25 July 2018
Written by: Gustavo E Perez-Lopez, Head of AI Practice, Geomant.
A true story
A friend of mine was about to celebrate his fifth wedding anniversary and was looking for an appropriate gift for his wife. The date was approaching, and he had not yet made up his mind about what to buy her.
He was on his daily train trip back home from work and was chatting with me on his smartphone using Facebook Messenger. I suggested that he buy a special anniversary ring, which several jewellers offer for such occasions, so he did a quick search on Messenger and found a few retailers with a Facebook business account. He felt lucky, so he picked the first on the list who claimed, “Typically answers within 24 hours”, and asked about special anniversary rings. He got an almost immediate answer, although not exactly what he was expecting:
He checked his watch and it was 5: 17 pm. He then picks the next few retailers on the list and got the same response. Bad luck, he thought as the train was arriving at his destination and he had to leave the coach without having yet found a gift for his wife.
Next day my friend visited a local jewellery shop and found a fantastic 5th anniversary ring which his wife loved. I don’t remember if he ever got a response from any of the retailers on messenger, but even if he did, it did not matter anymore, the moment was over.
What are the key take-aways from this story?
My friend’s experience of online shopping is not a unique story. It’s worth summarising the learnings in a few points:
- People are increasingly using mobile devices and messaging apps to do online shopping and find information about products from retailers and other online businesses that help in making purchasing decisions. This is especially true for Millennials and Gen-Z consumers. And it can happen anytime, anywhere, i.e. after work, on the train journey home.
- Modern consumers want to connect with businesses via the same channels they use to chat with peers and friends. In the above story, my friend was chatting on Facebook Messenger and he used the same app to find and communicate with online jewellers that also had a Messenger account.
- Answering an email within 24 hours is probably an acceptable response time, but when using digital messaging channels, there is an expectation of an almost instant response. Several studies have shown that digital enquiries answered within 5 minutes would produce the highest engagement rates; on the contrary, replying after 30 minutes would almost certainly mean that you have lost that customer, as by then most people would have gone elsewhere with their enquiry. This is exactly what happened in my friend’s story.
How technology could have helped my friend buy his wife a gift and the retailer gain a new and happy customer
Keeping a high number of staff and extending opening hours to respond to customer enquiries via online digital channels anytime and anywhere may not be affordable or cost-effective for most business. But AI, or intelligent technology, is here to help both consumers and businesses:
- Artificial Intelligence (AI) and in particular language understanding has evolved a long way in the recent years. In a stunning, though controversial demo1 last May, Google showcased a voice assistant using Google Duplex technology that can book haircuts and tables in a restaurant without the other party ever realizing they were talking to a robot.
- Azure Machine Learning, Microsoft Bot Framework and Microsoft Cognitive Services, key components of Microsoft AI2 platform, are making really easy and affordable for customers to implement the latest advancements of AI in almost any industry and business process.
- Chatbots – an intelligent software that can engage in human-like conversation with people across multiple messaging platforms – are a key tool to help your business get started with AI and take the next steps in the digital transformation journey.
Going back to my friend’s story, an AI-powered Facebook Messenger chatbot, using Microsoft LUIS natural language understanding technology would be able to process typical customer enquiries such as “I am looking for a wedding anniversary ring.”
It could then go and do an automated search in the retailer’s product catalogue and inventory system and present my friend with a choice of rings, including an image, a brief description, options, price, stock availability and even a “Buy now” button, so that the whole purchase journey could be easily completed without having to leave the Messenger app.
Furthermore, by using advanced machine learning and predictive algorithms, the chatbot could also make personalised suggestions on companion items (e.g. an anniversary set including also earrings).
Finally, once the chatbot had converted my friend from a casual consumer into a delighted customer, it would offer to send him automated notifications in advance of the next wedding anniversary to make sure my friend does not have to worry about what to buy his wife next time.
How retailers are using AI and chatbots today
One of the key messages from the last World Retail Congress was that “retailing today may be tricky, but it also has huge opportunities to create new revenue streams worth billions of dollars”, and one way to seize this opportunity is through AI 3
Many retailers globally have already embraced AI and are leading the way in the application of intelligent technologies, including chatbots for the benefit of their business and customers. A research paper published in March this year 4 revealed that sales at UK’s biggest online retailers have increased by 23% in a year, and one of the key contributing factors were AI-powered chatbots on digital e-commerce platforms, which “engages shoppers in conversation to advise and suggest purchases”.
