At its heart, everything that Netflix does is data-driven and customer-focused – and with more than 204 million subscribers, Netflix has access to extensive data that enables them to identify user needs and preferences. Everything from making ultra-personalised recommendations to users, to deciding which shows to take into production, or trying out new website features is determined by this data.
Whilst an online presence is becoming increasingly essential in order for retailers to survive, it is the leveraging of data, in order to make smart business decisions and deliver excellent customer experiences, that will ensure a thriving retail business. So what can retailers learn from the way in which the streaming giant utilises data?
Using machine learning to unlock the value of data
Understanding your customers is key in today’s world of retail – and businesses now must get to know their shoppers in the way that Netflix knows its subscribers. The streaming service’s data collection method is impressively unobtrusive to the end-user; customers, for the most part, won’t recognise the mechanisms and algorithms at play behind the scenes. The same can be applied to e-commerce platforms. Retailers can monitor visitors’ activity on their website, app and social media pages; collect personal information via an online account; and track a customer’s final basket spend.
Data is only useful, however, if it has also been properly analysed. Interpreting data on customer preferences, behaviours and past purchases, and applying machine learning technologies to this information enables retail businesses to provide highly individualised offers and tailored product recommendations to customers. This has been shown to boost sales, revenues and conversion: 35% of Amazon’s revenue alone is generated via its recommendation engine, and even minor incremental improvements can pay massive dividends. It is in this analysis – of data gathered both in-store and online, and brought together from a range of sources using data transformation – that retailers can find the most value.
Venturers Club roundtable: driving success within e-commerce
How IoT and computer vision help collect more data for bricks and mortar
For bricks and mortar retail, collecting data on customers is obviously more difficult – they don’t need to ‘login’ in order to enter a shop. Retailers, however, can track credit cards to group transactions back to a specific customer, and use this data to link both online and offline sales. AI, when used smartly in store, can also help retailers to get to know their customers better. Connected devices and IoT, along with Computer Vision technology (CV), allows businesses to collect data from sensors, cameras and mobile devices on consumer behaviours. This can include the items that are picked up or put back down, the directions visitors move in, whether the shoppers are regulars, or which areas of the store are most visited. Analysing the data gathered by this suite of technologies can, in turn, help drive brand loyalty with a tailored in-store experience.
Loyalty programmes continue to have a key role to play in supporting retailers with data capture (such as behavioural and transactional information), and analytics are helping loyalty schemes to become more powerful in driving sales than ever before. A vast majority of retailers – from supermarkets, to cosmetics and apparel – are using loyalty cards and programmes to capture and leverage customer data.
Collaborative filtering: the key to hyper-personalised experiences
Every Netflix user’s home dashboard is hyper-personalised according to their preferences identified in the data, and as a result, the subscription service has an impressively low churn rate and an almost unrivalled customer retention. Whilst personalisation has been on the retail industry’s radar for some years, it has become even more important in the COVID landscape: retailers that leverage data to deliver unique and targeted customer experiences will out-perform their competitors.
Fashion retailer ASOS uses machine learning and an algorithmic approach known as ‘collaborative filtering’ in order to recommend products to customers. Collaborative filtering finds customers that have similar purchasing patterns, and then recommends products that other customers like to those who haven’t yet purchased them. It’s not as simple as it sounds – rather than customer segmentation, this activity is in fact a matrix factorisation technique. The brand’s analysis of the ways in which users interact with products enables them to drive better, more personalised recommendations. As a result, ASOS is often cited as the most trusted online-only fashion retailer, and it’s focus on customer experience has helped the business to really stand out from the crowd. Retailers can also explore other types of customer-centric models – such as automated sizing recommendations, which help ensure the customer selects the right sized product (thus customer satisfaction), and also reduces the probability of a return.
Why ‘omnichannel’ retail could be the secret to getting shoppers back in store
Building data-driven products
How do you drive and manage the right product to the right customer, at the right time? Netflix uses its data to determine which movies or shows to buy in, or which to take through to production. Understanding their customers means that the streaming service can confidently develop and release products that they know will be successful with the end-user. By releasing content that customers want to see, Netflix is able to ensure a high standard of customer experience and satisfaction, and further strengthen retention.
A/B testing on Netflix users also allows the business to try out new features and capabilities on their service, the aim being to identify which features should be rolled out – using the data results from testing segments of users to make the best possible business decisions.
Retailers should similarly be developing and rolling out products that are data-driven. Which products will resonate and sell well? How might this differ from one store location to another? Larger in-store retailers have been doing this for some time already – capturing and studying customer demographics data for each local catchment area and optimising inventory and store accordingly. For fashion retailers in particular, analysing data around product lines, stocks and sales levels could prevent the need for end of season fire sales – in which goods that have failed to sell are reduced to a very low price (further reducing profitability). Targeted product development, taking demographic and behavioural data into consideration, will ensure that retailers are only investing in products which will be successful.
Netflix registered a 24% increase in revenue in the year 2019-2020, and a 74% customer retention rate in 2021. The answer to business growth, sales and product innovation undoubtedly lies in the leveraging of data, and this ability to do so will be crucial for retailers if they want to capture the marketplace in the coming years. It doesn’t take 204 million users to provide a decent data set either – but retailers should be gathering information on the customers they do have. With modern technology, it’s becoming far easier to build scalable algorithms, and cloud service providers are ensuring that data analytics capabilities are more accessible to businesses of all sizes. As a result, retailers across the board will likely see a good return on investment on analytics, and insight from this data can enable retailers to grow their revenues significantly. The message is clear – harness data now, or fall behind. Do it in the right way, and your business can save money and grow faster.
