Shopping online for clothes is a great convenience, but not being able to virtually try on clothes is a problem. One company, ASOS, has solved it. ASOS offers free delivery and free returns to mitigate customer fears about buying the wrong size. Then, to further improve the experience, the company uses data about past purchases that weren’t returned in order to determine the proper size for customers going forward. The learning algorithm can then take this information a step further and determine customers’ sizes across all the different clothing brands the store carries — with all their unique sizings — so customers can rest easy about ordering the right size in the future.
By using data to personalise and improve the customer experience, ASOS provides great customer experiences, while also reducing costly returns. As they put themselves in the shoes of the customer, the people at ASOS show marketers everywhere a customer experience worth emulating.
Overcoming the challenge of understanding customer expectations.
Traditionally, marketers have viewed their customers in terms of channels. This has led to a siloed view of marketing and a disjointed experience for the customer, all of which reflects poorly on brands. Customers view brands as brands. A customer doesn’t distinguish between ASOS on the web and ASOS on a mobile app — just like you don’t distinguish between your best friend on the phone, in person or via email.
To provide customer experiences that will keep customers coming back, brands need to shift away from a channel-based approach and into an audience-based approach. It’s about putting yourself in your customer’s’ shoes. Customers don’t experience devices. They don’t experience channels. They experience a brand holistically, so you need to think holistically about the customer journey your brands should provide and how to optimise it.
To optimise the customer journey, brands must focus on these three areas: technology, data granularity and revised marketing strategy.
Mastering experience delivery via cognitive technology advancements.
As analytic technology has advanced over the years, companies have gained the ability to automate more and more of the work required for creating effective customer experiences. Initially, analytic technologies were descriptive. This involves dashboards and report generation — reporting what has already happened. The next advancement brought diagnostic analytics. This applies statistics to better explain and organise data. Diagnostic analytics provide a better understanding of what’s really happening.
The third advancement was predictive analytics and it represents a pivot point from reporting the past to thinking about the future. This is how companies use historical data to predict events, like the likelihood of a sale. Predictive analytics can happen in real time by computer or off-line by analyst.
The next stage is prescriptive analytics. Once you can effectively predict outcomes, you can begin to influence or nudge, those outcomes by taking outside actions to influence customer decisions, like when ASOS recommends the right size while a customer is looking at jeans. This is where technology blossoms — there’s an actual recommendation taking place automatically and in real time.
The latest evolution of analytics is cognitive, which combines the earlier advancements of predictive and prescriptive analytics. By using machine learning and artificial intelligence to perform these functions, companies can process far more data in far less time. This makes way for near-instant, scalable decision making, which allows companies like ASOS to make recommendations to any and all of their customers whenever and wherever they engage.
Cognitive analytics is like a matchmaking service between a company and its customers. Via machine learning and real-time interventions, it pairs customers with the experience most likely to drive the desired outcome. These matches are the way companies nudge customers toward purchase decisions, in real time, by providing excellent experiences.
Getting personal via granular datasets and propensity profiles.
With machine learning and AI, companies can now build very accurate customer profiles that make relevant matchmaking possible. This opens the way for individualisation — but not in the way some people think. Customising experiences at an individual level would actually be cost prohibitive. Instead, companies use demographic and psychographic data to enhance their understanding of who the high-value customer is, bringing granularity to the data.
Using that information, companies can build customer profiles that represent large groups of similar customers. We often talk about one-to-one personalisation, but brands don’t generally create individual ads and offers for each unique person — though portions of an ad may be dynamically individualised. Instead, they create ads for a handful of key audiences and match customers to those audiences.
With an understanding of these key audiences, marketers can target offers and campaigns properly. Then, the next step of personalisation is for the technology to track an individual’s preferences and connect them back to one of these groups or profile clusters, in order to make the appropriate offers.
Traditionally, analysts at ASOS pulled data from an analytics tool and then reported that data to key stakeholders to drive their decisions. But, the tool was cumbersome in that it didn’t allow for dynamic or reactive reporting. For ASOS, with monthly site visits in the tens of millions across myriad devices, the inability to create a 360-degree, real-time customer profile at scale made it impossible to surface actionable insights. With tools like Adobe Analytics, however, they could segment their audiences in this way and then build actionable profiles that stakeholders could use to craft and deliver compelling customer experiences.
From there, the right analytics tools allow companies like ASOS to use data to build propensity profiles for customer interactions. A propensity profile is a dataset that indicates the likelihood of a specific future event. It’s like predicting the future, based on what similar customers have done in similar situations in the past. Though it’s not a perfect prediction, it gets marketers thinking in the right direction.
Understanding what customers are most likely to do next based on what they’ve done in the past helps marketers know when to intervene in an effort to move customers further down the path to a sale. This is how marketers know what experience to use in order to nudge customers forward on the customer journey.
Turning high propensity customers into lifetime customers.
Finally, using customer and propensity profiles, companies need to re-design their marketing plans from the ground up with retention in mind. Marketers should plan in profile-specific ways, building campaigns and offers for each major customer type, rather than the traditional method of channel-specific approaches. Simply put, companies need to break from the status quo if they want to succeed.
And, in doing so, they need to think long-term. Here’s why. Usually, brands tend to overemphasise the immediate sale at the expense of lifetime value. It’s a balancing act. Don’t just push a low-value purchase. Instead, nudge them toward additional value.
Taking this warning into consideration with the propensity profile, marketers can map out the customer journey and what interactions to take in order to lead customers to high-value interactions. This connection between propensity and prescription helps marketers know when to take action — and when to leave people alone.
The goal of the brand is to nudge those who are likely to create value, but need some help. Using analytics, identify the people who are already going to do what you want them to do and leave them alone. They’re already on their way. Likewise, people who don’t match your profiles are unlikely to complete the customer journey and so should be left alone. Instead, use your marketing budget to influence those individuals who are sitting on the fence. Ultimately, those are the people you want to nudge.
By harnessing analytics and personalisation, marketers can maximise the impact of their ad spend. Simply providing the perfect clothing size while a customer is shopping on site allows ASOS to nudge the customer toward a purchase. This kind of targeted effort brings excellent customer experiences to the customers who will actually care, increasing the likelihood of a purchase.
Brands of the future.
ASOS nudges its customers by offering a free return policy and then doing everything it can to make sure customers never need to return anything. By learning from its customer segments, the company can better target the potential customers who need the push. That means more conversions and a better return on advertising spending.
The successful brands of the future will need to re-evaluate their experience strategy today. Instead of diving in with touchpoints and strategies around channel- or device-specific experiences, get to know your customers first. Then, with this personal understanding, you can drive your programme decisions, marketing strategies, data planning and the associated technology that’s needed to support the right experiences. That is the key to future success as an experience business.