2020 is coming up fast, but did you know it is predicted that by then, 85% of customer service interactions will be powered by AI bots? Did you also know that by 2020, 100 million consumers are expected to shop using AR technology? The use of AI in commerce is the future of customer experience – are you using it to the best of its ability?
Maximize the use of AI in Commerce to Improve Customer Experience
The use of AI in commerce has so much potential in part because AI is a chameleon — it morphs and adapts depending on consumers’ needs. Perhaps AI’s greatest trick though, is how it plays a dual role of being both a money-maker and, a money saver. For example, augmented reality is as close to a living, breathing purchase experience a customer can get. From an operational standpoint, integrating AI into data analysis means infinitely more information can be processed than a human being could handle, freeing up more time for your team to spend engaging and selling to customers.
In this article Elastic Path share the three main areas within AI you should focus your attention right now to maximize its effectiveness, how leading brands are using AI to not only deeply engage their customers but also eliminate major pain points their customers are experiencing and, how you can understand the digital maturity of your team in relation to effectively leveraging all that artificial intelligence has to offer in the future of customer experience.
Content Summary
Introduction
Professional vs. public perceptions
Voice assistants and chatbots: Smarter than the average bear
Current best use: Sephora unleashes a robust beauty tool
Augmented reality: Moving beyond awkward headsets
Current best use: Zara invites “models” into its stores
Predictive preferences: Tell me what you want… or don’t
Current best use: The nine phases of Amazon’s on-site recommendations
Conclusion: Assessing your state of AI-readiness
Introduction
Commerce is moving in an increasingly AI-centric direction and by 2020, artificial intelligence (AI) bots will power 85 percent of customer service interactions. Considering that is just around the corner, it’s in everyone’s best interest to get started with implementing or improving their use of AI in the customer experience.
The partnership between AI and commerce has so much potential in part because AI is a chameleon — it morphs and adapts depending on consumers’ needs. Most organizations see AI as a cost-savings strategy but unfortunately some are already losing revenue due to poor customer experience. In this case, the opportunity for AI is clear. Improving the customer experience through the use of AI will result in less customer turnover, establishing a cost-saving strategy to increase sales and improve profit margins.
In this article, we’ll identify the three key areas commerce companies should focus on right now if they want to ensure a successful future using AI in the customer experience: Voice assistants/chatbots, augmented reality (AR) and predictive technology. We’ll share actionable takeaways based on real-world success stories and the common pitfalls to avoid in your own implementation of AI.
Professional vs. public perceptions
The role of AI in commerce was borne of chaos in a rapidly changing world. Once brands gained the ability to sell products directly to consumers (think: buying your jeans directly from Levi’s), third-party vendors began slashing prices to keep up. While scrambling to stay afloat in this new fiscal world, retailers started losing sight of the customer experience and at the same time, losing consumer loyalty.
This may be where businesses have the most to gain from AI. AI has an outstanding ability to process data with far more speed and accuracy than a human being ever could, giving retailers far better insight into things like pricing structures, inventory management, target prospective customers and more. This data is also rich with actionable insight on how to better serve customers, somewhere AI can also play a huge role.
Chatbots are becoming one of the more recognizable uses of AI in commerce and with that, one of the better use cases to show its power to learn behavior and supplement human intervention. Commerce interactions generate a neverending stream of data and feedback, the perfect fuel for AI technologies. The more data they have at their disposal, the better they perform.
Because of this, AI has a bright and robust future when it comes to public-facing interactions, but how does it truly fare beyond the computer scientists that bring it to life?
A programmer’s interactions with AI don’t need to be particularly personal, after all, people don’t expect their AI tools to ask about their weekend or their lunch plans. But when it comes to public-facing applications, AI has to be able to apply its learning to show some level of relatability and compassion, especially in the event it is handling frustrated customers through chat.
