olly - Fotolia
Customer analytics applications add artificial intelligence to reduce market ‘friction’
Customer analytics applications developers are seeking to add layers of intelligence to improve sellers’ interactions with buyers
Customer analytics tools have always been at the forefront of adoption in the business intelligence space – and given the importance of customers in revenue generation, it is unlikely this situation will change any time soon.
Continue Reading This Article
Enjoy this article as well as all of our content, including E-Guides, news, tips and more.
While pure-play tech firms such as Google and Amazon are ahead of the pack in how they are using such technology to enhance the customer experience, large retailers, banks and telecoms organisations are not far behind. Most of them have been using customer analytics products ranging from customer segmentation to social media listening tools in one shape or form for years, particularly in marketing departments and customer service functions.
But with the growing uptake of artificial intelligence (AI)-based technology, ranging from product and pricing recommendation engines to natural language processing in search and virtual assistants such as chatbots in a self-service context, the market is now starting to move on – in the corporate sector at least.
“Customer analytics technology is widespread where it relates to call centre agents’ desktops or personalising the web experience,” says Sheryl Kingstone, research director for customer experience and commerce at 451 Research. “But where it’s not so common is in making customer service applications more intelligent and contextual, to optimise customer interactions based on profile or location and reduce friction points.”
Although suppliers such as SAP and Oracle are starting to upgrade their systems to this end, other players such as DataSift and Janrain are also coming out with broad-based customer intelligence platforms. These platforms collate and analyse customers’ personal information, as well as transaction-based and other more subjective data such as findings from satisfaction surveys. Background about competitors, wider industry conditions and general trends are then added to the mix to put the information into a broader context.
Revolution in customer experience
Ray Eitel-Porter, managing director, AI, at Accenture Digital, says the customer experience will be revolutionised in coming years, and it will be a win-win for customers and sellers alike.
“Consumers will be offered the right product at the right time and at a price they can afford, which is what sellers want too as it makes people more likely to buy and reduces inventory and, therefore, waste.”
Certainly within the next five years, Eitel-Porter expects very customised offers to become the norm and for many elements of traditional customer service to be handled by AI systems based on big data. They will analyse past and current consumer behaviour, as well as a range of patterns and trends, to optimise customer interactions.
As a result, activities such as online pricing promotions will be much more targeted, while chatbots will handle most standard customer queries.
Artificial intelligence software will likewise be used to reduce customer friction points, says Kingstone.
“It’s about contextual relevance – getting people the right information at the right time or pointing them to the right person more quickly,” she says. “There’ll simply be more intelligence in the system, which will optimise the customer experience.”
For example, when consumers visit a retail website in future, the entire site will be assembled for them in real time, and each unique visitor will see something quite different to their friend or neighbour, as it will have been optimised specifically for each individual.
On the other hand, if they go into a physical shop’s changing room to try on clothes, the augmented reality mirror will be able to provide them with other options and/or personalised offers. Customers will be informed if an item is in stock and be given the option to order online if not.
Ultimately, the aim is to create a so-called “psychic pizza” experience, a term first coined by customer service researcher and entrepreneur John Goodman.
Ron Tolido, chief technology officer at Capgemini’s insights and data practice, explains this term: “Imagine it’s 9pm on a Friday evening and you’re getting hungry, and suddenly the bells rings and your favourite pizza is delivered before you’ve even ordered it. It’s about being so predictive and understanding customer behaviour to such an extent that you can anticipate their needs and provide them with a completely seamless experience in real time.”
But Tolido acknowledges that there is a “careful balance” to be struck by sellers between “being trustworthy” and moving into the “creepy zone”.
“This approach enables companies to be more proactive, but they also have to get privacy and security right or you can forget the rest,” says his colleague Kris le Sage de Fontenay, head of financial services at Capgemini Consulting UK.
The idea is that to add a great deal of convenience to the customer experience, organisations need to know more about the consumers they are targeting. This means getting hold of and working with their personal data using analytics tools.
But for customers to be prepared to hand it over and allow this information to be manipulated, they have to understand the benefits to themselves and evaluate whether they are significant enough to warrant the exchange.