To give you a closer look at how retailers are using AI technology and in particular intelligent chatbots to improve the customer experience and boost their sales, I have selected a few examples below of chatbots that you can try yourselves and see in action. This is an entirely subjective selection and represents a very limited scope of what you can do with AI.
Cami, Currys PC World, Electronics
Currys PC World is a brand of Dixons Carphone, a major European electronics and telecom retailer and service provider, based in the UK.
Cami is Currys PC World conversational chatbot 5 that helps online shoppers to:
- Find the right product
- Check if a product is in stock in your nearest store
- Save products you are interested in for later (wish list)
Cami understands natural language searching so that you can include things like your budget or your favourite colour, i.e. "I'm looking for a red toaster" or "I want a Samsung TV between £350 and £450". It can also scan pictures of product labels taken by the user. You can chat with Cami on Facebook Messenger or via Curry’s chat portal.
Margot Winebot, Lidl UK
Lidl6 is a German global discount supermarket chain that operates over 10,000 stores across Europe and the United States.
Margot 7 is a conversational Facebook Messenger chatbot that understands natural language and helps shoppers at Lidl UK get the best out of their wine range. Margot helps users in various ways, e.g.:
- Helps to find wines by country, region, grape, colour and/or price
- Gives tips on food pairing
- Tests your wine knowledge with a quiz
Margot is accessible from Lidl UK Facebook page either via a browser or via the Facebook app.
H&M
H&M8 is a Swedish multinational clothing-retail company known for its fast-fashion clothing for men, women, teenagers and children. It has developed a chatbot9 on the Kik Messenger platform that helps shoppers building an outfit based on personal style preferences. Using the H&M chatbot users can:
- Search, shop and share outfits
- Create outfits based your style profile
- Vote and browse outfits created by other users
The AI Retail Playbook
Chatbots or Virtual Assistants are a great example of using AI to improve business outcomes for retailers, but they are just one of the endless applications of AI in the Retail Industry. Going in-depth into each of the potential use cases of AI in retail is way outside the objectives of this post, but Microsoft has recently published an AI playbook for the retail and consumer goods industry10 that goes beyond the chatbot scenarios and envisions a complete retail experience infused with AI.
The playbook includes four detailed scenarios of AI-integrated retail, which cover using virtual assistants to understand and adapt to customer needs, building personalised service and shopping experience based on comprehensive shopper profiles, optimising in-store operations, and data-driven marketing.
I cannot agree more with the following statement from the playbook:
If you are interested, drop me a line and I will send you a copy of it.
Conclusion
I firmly believe that online retail is the sector that can probably take the most advantage from the AI revolution that we are witnessing, and chatbots are their starting point. Many retailers have recognised this, have already started the AI journey and began enjoying its benefits. Are you ready for your Quest of the Holy Grail?
About Geomant AI Practice
Over almost two decades, Geomant has been helping companies to enable better customer service and effortless customer interactions. Building on our strong software development, integration and innovation heritage in contact centres and unified communications solutions, Geomant AI Practice is now assisting our partners and customers to step into and successfully ride along the new journey of digital engagement and artificial intelligence (AI). Visit our website to find out more.
If you want to learn more about Why AI and chatbots should be a priority for online businesses' take a look at our informative webinar, available online now.
Related blogs
If you find this reading of interest, you may want to have a look at some related blogs:
- How to take those critical first steps towards implementing AI in the contact centre
- How much does a chatbot cost and where does the value lie for my business?
- Why chatbots are the future of m-commerce: statistics, benefits, use cases & startups
1https://www.theverge.com/2018/5/8/17332070/google-assistant-makes-phone-call-demo-duplex-io-2018
2https://www.microsoft.com/en-gb/ai/default.aspx
5https://customers.microsoft.com/en-us/story/dixons-carphone-retail-cognitive-services
6https://en.wikipedia.org/wiki/Lidl
7https://www.lidl.co.uk/en/How-to-use-Margot-the-Winebot-11782.htm
8https://en.wikipedia.org/wiki/H%26M
9https://www.topbots.com/project/hm-kik-bot-review/
10https://enterprise.microsoft.com/en-gb/industries/retail-and-consumer-goods/