Forrester research finds that consumers who engage with a retailer’s site search are 2-3 times more likely to convert.
The data that users provide in these search queries can be used to fuel personalisation, yet too many retailers miss this opportunity through a sub-par site search experience.
61% of sites require the user to search using the exact same language as the jargon the retailer uses. For example, the synonymous nature of “trainers” and “sneakers” may go undetected.
46% of sites do not support thematic searches (such as “summer dress ideas”).
27% of sites can grind to a halt if users misspell even a single character of their query.
If retailers are not getting the basics right, how can they expect to deliver on more advanced areas of ecommerce search?
Users are more likely to convert if they use site search because they have a clear purchase intention in mind. But if retailers overhaul their search experience to make it a source of inspiration and discovery, they can encourage much deeper interaction. In turn, they can convert this data into truly personalised fashion search.
What does personalisation mean in fashion ecommerce today?
The definition of ‘personalisation’ in online retail is changing to incorporate the stylistic elements that make fashion what it is.
A crude summary of a person’s name, demographics, and past purchases does not equate to a personal taste profile. In truth, customers hold preferences for different cuts, styles, and shapes that dictate their eventual purchases.
We now have the technology to capture these preferences and match them to personalised recommendations from a retailer’s inventory.
This starts by defining the characteristics that blend to form a taste profile.
What are the components of a shopper’s taste profile?
Visual preferences: The way a consumer interacts with – and shares – images on a website can reveal clues about their visual preferences. For example, do they prefer certain patterns?
Purchase behaviours: Past purchases by this shopper (and similar users) add context to their current browsing patterns.
Styles: Does the shopper prefer formal, playful, or informal styles? This is a more qualitative element of fashion, but it is crucial to understanding an individual’s tastes.
Brand/product affinity: The brands and product categories that the consumer holds in high regard.
Fit preferences: Does the shopper prefer a looser fit, for example?
These elements can all change over time and the retailer’s multimodal search recommendations must adapt accordingly. Yet they also provide the basis to create a consistent experience for each user, tailored to their longer-term preferences.
Other data can contextualise these components in real-time for each visit:
Shopping intent: Does the consumer have a specific product in mind, or are they open to ideas?
Occasion: Are they shopping for a particular occasion that will dictate their interests for this session?
How can brands find out these consumer preferences?
There are different types of data to consider, which can all be captured in a variety of ways.
Ask: Retailers can encourage consumers to interact with the website and adapt the experience on their own terms. This creates zero-party data (also known as “declared data” or “trusted data”) that the brand can use for personalised search results.
Observe: Retailers can track customer interactions with their products to capture first-party data. It is important to consider this data when designing a website, as brands can structure their content in ways that lead to more insightful customer information.
Infer: Rather than waiting to capture individual consumer data, retailers can pick up on cues from a consumer as soon as they visit a website. They can then use data from other, similar users to start the recommendation process. This avoids the need for a “cold start” with each new consumer.
How Cadeera brings personalised search to fashion retailers
Cadeera develops a taste profile for individual users by blending multiple data input sources. Our technology also augments a retailer’s inventory data to ensure that it can be matched the correct user intent states. This means fashion brands can not only capture preference data in real time, but also guide each user on a personalised shopping journey to the right products every time.
