Today’s interview is with Eric J. Hansen, founder and Chief Technology Officer of SiteSpect, a leading software platform that offers optimization, testing, targeting, and personalization solutions that allows businesses to create superior customer experiences. Eric joins me today to talk about the potential of artificial intelligence (AI) in customer experience and personalisation, what brands should be thinking about and what they should be doing to prevent this.
This interview follows on from my recent interview – The meaning of personalised customer experience – Interview with Jamf, Paycor and Qumulo – and is number 235 in the series of interviews with authors and business leaders that are doing great things, providing valuable insights, helping businesses innovate and delivering great service and experience to both their customers and their employees.
Highlights from my conversation with Eric:
- SiteSpect have built a tool that helps enterprises do large scale experimentation, personalisation and optimisation.
- Eric is excited about the potential of AI to drive improvements in customer experience, particularly in the way it can help facilitate customer choice and remove areas of friction.
- Eric says: I’m a customer too and when I do things online it’s not so much about saving time but making sure I have made the best choice possible.
- Consumers have this fear of making the wrong choice.
- Personalisation in the ecommerce space is largely about utilising AI to help customers make better choices, reduce risk and remove friction.
- However, many brands are taking a ‘throw the kitchen sink at it’ approach, are offering too many options to their customers and, in the end, are only confusing them.
- Whatever happened to less is more.
- This is reminiscent of the Paradox of Choice situation described in the 2006 HBR article: More Isn’t Always Better.
- You don’t want to design a customer experience that is akin to getting in the cockpit of an F16.
- In the realm of AI and personalisation, the cross-sell recommendations that are common across many ecommerce sites come about from inferences that are made from data about your purchase history and/or previous behaviour/choices.
- However, the inferences made and subsequent recommendations are only as good as the underlying product and attribute data, which in many product cases is still not that granular, detailed or consistent across product categories.
- An additional problem is that an erroneous or even just curious click can be construed as being an important bit of data and can end up producing recommendations that don’t fit with the customers actual interests or preferences.
- Therefore, to do personalisation well still requires, at times, manual over-ride or human intervention.
- Quality Assurance (QA) is an important wrapper that needs to be added to any personalisation effort.
- To personalise or not to personalise?
- If I log in and give you access to my historical data and, in the effort to personalise my experience, you recommend something or change my experience in a way that is supposed to be better and you miss and it’s actually worse or wrong then it’s going to stand out much more starkly than anything in a non-personalised experience.
- Brands when they are considering personalising the experience that customers receive should be being transparent and open about what they are trying to do.
- Experimentation can help improve and prove the efficacy of personalisation strategies.
- Voice of the customer feedback should also feed into the process and help refine the personalisation process.
- Brands should have systems in place to uncover aberrations in experience based on tight limits/expectations of behaviour.
- Eric provides Staples as an example of a brand that is making great strides towards providing a more personalised experience for their customers. This is driven by a set of tailored recommendations that they have been able to develop through following SiteSpect’s experimentation approach. As a result they improved their customer engagement by 114% and, at the same time, have boosted their conversion rate by 15%.
- To any leaders or executives that are interested in getting started on or improving their personalisation efforts then Eric recommends going out and learning and finding out more about what others have done. In the first instance, they should look up the large amount of material that has been written about open source personalisation projects. This is the first they should do before buying any new software.
- The educated buyer or business person is going to be the most effective.
- There is no one size fits all and experimentation can help you figure out the best way forward in defining and delivering a personalised customer experience strategy that works.
- Experimentation is a mindset.
- Wow service/experience for Eric, whether it is is reactive or proactive, should always be personal andpersonalised.
- Check out SiteSpect.com.
About Eric (taken from his company bio)
Eric is the founder and Chief Technology Officer of SiteSpect, and resides on the board of directors. He leads overall product and technology strategy for the company, as well as entry into emerging markets.
Prior to SiteSpect, Eric was the founder and CEO of Worldmachine Technologies, an Internet development and consulting firm specializing in large-scale web engineering projects for organizations such as John Hancock Insurance, Putnam Investments, Hearst New Media, and The New England Journal of Medicine. Prior to Worldmachine, Eric held product management and software engineering positions at several Boston-based technology firms, including Princeton Transportation Consulting Group (Logistics.com), Raytheon Company, and the Center for Clinical Computing at Harvard Medical School. Eric is a frequent speaker at conferences covering web analytics and optimization, and he writes regularly on topics dealing with the intersection of marketing and technology.
He received a degree in Cognitive Science and Psychology with honors from the University of Rochester in Rochester, NY.