McDonalds mastered the upsell with one simple question at the time of purchase: “You want fries with that?” A simple and relevant question at the right time that has likely generated millions of extra dollars in revenue through the years for the company. Ever since then, companies have tried to emulate their success by identifying complementary products in their offering and training sales staff to ask customers the right question at the right time.
Today, the generation and tracking of customer data, transaction data and purchase behaviour data are enabling companies to move away from a generic upsell and cross-sell to a personalised one, and machine learning is ensuring data-driven recommendations reach the right customer at the right time.
In this blog post, we look at how Amazon and Hyatt Hotels are using machine learning to personalise their approach and improve upsell and cross-sell effectiveness.
Amazon’s machine learning knows what you “might like”
Amazon, a veritable cornerstone of the online world, still manages to not only on-board new business at an impressive pace, but continues to demand a large share of wallet of existing customers with targeted, data-driven cross- and up-selling strategies thanks to machine learning.
Using data correlation techniques on its massive database of over 150 million customers, Amazon is able to gain insights on past purchases, reviews, customer preferences and product popularity to make relevant and personalised recommendations to users that match their buying history and preferences. On every product page, Amazon is using machine learning to display products other customers bought when they bought the product you’re viewing. By analysing purchase behaviour and identifying patterns, Amazon identifies which products are often bought together and displays the complementary product at the critical moment that can trigger an impulse purchase. For Amazon, their use of machine learning increases share of wallet and total size of purchase, as well as improves the customer experience through the relevance of their recommendations.
Hyatt Hotels’ predictive analytics boosts revenues
In March this year, the Hyatt Group announced that it had aligned its operations to use predictive analytics to improve cross- and up-selling to guests at their 500-plus hotels across the globe.
By analysing guest history and preferences and comparing them to those of guests with similar profiles, Hyatt is able to automatically display relevant messages that tell desk agents that the guest they are checking in is likely to want to upgrade their room to one with a view, or might want to know more about amenities the hotel offers. It’s very similar to Amazon’s “you might like this product” which Hyatt admits was their inspiration for this new data-driven step in the guest check-in process.
According to the group’s SVP for Strategy and Analysis, Chris Brogan, “In 2014 in the Americas, we rolled out a program that has increased the average incremental room revenue, post-reservation, by 60%, 2014 versus 2013. That’s compared with similar programs in the past that lacked the sophisticated analytics.” Among Hyatt’s chief data sources was its membership program that gave the group the per-individual insights they needed to offer special discounts or amenities, based on a particular member’s past traveling, accommodation and other preferences. The success of the group’s initial foray into big data in the Americas has led to the decision to adopt the predictive model on a global scale.
Learning from every new byte of data
As people’s lives go through their natural ebbs and flows, their requirements and expectations are subject to change, and one of the biggest mistakes any business can make is to assume that this never happens for their customers.
The beauty of machine learning is that it is constantly learning from new data and improving its predictive ability with every new byte of data it’s fed. Analysis of historic data can map changes in behaviour back to new recommendations. Knowing and predicting what your customers may want – even before they know it themselves – and meeting that need at the right moment is the competitive-edge machine learning and predictive technologies can provide.
Read our media coverage to learn how we applied the same predictive analytics techniques we use to predict consumer behavior for our clients to predict the this year’s Oscar winners, as well as the results of all of the 2015 Rugby World Cup matches with a 91% rate of accuracy.