Professional perceptions
If your professional future is in tech, it’s also going to be in AI. Stanford University’s AI100 study is an ongoing look at how AI has shaped our world and will continue to do so, finding the share of jobs requiring AI skills has grown 4.5 times since 2013. Additionally, it found:
- 84 percent of enterprises believe investing in AI will lead to greater competitive advantages.
- 75 percent of enterprises believe AI will open up new business opportunities.
- 63 percent of enterprises believe the pressure to reduce costs will require the use of AI.
Reflected in those statistics are two important points: AI will be both a money-maker and a money-saver. As mentioned above, one of the greatest benefits of integrating AI into data analysis is that it can process more information than a human being could possibly handle. For programmers, this frees up their time to handle other, more interpersonal tasks. For sellers, AI bolsters their capabilities through things like recommendation engines and sort order. From an external-facing standpoint, AI tools have the potential to increase revenue by enabling companies to communicate with more customers in a shorter period of time.
While enterprises are excited about AI’s potential, individuals in certain roles within the company may not be as enthused. It is the organization’s responsibility to help workers see the benefits of AI but, that is a topic we will explore later in this article.
Public perceptions
Consumers’ experience with AI has not always been a positive one. From the beginning, telephonics, chatbots and other AI customers have interacted with has been inconsistent, frustrating and sometimes felt like part of the problem rather than a solution. In short, AI wasn’t human enough.
At this point though, businesses that still hold onto the notion that customers don’t like AI are doing themselves a disservice considering how much AI stands to improve the customer experience.
Companies now have to meet consumers’ expectations on all fronts; one wrong move, and they’re off to a competitor.
Two-thirds of respondents to New Voice Media’s survey (and 80 percent of 25- to 34-year-olds) reported leaving a business due to inadequate customer service. In 2017, that meant businesses lost out on $75 billion because of poor customer service.
It’s time again to put customers first. AI makes that goal both realistic and easy to implement. Here’s where you can get started with implementing voice assistants/chatbots, AR and predictive technology. 4
Voice assistants and chatbots: Smarter than the average bear
Alexa, Siri, Cortana, Watson, Google, the gang’s all here! Most Americans have met, or rather, interacted with at least one of these digital voice assistants. The reason these names have risen to the top of popularity is due to their exceptional predictive analytics and, while they’ll never be mistaken for humans, their quippy replies when asked the right questions could be thought of as endearing.
The interactive nature of this branch of AI presents a great opportunity for commerce brands to show off their voice and personality. Do you want your bot to be able to respond with a bit of humor? Do you want it to convey academic expertise, or sound like a friendly neighbor or colleague? This decision could make or break your engagement with a consumer, so make sure you give it some thought.
Current best use: Sephora unleashes a robust beauty tool
Beauty retailer Sephora knew the massive selection of products they offer had the potential to be a huge opportunity when it came to their use of AI. No two customers would have the same needs or preferences, but a lot could be learned in the process of a chatbot helping them figure out what to buy. Enter: their suite of chatbot-based apps.
Kik chatbot. The bot asks users questions about themselves to learn preferences, then offers up videos, personalized beauty tips, products and more. The Kik bot allows users to select from suggested answers so they don’t waste time typing long conversations. Shoppers can also make in-app purchases, a key feature for customer retention.
Reservation assistant. This bot uses Facebook Messenger to help customers book appointments for consultations and beauty services at a nearby Sephora. The bot’s smart learning system allows it to carry on conversations based on the user’s natural language.
Color Match. This pairing of Facebook Messenger chatbot and AR is perfect for customers who don’t have immediate access to a Sephora store. Color Match users hold their camera up to any image or face, and the app will present them with the identified shade, as well as other matching Sephora products. The face doesn’t have to be the shopper’s own; Color Match can scan advertisements, Instagram photos or any image the shopper wants to recreate.
Actionable takeaways and opportunities
Voice will become a critical part of customer interactions in both B2B and B2C environments. In the next four years, voice assistants are expected to be used in 55 percent of American households.
