These days, there is an online shop for just about any product: food, cosmetics, clothes and even medicines can be ordered around the clock on the Internet. Aside from an appealing design and fast loading times, the search box is a key element in such shops.
We’ve all seen it – the small field at the top right or in the middle of the screen, often accompanied by a magnifying glass or the word ‚Search‘. Users are intuitively familiar with the search box and the subsequent list of search results, but few realise just how incredibly powerful these search engines really are. Search terms are reduced to their stems and compared against content and meta data in site articles, while suggestion functions along the lines of ‚Did you mean …‘ are now par for the course for any professional organisation.
It’s even possible to refine individual results by means of a replacement strategy, whereby search terms are partially modified: searching for ‚Television‘, for example, may call up the results for ‚4K television‘, ‚LCD television‘, ‚Smart TV‘ and so on. All this is made possible by operator-defined relevance models or settings for specific user profiles. As a result, bargain hunters searching for ’special offer‘ or ‚clearance‘ will always be presented with the best and most recent offers.
Fast results thanks to filter function and guided navigation
Even the best relevance ranking cannot predict the life situations of every visitor to a site. That’s why tools are needed to call up the information the visitor wants quickly. In just a few steps, this can be done with filters and guided navigation. In this case, results lists are sorted in such a way that the visitors can find the information they’re looking for instantly. With guided navigation, the users are never directed to a final dead end; they are continually forwarded according to their own wishes.
Hierarchy of search products
Also of interest are search products for which filters and faceted classification create a hierarchy: subordinate levels are assigned to certain filter values, so that when the user selects a filter value, the next level down is automatically displayed as a new filter. Filters of this kind may be used in the automobile area, where brands, models and products can be distinguished by type.
Even wines can be classified as red, white or rosé, with the filter ‚red‘ calling up the values ‚Merlot‘, ‚Bordeaux‘ and so forth. If a customer then chooses ‚red‘ to narrow down the search, he will be invited to choose from the filters of Merlot, Bordeaux and so on. It is therefore possible to present complex filters with numerous filter values in a clear way. The main advantage for users over separate facets and filters is a greatly simplified display.
Customised recommendations from search-based content
Modern searches also provide context: for example, searching for a type of red wine may also call up a selection of dishes suited to the chosen wine from a database of recipes. The overall effect intensifies the motivation to purchase, even where a recipe database suggests suitable wines for particular dishes. The important point is that the application can be incorporated into various partner sites relatively easily (ideally by dragging and dropping a pre-defined embed code).
To give one example, a portal offering information on recipes can include dynamic content from a wine shop in a straightforward way. In addition to the currently displayed recipe and list of ingredients, the reading conduct of a visitor browsing for recipes and dishes can also prompt the search engine to offer suitable wines from the wine shop in the background.
To give one example, a German-language portal offering information on recipes can include dynamic content from a wine shop in a straightforward way.
Automatic suggestions for suitable wines
Search-based and dynamic landing pages
Today, intelligent search solutions are capable of generating not only text extracts but entire landing pages – and the variables for dynamic landing pages are more diverse than ever. This means the virtual shopping experience can now be ‚made to measure‘ and aimed at the specific needs of individual visitors. Leading manufacturers are offering the possibility of generating landing pages that display search-based information at low cost. Contextualisation of this kind has been a major trend among search function developers for many years.
Utilising context for targeted landing pages
Thanks to contextualisation, intelligent search solutions can create targeted landers, thereby distinguishing between implicit and explicit context. Implicit context is information derived from user conduct or the environment: it includes typical activities in social networks, search and click patterns, main activities performed and the age and location of the visitor. In fact this and other, similar information regarding the user is collated before the actual search.
By contrast, explicit context is derived directly from the search terms entered in a company system as well as selected sort and filter criteria. In defining these and other criteria, the user enables the search engine to produce the most precise results possible. To an increasing degree, the boundaries between applications are blurring: today, for instance, intelligent search solutions automatically take account of social media, stored PDF documents, databases, pictures and videos. Publicly available data from the machine-to-machine network could also be integrated in future. The Internet of Things may well lead to new and expanded scenarios.
Critically examining search solutions
Those selecting a solution need to critically examine not only the added value for site visitors, but also the way in which the business is organised and how it treats the security of sensitive data. Is a cloud service offered? Where does the data come from and what kind of security standards are applied? As an alternative to cloud solutions, some producers offer appliance boxes with pre-installed software that can easily be integrated into existing infrastructure.
Search technologies provide an excellent basis for market-leading companies that have been confronting precisely these challenges for more than a decade, and coming up with many highly accurate solutions. In particular, sequential questions leading to dynamic landing pages promise major benefits. These days, a web site must be capable of creating a ‚dialogue‘ with a visitor that is convincing enough to ensure the visitor accepts the offer and returns to the site.