Personalisation is spending yet another year as one of the key retail and marketing trends driving success. Gartner cited it as key focus for 2018, calculating that 64 percent of marketing leaders are already using or planning to deploy personalisation technology across the buying journey.
And with figures from McKinsey showing that well executed personalisation can boost total sales by 15 to 20 percent, it’s no wonder why. In the last few months alone, online marketplace eBay has launched a new personalised Interests feature; while US brands Sportsman’s Warehouse and JCPenny are now using live personalised product alerts to allow shoppers to personalise their marketing and re-engage non-buying online customers.
The key to the success of these ventures is of course the data upon which they are built. By linking online and offline consumer data together, marketers can build a stronger profile of a consumer’s preferences and then use it to perform more advanced personalised predictions through analytical models. Digital behaviour can be a strong indicator of product purchase propensity or brand affinity. This can be used to further personalise digital marketing campaigns and offers across both the initial digital channels as well as in the call centre or branch, as well as informing direct outbound campaigns via email and SMS.
We’re now starting to see some brands take personalisation a step further, combining data from internet connected devices to improve the in-store experience. In the US, Nordstrom has been using the consumer’s digital footprint and past purchase history to recommend products in its new Nordstrom Local stores. While Amazon Go sites use facial recognition technology, product pricing data and a user’s Amazon account to allow them to automatically pay just by picking up an item and leaving the store with it. Here in the UK, new billboards being trialled in London use demographic data automatically collected by IoT enabled surveillance cameras and run through personalisation algorithms to select which adverts to serve to the audience around it.
While not all brands will feel comfortable engaging in this level of personalisation now, it shows the market just what’s possible and the level of experience that consumers will eventually come to expect.
However deeply a brand or marketer decides to invest in providing a more personal experience, one of the early and most critical steps in the process is preparing this variety of multi-structured customer data into an accurate data output for downstream use.
Without a reliable data foundation, it’s impossible to conduct analyses that inform when to deliver the right personalised experience at precisely the right time. And more often than not, this lengthy process – which takes on average 80 percent of a data project’s time and resources – is still being done manually through tools like Excel, which can introduce human error.
In a world where consumers expect personalised offers served in an instant, time and accuracy is of the essence. Just one wrong move can mean losing a customer to a rival brand forever. That’s why savvy marketers are investing in intelligent data preparation solutions to reduce the time required to make sense of the raw data which feeds their predictive models.
These tools build on decades of work in human-computer interaction, scalable data management and machine learning. Data wrangling solutions significantly enhance the value of marketing data. Regardless of how unmanageable the data may appear, these solutions can help alleviate the manual work of data wrangling, which usually requires laborious hand-coding. As a result, the analysis process is drastically accelerated, and data that might have been untouchable becomes accessible.
Data wrangling tools quickly format siloed data from transactional, social media, and other sources to turn customer interactions into insights that can be used to optimise marketing efforts. For example, offering personalised services and products based on a 360-degree view of customers and their touchpoints – a capability that Amazon has long excelled at.
Similar analysis can be done on data from social media and web browsers to identify new customer prospects based on insights gathered about their lifestyle, preferences and purchasing habits. When that’s combined with examination of data from CRM systems, marketers can identify attributes of existing customers and target similar groups in the wider market.
Marketers across the spectrum are already using this kind of technology and process to segment the large, rapidly growing and increasingly complex quantities of customer data into personalised customer profiles, gaining a more accurate understanding of their consumer base. For example, The Royal Bank of Scotland has used it to gain a faster and more accurate 360-degree picture of its customers so that it can build more personalised products.
With the amount of data marketers are required to include in their analysis only set to grow – the number of connected IoT sensors and devices is due to exceed 20 billion units by 2020 – so too is the complexity of the road to accurate personalisation. With the economic outlook uncertain for the next few years, now is definitely the time for marketers to strengthen their data foundations to withstand whatever may lay ahead.
