fbpx

loading...

RiTribes

RiTribes is home to a variety of people, perspectives, ideas, and information.

Email: support@ritribes.com

How Data Analysis Can Refine the Customer Experience

  • By Sergei Kosiachenko
  • November 21, 2019
  • Comment

How Data Analysis Can Refine the Customer Experience.

There are three big brands that get a disproportionate share of many people’s disposable income, not just because of the products and services they provide, but also because of the way they go about talking to customers. Amazon, Nespresso, and Netflix do one thing right above all else – they create an experience that can leave people feeling satisfied and happy. And an important part of that experience is created by what they say and when they say it because they say the right things at the right time.

Good and Bad Examples

Nespresso knows when to remind you to buy more capsules and how to remind you so that it doesn’t become a nag. Netflix gets its recommendations right, and as a result, scores of people have discovered dozens of new films and series they otherwise would never have seen. Amazon does a good job with its promotions, suggesting things people might actually want to buy. It feels as if they’re all listening, taking notice of what people like and trying to give them what they want.

Contrast that with the experience I had when I recently bought a fridge, from an online-only appliances retailer. The buying experience was perfect, the product was spot on and I was very pleased with it all.

Until I started getting emails from them promoting fridges. Endless emails trying to sell me a fridge. Having just bought a fridge, I didn’t need another one and I was somewhat surprised that they hadn’t worked this out. I haven’t bought anything else from them.

So, what did this appliance retailer do wrong that the three other brands don’t? They messed up their use of data. Someone there knew I’d bought a fridge but hadn’t told the marketing people, or the marketing people hadn’t bothered to look or, most likely, the data wasn’t available.

Humans Need to Interpret AI-Collected Data

Using data effectively is an art and a science and, even if you gather and maintain it correctly, it’s how you interpret it that matters. Faced with an increase in sales of kitchen utensils every September, a retailer needs to be able to figure out that it’s because students are buying their homewares before leaving for university, and not because people replace their spatulas every autumn.

The trouble is, data is not taken seriously. We’re always going on about how vital it is, but how many companies employ proper data scientists to deal with it? Too often that job is left to marketers and although we’re pretty good at many things, we’re not data scientists. Relying on artificial intelligence to do the job for you has its limitations; while it can spot patterns, it might not be able to work out what they mean (see the spatula example above.)

Invest in Data Scientists

And the fabled single view of a customer can be very hard to attain. The car industry works almost exclusively on the franchised dealer model, and that means the car manufacturers have little direct contact with the customer. The dealer owns the customer relationship, which leaves the customer isolated from the brand. This isolation means it’s easy for customers to assume that the brand they’ve loyally driven for years doesn’t care about them.

There’s no single solution to improving this situation. But investing in data scientists is a good start, especially if it stops marketers from having to do the job. Leaders need to do a better job of integrating the left and right brain parts of a company, getting operational data to mix with experience data. Understanding customers and giving them a positive experience is fundamental to remaining competitive; data provides the means to do so, and we need to start using it properly.

How do you ensure accurate data interpretation and seamless collaboration between your marketing and data science teams? Let us know in the comment section below!

Link to the source