“We’re bound to see a few disturbances along the way, with some getting it right and others getting it disastrously wrong, but the secret to success will be in establishing a balance between the two worlds,” concludes Tolido.
Case study: Ve offers hyper-personalisation
“Our vision for the world in five years’ time is that no two people will have the same customer experience because websites will respond to who each individual is and what they’re doing,” says David Marrinan-Hayes, managing director of Ve, a digital marketing services firm.
In fact, this kind of approach will become such a “fact of life that if it’s not available, people will feel the customer experience was a bad one”, he believes.
The company, which handles online advertising and marketing campaigns for clients such as Virgin Media and Nissan, already uses its own customer acquisition, conversion and re-engagement technology to understand consumers and their intentions.
But as of June 2018, it intends to release personalisation software that will take things to a whole new level in terms of analytics and real-time optimisation, says Marrinan-Hayes. A key aim here is to cut site abandonment levels, which across the industry currently stand at about 60%, or even as much as 98% for new customers, due to issues with price and relevance.
“Big data and machine learning technology are going to fundamentally change the way the web works,” he says. “There’ll be enough information – and not just dumb data like someone has been to the site before, but intelligent data such as how much they spent and in what areas – and it will enable systems to tailor customer offers in real time.”
But rather than simply go for mono-brand onsite personalisation based only on a single client’s data, Ve also plans to add to the mix anonymised customer information from other clients worldwide in a bid to provide a more rounded view of individual consumers.
“We’ll aggregate anonymous data,” says Marrinan-Hayes. “In the old world, it takes a long time, but with AI, after two visits you can infer things with a few data points and start building up a picture, not of who people are but what their behaviour is.”
The goal, he says, is to make the processing of billions of data points so speedy and, therefore, invisible that the personalisation process “feels like magic”. This means consumers will receive a display of relevant products that are “constantly self-optimising” and price points that vary based on their response to discounts.
“Digital advertising drives 68% of all visits to e-commerce websites, but 80% of them currently bounce, which means the majority of advertising spend isn’t hitting home,” says Marrinan-Hayes. “But as AI becomes more sophisticated over the next five to 10 years, we’d hope that figure would drop to sub-20%.”
In the first instance, the goal will be to see bounce rates fall by a couple of percentage points each year, but “for big retail, that can still mean millions of pounds in extra revenues”, he adds.
Case study: Meon Valley Travel Group aims to be a step ahead
AI systems will play a big role in the travel industry in future as the technology will be able to react to customer needs much more quickly than is the case today, believes Colin Boddy, corporate director of travel management company, Meon Valley Travel Group.
“I can see a time when managers could keep abreast of their travellers and how much they’re spending simply by asking technology like Alexa or Siri, and the information would be fed back to them instantly,” he says. “They could also ask for a quote for a travel plan and we’d be able to draw them up a policy at once.”
Such systems would also automatically alert customers to any problems such as flight delays and arrange for coffee to be ready for them when they arrive at their hotel to provide a positive experience in the most holistic sense.
“It’s going to be a natural progression from where we are today, but it will be a big part of the industry going forward,” says Boddy. “Down the line, it’s about being able to deliver value to customers much more instantly.”
At the moment, the average industry response time for customer email queries, for example, is about four hours. But Meon Valley has already managed to reduce this to one-and-a-half hours by automating its offline booking system.
Just over a year ago, it replaced its existing Outlook email offering with Freshworks’ Freshdesk customer service software, which includes customer analytics functionality. The technology centralises customer communications in one place and enables the company to do everything from understand agent response times to recognise peak booking times and track conversion rates.
This information, which includes the results of customer satisfaction surveys, can then be used to generate performance reports.
“In a world that’s becoming more instant, having information at their fingertips helps staff to become more productive and gives us a good insight into what’s happening with our customers so we can enhance their experience of working with us,” Boddy concludes.
Read more about customer analytics
- What is customer data analytics?
- Find out how a cloud managed services firm speeds customer analytics with Pure Storage arrays.