Data is at the heart of e-commerce strategies, but knowing how to leverage it effectively is key. While the tech giants have made it their main asset, e-retailers also have a true goldmine at their disposal: their own website and its data. Yet, between Analytics, CRM, and margin tracking, one crucial resource is often overlooked: the product catalog.
In this third article, we’ll explore how to structure and activate your data to optimize your offering and refine the shopping experience. Thanks to new AI technologies, it is now possible to anticipate consumer expectations and adapt merchandising in real time—a powerful lever to sell better, without necessarily selling cheaper.
You need better data!
Data has become a tangible and monetisable asset, which is why big tech companies are doing everything they can to collect it, use it and, above all, keep it.
There’s no doubt that data is now king. With the advent of the digital age, its is all around us. For years we have been talking about data as the oil of the 21st century, but initially it was hard to understand what that might mean.
Today, between mobile, e-commerce and the digitisation of the entire value chain, everything has become clearer. According to Gartner, in 2020 each person produced 1.7 megabytes of data per second. That’s the size of a 2-minute MP3 file.
This state of affairs has not gone unnoticed by tech giants like Google, Apple, Facebook and Amazon, who are fighting over your data - whether you have shared it voluntarily or not. Data is now so central to the business models of large organisations that it is considered an asset in its own right by Big Tech companies as well as by financial analysts.
Online retailers - your website and its components are a real gold mine for measuring, understanding and improving your business performance, provided you are able to exploit them.
When we think of data for online stores, we immediately think of analytics, CRM data, profit margin ratio and, to a lesser extent, product catalogues… Without searching exhaustively, the three main things to consider in relation to your data should be WHAT, HOW and WHY.
The WHAT question relates to your source of data and its reliability. If the data quality is low, your results will be worthless - as the Americans say, “garbage in: garbage out”.
The HOW is a question about the nature of the data, what format it is in and how you can make use of it successfully.
Finally, the WHY relates to the purpose of using this data.
For example:
- the use of data from unsuccessful searches to enrich the range of products which aren’t in the catalogue,
- the use of product data to anticipate disruptions in stock.
The example below offers you a concrete illustration of how new technologies for behavioural prediction can take advantage of unexplored areas of your product catalogue.
Take the example of an online seller specialising in spirits. During the product data classification process, the retailer took care to fill in as many attributes as possible. At this stage it is possible to add a great deal of value to your online sales offering.
For its “rum” category, the wealth of data made it possible to create very different attributes, starting with the most obvious - such as region of origin, colour (white, brown) and price range. Following a deeper dive with top in-store salespeople, however more detailed attributes emerged:
- the ingredients: the presence of spices or molasses; whether production was organic or not
- maturity: single casks, resting time, vintage product marketing
- attributes: numbered editions, box presentation or additional gifts
Thanks to this in-depth reading of the product catalogue, retail focused Artificial Intelligence can propose the products most suited to the purchasing intention of each shopper.
Based on responses to what is on offer, business algorithms qualify the purchasing motivations in real time and bring the most relevant rums to the top positions by combining an infinite number of possibilities:
- Is this a gift or a purchase for yourself? What is the appropriate price range?
- Are we talking about celebratory consumption or savouring?
In addition to increasing your sales, this is a real opportunity to present your products according to the specific search angle of each customer.
Real world example
To illustrate our point let’s take the real-life example of two of our customers in the consumer goods sector. We were able to spot that with the same frequent search, the first company was limited by the quality of its reference data and only offered two filter options, while the second provided nine. The second customer had a rate of filter usage three times higher than the first. Knowing that conversion rates are on average doubled when a filter is activated during search, we can understand how itemising every tiny aspect of our products can have a direct impact on turnover.
Cheat-Code 3 : Offer your visitors the chance to understand the full detail of your product range
Photo credit: ©istock ©shutterstock