Customer experience: Get the basics right and the sky is the limit
To measure up to customer expectations, businesses should start with the data they already have and work towards building a solid base before investing in the latest technology.
A great customer experience (CX) is crucial to retaining customers, but in an age when brands are fighting harder than ever before for savvy consumers, the pursuit of designing an outstanding CX can send a business down a rabbit hole of the latest advancements in artificial intelligence (AI) and machine learning before it is ready.
Make no mistake, technology can drastically transform the experience customers have with a business and drive valuable operational efficiencies; however, investing in technology for technology’s sake will not do much for an authentic CX if simple data challenges are not addressed first. Businesses need to get the basics right, build a solid base and then methodically add layers with a clearly-defined CX strategy.
Try this: think about a good experience with a company that kept you accurately updated along every step of your journey. An expectation was created, and you were able to measure how the business fared in meeting this expectation. Now, compare that to an experience where you had no idea what was going on. Try and remember the frustration you felt when trying to get answers from this company.
No business wants to end up like the second example. However, unless some basic challenges are addressed, that risk is real.
The first challenge is that either companies don’t have data (though most companies do have some collected information), or they are unable to make sense of the data to use it properly. In practice, many companies don’t have a process to efficiently associate a customer with a profile or context.
Ultimately, a business with a great CX can become proactive and not reactive.
The second challenge arises when a business does have access to data, but it is primarily focused on information about the customer’s relationship with the business. In other words, there is an association with product but not preferences, and by extension, context. The latter is needed when predictive models are used.
For example, if a customer had a great outcome with a particular sales agent, that information needs to be stored to build a picture of their preferences and context. This could enable AI models to determine the best steps to follow to match the same positive outcome in the customer’s next interaction with the business, or, what not to do to repeat a negative outcome.
Putting it into practice
Imagine for a moment that a business’s leadership team were to sit around a table and commission someone to overhaul their CX. There are a few steps to follow which would ensure they lay a solid foundation from which to build out use cases and benefit from the best that technology has to offer.
Start from the bottom up, and not from the top down. One needs to learn to swim before buying the latest scuba gear – imagine jumping overboard having forgotten the basics?
It starts with the data you have on the channels that you have them. There is already a lot to work with there. Then, make sure you can identify customers and associate them – not only with product but also preferences.
Very few businesses have no data: if a customer contacted the company by phone, it needs to associate their telephone number with their account; if it is via e-mail, then their e-mail address needs to be associated with their account. These are just two channels – when a customer walks into a physical store and has a store card, that card identifies them. The point is – there is data in some form, so the first step is using it efficiently to identify customers accurately.
The business can then embark on a structured plan to expand its data pool to include associations, preferences and a behaviour history. This is when predictive models can be applied to great effect. For example: if Agent A is assigned to Customer X, then there is a higher probability of a sale than if Agent B were assigned to the same customer, but it may be different entirely for another customer with another set of preferences and behaviour history. The efficiency and efficacy of a system like this is self-evident.
A carefully structured CX strategy, then, is not about using a chatbot for a chatbot’s sake or investing in AI willy-nilly. It is about understanding the data you have, using it, and then putting plans in place to improve this data so that you can build out relevant use cases and use predictive models and other technology interventions to personalise a customer’s experience. Detailed analytics drastically improves business intelligence and can change the game in how a business designs its processes.
Ultimately, a business with a great CX can become proactive and not reactive. Let’s go back to the two examples you imagined. Let’s change it slightly, and imagine the first business made a mistake, but before you called in anger or took your business elsewhere for good, its systems proactively picked up on your unresolved ticket, your favourite agent made contact and apologised for dropping the ball and then went the extra mile in ensuring a positive outcome.
At its core, a great CX is about measuring up to a customer’s expectations. However, to get there, a business must first get the basics right and that starts with the data they already have.