Scale can mean many different things. In the context of the delivery of an omnichannel customer experience (CX), scale is primarily thought of as personalizing an experience for as many individuals as an organization’s people, processes, and technology will support. Scale encompasses the volume of data a decision is based on, the number of branches in a decision tree, the available variations of messages or offers and how quickly a decision is made.
When real-time is added to the mix, the ability to do all those things becomes exponentially more difficult. Nevertheless, a real-time capability is a vital component for scaling CX. While it is possible to scale without real-time – meaning you engage with every customer, make decisions using all available data and on hundreds of decision tree branches – the problem is that without a real-time capability a personalized experience will – for many customers – fall short of expectations.
The reason is that a real-time component ensures that decisions are made on timely, accurate, and complete data, which is required to keep up with the cadence of a modern, omnichannel customer journey that consists of a combination of physical or digital channels. Consider the results of a 2021 Harris Poll survey, where consumers ranked a consistent experience across all channels as the most important element of CX.
Birds of the Same Feather: Consistency and Real-time
In many cases, consistency depends on real-time. Consider a customer visiting a retail brand’s website. The brand personalizes an offer on the homepage using yesterday’s data. But if it fails to account for every action the consumer may have made in the interim, such as an in-store purchase, an engagement with a call center rep, a social media post, the pages that they’ve just viewed, or a host of other interactions that might change the personalized offer, it runs the risk of alienating the customer with an irrelevant offer.
Real-time analysis of an online session will factor in pages visited, time on page, links clicked, etc., with content generated perhaps also dependent on other sources of data compiled in real-time such as purchase history, previous website visits, or in-store activity. That real-time analysis and real-time decisioning could affect dozens of elements of the website experience – images, text, links, and any combination thereof. And what the customer sees and experiences might itself be culled from thousands of options. A major retailer might have 10,000 or more available products in stock – all of which will be available to drive personalized recommendations.
It’s important to recognize the distinction between real-time and speed. While the speed with which a brand interacts or engages with a customer across various channels is certainly one element of real-time, it is also possible to make the wrong decision quickly. Mitigating this possibility brings us back to the need to make decisions based on all pertinent data. If data with which a decision is made comes from 10 different sources, the data must from all those sources must be ingested, cleansed, matched, and readied for decisioning in real-time. Unless the original data sources themselves are consistently updated, a brand can quickly compile a unified customer record, but it may be making an inaccurate decision quickly if it’s based on stale data.
Delivering a consistent experience for an individual customer requires determining with a high degree of accuracy that a brand is showing the right thing to the right person at the right time, even with an enormous amount of variables. Real-time at scale has a lot of moving parts, in other words, that can grow in complexity depending on the level of personalization a brand wants to reach.
Ease Up on the Brakes: Real-time for Every Channel, Every ID + Key
The operational overhead needed to deliver real-time capabilities at scale can be overwhelming. It’s an enormous change management issue. For starters, businesses must adapt to a world where machine learning algorithms instead of people are making decisions. In the website personalization example, if there’s a universe of say 10,000 products available to recommend in a real-time personalization engine, everything must be tagged in a way to make intelligent decisions about what a piece of content is relevant for (and who it’s relevant for) and what it’s not relevant for.
The website, of course, is only one channel. That same calculation must be made for every channel, with a decision returned within milliseconds on not just a web page but for the mobile app, call center, chatbot, an in-store associate, a hotel clerk or any synchronous engagement channel. A consistent experience depends on real-time, yes, but also real-time in the precise cadence of the customer’s journey across all touchpoints.
A real-time decision, in other words, cannot be made in isolation. One issue is the risk of presenting a fragmented experience, the other is a brand may limit its options for personalization. For example, many solutions may promise a real-time capability if they’re doing real-time based on a website log-in key, or perhaps on a cookie and a web ID. Here, the brand will limit personalization to the observed behavior on the website, limiting both what’s available to personalize as well as how to identify a customer. If a customer visits a website from multiple devices that a brand does not recognize or tie to a unique customer profile, the brand narrows its options.
It’s important to be aware of all of the factors that go into delivering a hyper-personalized CX at scale. Real-time, as we’ve seen, has many different components including speed of decisioning, latency of data driving decisions and availability of a breadth of content to deliver personalized experiences. Scaling real-time is vital to ensure the consistency that customers demand in an omnichannel customer experience. For more on how Redpoint provides customer with relevant interactions informed by real-time data at every touchpoint, click here.