Taking personalisation to the next level
Identifying how much personalization to offer – and to whom – will separate winners from losers.
Hyper-personalization is one of three areas we focus on as part of the digital consumption cross-industry theme. The other themes we examine are products and services to experiences and ownership to access.
More than 70% of customers now expect more personalized experiences with the brands they interact with,¹ and digital technology is enabling companies to meet these expectations by delivering personalization to large numbers of customers at a low cost.
Spectacular advances in artificial intelligence (AI) and software intelligence are enabling companies to take personalization to the next level, making products and services highly relevant to a very large number of customers at the same time.
Personalization is not a new concept – companies have for a long time offered users the opportunity to shape or customize products/services, characterized by offerings such as My M&Ms and NikeID. However, another form of personalization is becoming increasingly prevalent, in which customer data is analyzed in real time to deliver more relevant interactions. An example of this type of personalization is Netflix’s Emmy Award-winning engine for generating recommendations, which drives almost 75% of viewing activity on the site.²
As the case studies below illustrate, other companies are also using personalization to underpin innovative offerings.
Macy’s and Shopkick³
Macy’s has partnered with Shopkick, a shopping app with more than 13 million users, to offer personalized recommendations, location-specific deals and rewards (known as ‘kicks’) to shoppers. Since 2013, the shopBeacon service, which uses Bluetooth Low Energy beacons to detect when users enter the store, has been trialed at Macy’s flagship stores in New York and San Francisco. The shopBeacon service can welcome a shopper upon entering the store, even if the Shopkick app is not open on their smartphone. To protect customer privacy, data is anonymized and aggregated. ShopBeacon technology significantly lowers the average cost of engaging a customer and also boosts the conversion rate for engagements, reducing the cost of transactions.
Duolingo is an AI-powered platform that offers personalized language learning to its 100 million users around the world. Valued at $470 million and backed by funding from Google Capital, Duolingo currently offers courses in 24 languages. It breaks lessons into short blocks and brings elements of gamification to the learning experience. Data about how strongly a concept has been learned is used to customize the user’s next lesson. The app is free for users, with language certification tests only costing $20, far cheaper than standard language exams. It also offers a service for schools that provides teachers with a dashboard to track the progress of individual students.
Ginger.io is an AI-driven mobile application for mental health patients that safely and securely uses data from a patient’s everyday mobile usage (time spent on calls, text messages sent) and activity (distance traveled, sleep) to map behavior over time and then track variations from patterns as potential risk triggers. The app in several cases can predict signs of depression for individual patients up to two days before outward symptoms manifest. Currently, it offers programs to help people with chronic conditions such as bipolar disorder, depression, schizophrenia and anxiety.
By continuously monitoring the user, Ginger.io claims it is more effective at targeting care when the patient really needs it than regular visits to a clinic would be. In this way, the Ginger.io service has the potential to improve clinical outcomes while reducing healthcare costs. Ginger.io has raised $28.2 million in funding so far with the last round in December 2014.
Customer data and hyper-personalization
As the possibilities for different types of personalization – and the specificity with which it can be delivered – increase, companies are facing a new challenge: deciding how much personalization to offer and to whom. The first type of personalization (customers choosing how to customize their products) usually does not require businesses to access significant amounts of customer data and is largely uncontroversial, especially as customers have usually opted to customize their product or service in the first place. The second type of personalization (delivering hyper-personalized services, offers or recommendations) relies on regular access to customer data and advanced analytical abilities.
Regular access to customer data is a key requirement for most types of hyper-personalization but many customers are reluctant to share personal information. Up to 90% of customers would limit the access to certain types of personal data and would stop retailers from selling their information to third parties, while 88% would prefer to determine how their data can be used.⁶ At the same time, some customers are more willing to share certain kinds of data than others.
Company strategies on hyper-personalization also need to factor in the fact that preferences about levels and forms of personalization vary significantly across customer segments. Customers do not appreciate all forms of personalization and are likely to switch if personalization becomes invasive.
Imperatives for companies
1. Establish the right incentives for customers to share data. Collecting real-time customer data is the first hurdle that companies must overcome. Both understanding how to directly incentivize customers to opt into data-sharing initiatives and establishing partnerships or platforms to gain access to second-party data will be critical for companies looking to successfully deliver on hyper-personalization.
- Deliver a clear value proposition. Persuading users to share personal information relies on providing them with tangible and clear value in exchange for their data. Research by Microsoft suggests that 49% of global consumers are aware that companies benefit from their data, but they don’t know how to trade for value in return. Here, emotive drivers of value can be as effective as monetary drivers – these include a sense of achievement, order or discovery.
- Establish customer trust. Communicating trust and security has become especially important, both when asking customers to opt in and when using secondary data from other sources. Customers are more likely to opt in if they feel confident that their data will not be misused or stolen. The Personal Genome Project has redefined ‘informed consent’ by requiring a perfect score on a test before participants can even enter their name.⁷ Companies can alternatively form partnerships or join platforms such as Genecloud that securely extend access to customer data across companies or industries.
2. Draw insights from ‘digital segmentation’. Different degrees of personalization will need to be defined for different customer segments, but companies must first re-evaluate their segmentation strategies. Customer segmentation models will need to change from traditional approaches limited to standard demographics to more sophisticated approaches incorporating digital behavior. Customer data platforms empowered by AI or machine-learning tools, such as Umbel, can help companies address segmentation in this way at a hyper-personalized level. Insights can be captured in real time to feed back into personalized experiences for individuals or customer segments that are deemed to be high-value.
Companies considering investing in personalized offers need to understand the degree of personalization that would maximize lifetime value for each customer. For many, this will be a function of their willingness to share data and their level of engagement with the brand.
3. Identify the right level of personalization to offer to individual customers. Identifying whether a large part of your customer base is geared toward hyper-personalization based on their individual data versus customization is only part of the challenge. Understanding and delivering the right level of personalization to individual customers (and at low cost) is where there is greater differentiation scope for businesses.
Analytical tools have been developed to enable companies to rely on techniques such as attribute analysis to map individual customer attributes based on emotive and psychographic factors; event sequence analysis to map individual customer journeys prior to and after a purchase; and collaborative filtering to identify information for decision making based on collaboration among multiple data sources.
The final decision on personalization (see Figure 1) will still have to be based on the company’s assessment of the impact of these initiatives on customer lifetime value, and consequently the return on investment in new software tools and processes. In many cases, especially where companies have to choose the level of personalization to offer, this analysis will have to be done at individual customer levels rather than at a larger segment level.
6. Accenture Retail Personalization Survey, March 2015
Digital consumption is one of four cross-industry themes (along with digital enterprise, societal implications, and platform governance) that have been the focus of the World Economic Forum’s Digital Transformation of Industries (DTI) 2016 project. An overview of the DTI program can be found here.
Our in-depth analysis of the digital consumption cross-industry theme is available in a white paper, which can be downloaded here.
To explore a selection of related articles and case studies, please select one of the tags below.