What drives a successful e-commerce offering? Part II: Three trends in Search & Merchandising
What are current trends in Search & Merchandising? How can your company use them to satisfy customers and retain them in the long term?
In Part I of his two-part series, Fabian Engeln from foryouandyourcustomers highlighted what companies need to offer to their customers in order to achieve a good experience of their e-commerce offering. In this article the Multichannel Consultant will share what customers expect from Search & Merchandising capabilities in e-commerce, both today and in the future, in times of ever faster digitalisation and shortened technology development cycles?
Innovative technologies open up new opportunities to address potential customers, through both existing and new channels. What are current trends in Search & Merchandising? How can your company use them to satisfy customers and retain them in the long term? The following three developments have an increasing importance in the area of Search & Merchandising:
Trend 1: personalisation
Numerous e-commerce companies are already focusing on personalisation based on customer segments. For example, information about gender, age and location is used to personalise the homepage in online shops. That way, customers receive personalised messaging as soon as they can be assigned to a segment via a login or session. E-commerce solutions for 1:1 personalisation based on customer and analytics data go one step further. These enable companies to automatically create highly personalised customer experiences in their online channels. Data collection begins as soon as a user clicks on a specific page element or sends a search query. With this behavioural data, personalised rankings, filters and recommendations can be displayed in the user’s context. For example, product recommendations on a detail page take into account what the user has previously searched for.
Comparatively new in the area of 1:1 personalisation are algorithms that calculate affinity scores for products per user. Collaborative filtering is one of these methods that generates recommendations or personalised product rankings based on user comparisons. A tracker collects data on clicks, views, add-to-baskets or ratings from each individual user. The algorithm compares users with each other and calculates preference scores per user and per product. Let’s say user A is interested in products X and Y. In addition, there are several other users who are interested in products X, Y, and Z. Then, the algorithm is based on the assumption that user A will also be interested in product Z. In the movie and video streaming industry, Netflix.com uses this method to suggest movies and series to users which similar people have rated well. The implementation of 1:1 personalisation can lead to high technical requirements in e-commerce system architectures. Companies with a lower degree of maturity therefore first need a basic infrastructure in order to update data in the digital channels more quickly. If properly implemented, Search & Merchandising solutions can use these advanced personalisation methods to improve their customers’ user experience and conversion rates.
Trend 2: Visual search
Visual search is a rather recent technology that supports multiple use cases in online shops. In most cases, it is enabled through machine learning algorithms that are “trained” with thousands of photos. They are tuned to recognise features and shapes of products. A use case for this is the comparison of product photos within a catalogue. For example, alternative products can be automatically generated for the product shown on detail pages based on its visual properties. This results in a great reduction of effort for online merchandisers, as manual data-based rules do not have to be developed any longer.
Visual search and recommendations at Asos.
Another use case is to start a product search through a photo taken with the smartphone camera rather than by typing in text: Instead of searching for a general term like “blue T-shirt”, users often find more specific results via this visual search form, as it can also recognise attributes such as the collar shape of the shirt. An example of this is the fashion retailer Asos.com. A few years ago, the visual search functionality was integrated into the company’s own smartphone app and made it possible for users to search for clothing items via their smartphone camera. In addition, relevant outfit recommendations are displayed for each individual product.
Trend 3: Voice search
Smartphone owners often use one of the integrated voice assistants Siri or Google Assistant – a feature that is also available on devices such as Amazon Alexa, Google Home and Apple HomePod. With the help of these virtual assistants, users can play music, set wake-up times, obtain weather information or do other things solely through voice input. Why is this relevant for digital businesses? Because voice search is rapidly growing. In a study published by Forbes, by 2020 50% of all searches will be done by voice.
However, voice searches can only provide a limited overview of a broad product range. Therefore, it makes sense to use customer data and predict what users are looking for based on previous purchases or personal preferences. Also, a voice search is more relevant for fast-moving products than for higher priced premium products. From a technical perspective, voice searches pose some challenges in the area of data processing. While users typically search for certain keywords in traditional search boxes, a voice search must first convert the language into text. In addition, the user often speaks complete sentences rather than only a few keywords for voice search. The British online food retailer Ocado.com is an example of the implementation of a voice search. It offers this functionality as an Amazon Alexa skill. For example, when customers are in the kitchen and say “Ask Ocado to add flour to my order”, flour is automatically added to the personal shopping list. This list can then be sent as an order. With this Alexa skill, Ocado simplifies the entry into their customer journey for their users.
To conclude, I see the continuous urge of companies to invest in new technologies in the field of e-commerce search and product access. The more innovative technologies are on the market, the better will companies be able to gather data about its usage and optimise them step by step. Retailers should not lose the opportunity to invest in those, since their competitors might overtake them once they have identified the most useful way to deliver a better customer experience in the area of product access to their customers.