trackers (who know exactly what they want), hunters (who know the sort of thing they want) and explorers (who want to discover things).

Every site should incorporate navigation and service devices to meet these different styles of exploration: trackers want accurate search by catalogue or model number, hunters what to search by description and explores want to browse for example. The report suggested that customers would only click on links when they feel they are making progress toward their goal. Incorrect or inappropriate labelling of links, products or categories will stop visitors in their tracks.

Search results
The comments on search are particularly illuminating with the major criticism being the limited help provided by most sites in helping customers distinguish between results: few pictures, poor filtering, poor descriptions, no prices. How is a customer for a digital TV to make a choice when 200 TV’s are returned all of which look similar? 56% of search results pages in this survey were found to present significant usability problems.
The authors argue for significantly more information to be provided in results pages to help customers make their decisions and tools to help with the decision support such as allowing selection of candidate products and then providing comparison. They evaluated the following aspects of search: was the search term displayed with the results, search string still in search box (to allow editing), number of results shown, can be sorted, can be filtered, pictures of product, brand info, product info, model, price, saving shown, availability, descriptions, can add to basket directly. One interesting observation was that not one of the sites selling cameras made any specific recommendations when searching for a 4 megapixel camera. This is a missed selling opportunity.

Search tolerance
Another interesting test was looking at how well search worked on each site. While most sites were tolerant of upper and lower case, partial model numbers, spelling mistakes and poor formatting were generally poorly handled. Another area for attention.

Supporting the shopping process
The report examines some specific recommendations from other sources for improving the shopping cart. This is another useful checklist:

• In the UK call it a basket
• Don’t make them register before they buy
• Use add to basket instead of buy
• Provide access to the basket from every page
• Include links back to the products within the basket
• Avoid making additional offers as a product is added to the basket
• Confirm each item has been added
• Let customers remove items easily
• Let customers return to where they were when they continue shopping
• Save the contents of abandoned baskets for future visits
• Show the number of steps
• Show the total cost of purchase (include delivery) before making them register.

Most sites in the survey complied with most of these suggestions with the exception of continue shopping, links from basket to product, confirm add to basket and saving abandoned baskets (which was considered controversial).

Promotion
E-commerce allows a large amount of personalisation and computation; regular customers can be rewarded, shoppers can be upsold and additional accessories and add-ons can be offered.

The survey was surprised how few companies make any significant use of targeted promotions. There is very limited analysis of what is in the shopper’s basket or attempts to engage the customer in further business. This is a major area of opportunity.

Future stores
The report concludes with some recommendations for features of improved stores in the future. Here are some of them for you to consider:

1. Think about improving how products are categorised to make it easier to find things
2. Provide filtering and decision support tools to help people make choices
3. Provide interactive product guides and combine these with feature filters
4. Consider adding meta-data to product databases to allow more manipulation of products in the light of customer behaviour and improve search and filtering
5. Provide synonyms to increase the power of the metadata
6. Make it easier for the customer to justify their purchases through extra information on features, comparisons and benefits

Finally
This is an interesting report and I have only pulled out a few of the ideas that I think will be relevant to my student’s assignments. It is worth joining e-consultancy if you are interested in reading the whole thing.

Recent comments:

What do you think?

Add your comments