As industries go, the world of eCommerce is still a newborn baby. From its origins in static ads and interstitials to the birth of increasingly intelligent algorithms, the technology being used is evolving quickly and without respite.
To reflect this in our latest series of blogs we’ve been speaking to the best and brightest in eCommerce and retail to find out what the future will hold.
First up is Caroline Baldwin, Deputy Editor of Essential Retail.
A lot has been written about the importance of investing in personalisation tech. But have retailers started reaping the rewards of this endeavour?
Definitely. And we’re starting to see it cropping up more and more. One interesting example is Shop Direct which just reported a 43% increase in profit. They put this down to technology, and specifically personalisation.
I spoke to the CEO last week and he said the next step for them is machine-learning. This will be one of the biggest shifts for retailers, and a transition to what’s called conversational commerce.
So what’s changed? What’s driving this shift?
Actually Facebook really brought it onto the agenda. People have been talking about machine learning for decades. Then IBM’s Watson and tech like that turned these concepts into a reality, but only for a number of niche use cases. It has taken something as consumer-facing as Facebook to make the rest of the industry sit up and start listening.
Shop Direct are actually working with IBM on various projects to see what they can do, such as artificial intelligence automation in its Very app. There’s also a machine-learning concierge robot in an American hotel chain that can answer standard questions such as “what restaurants are open nearby?”
This is being driven from the demand side, customers are expecting more from retailers. Everyone wants a personal connection. I’m guilty of this and after most interactions I head to Twitter to rant or praise them.
What do you think about the impact of tech on the customer experience over the past few years?
Retargeting hasn’t been done very well, I don’t think. I get annoyed when I’m off to see a play and then keep getting retargeted for tickets of the show I’ve just seen. Whenever I go on a certain shoe retailer’s site, I almost dread it because I know I’m going to get retargeted all over the internet – no I can’t afford those boots right now!
I wouldn’t say for many it has been damaging, but the ones doing it really well have seen the benefits. Eurostar is trying to get it right, by understanding that customer behaviour changes and shifts. Getting context is the tricky bit.
Interesting, so it’s about combining social information with other sources of data?
That would be the dream. I thought this the other day, take weekly food deliveries – what would happen if you let them connect to your Google Calendar so they could get more information on you, if they knew when your payday was, they could push more products around that specific date. Thinking about wider retail, that personal approach would be much better than everyone receiving a blast of payday emails on the 30th.
Do you think machine learning will lead to more democratisation in retail?
Definitely, 100%. The main challenge at the moment is that, even Facebook, have to have hundreds of human handlers to intercept, educate and correct chatbots. This required an enormous investment of resources. Just like the real time chat popups we’ve had on websites for years.
There are a few concierge services that have caught on too. One such example is BarChick, a website that provides information on bars and restaurants. They launched a service where you text a request like “I’m off to a concert and need to find a nice bar near the O2”.
At the moment a human gets back to you but over time it’s about using machine learning to educate bots so they can pick up the bulk of the work.
So machine learning is about enabling scalability?
Exactly. Shop Direct told me they don’t want to just use it for customer service, they want to use it for personal shopping as well. It’s not just about answering the logistics of an order, it’s about asking things like: “What might look good with *this*?”
So what’s the balance of investment between the visible and behind the scenes tweaks?
Shop Direct mentioned this in terms of A/B testing. So if you show customers X% of layers on a website and make small adjustments, how much can you increase conversion? I’m presuming machine-learning could cut out a lot of the human element of this.
What about the relationship between acquisition and optimisation?
There’s a little bit of worry that robots are going to take over the world. I don’t think this is the case but it’s undeniable they are going to get smarter. Their potential is only just being tapped into and the more we use them, the more they learn and the more valuable they become. Take Amazon’s Alexa, which keeps getting smarter with third-party involvement. It seems we’re at the tip of the iceberg for optimisation with the first evidence of companies going in this direction.