Know yourself, know your chatbot. Before introducing any voice assistant or chatbot to your customer support systems, make sure you fully understand your company’s persona and voice. The bot isn’t human, but it is acting as a representative of the brand. Give your bots a personality and script that align with your overall tone.
Maintain and retain existing customers. Morgan Stanley Wealth Management is piloting an AI program that helps its financial advisers stand out from the rest. Its bots evaluate client communications, then other potential suggestions for that client, such as stock movements, insurance tips and emergency resources. While this application is specific to insurance and stocks, the same machine-advising-human-advising-client model could be applied in any number of B2B settings.
A global ambassador. Looking to increase B2B sales in Asia or Europe, but don’t have the financial bandwidth for an international office? A chatbot that’s able to answer questions around pricing, shipping dates, production capabilities and more in local languages could be the difference in making a sale and growing your brand worldwide.
Potential pitfalls
The influence voice assistants have over what consumers purchase will continue to grow which means placing Amazon, Apple or Google between a retailer and its customers has to be done strategically.
Consider: Do you want to use an existing chatbot or voice assistant in your commerce stream, or develop your own? Reliance on a big-name company means ready-made customers and a broad reach, but also potential restrictions and drawbacks that go along with a partnership with a company that has the power to make or break your future.
Investing in voice assistants and chatbot technology is only a smart move if it is paired with the training to back it up. AI isn’t a plug-and-play technology — it must be told how to do its job, which means human programmers must be properly trained. Before investing in AI communication technology, make sure you have the human infrastructure to make it worthwhile.
Augmented reality: Moving beyond awkward headsets
In a sense, AR has the potential to bring ecommerce to life. Traditionally, websites and apps could recommend products based on browsing history and past purchases but they’re never more than static product images on a screen. AR stands to finally bring sensory immersion to the shopping experience.
When customers can interact with products in a more meaningful, personalized way, sales improve dramatically. By 2020, 100 million consumers will shop using AR technology. It’s no longer a technology awaiting us in the distant future, it’s here right now.
Current best use: Zara invites “models” into its stores
Seeing an outfit on a human is certainly more engaging than combing through racks of clothing or staring at a faceless mannequin. Unfortunately, retailers can’t exactly hire live models to walk through their stores all day.
Fashion retailer Zara addressed this problem head-on in 2018 by introducing AR into 120 of its stores worldwide. By downloading its AR app, shoppers can point their phones at Zara store windows or displays to watch a model show off a variety of outfits. Everything seen on screen can be purchased immediately through the app or in store. Codes on websites and shipping boxes from online orders also trigger the AR experience, giving Zara a range of digital touchpoints with the customer.
Actionable takeaways and opportunities
Simplifying (but not dumbing-down) the intricate. AR can explain complex value propositions or highly technical products in an engaging way. It helps brands take multiple stakeholders through demos or explain features to cross-functional teams. It cuts through the “let me set up a meeting with the IT team” or “I’ll ask the product team to send the information” into an invitation to “experience the product now.”
Providing real value beyond the big purchase. AR also enhances after-sales service. Coca-Cola Germany uses the technology to help B2B buyers plan where to put their vending machines in store. In the chemical, medical and transportation industries, AR is used to train workers how to assemble and use products to reduce injuries and misuse. This deep expertise helps consumers learn how to use products over time, deepening the relationship.
See it, snap it, buy it. The next wave of AR sales is something Instagram junkies and fashion gurus have long dreamed of: Snappable sales. eBay President and CEO Devin Wenig sees a future where typing into a search bar becomes obsolete. Instead, consumers will be able to take a photo of an item they like and buy it within seconds. The Future of AI in Commerce
Potential pitfalls
AR is as close to a living breathing experience as a consumer can get, which means organizations have to leverage the channel entirely differently than static forms of marketing like direct email etc. It’s not a channel that can be properly sustained through the duplication of content, to realize the medium’s potential you have to push yourself outside of the box and get creative with what your customers are engaging with.
Consider: Is your AR experience taking full advantage of the technology?
On the other side of the scale, organizations need to ensure they’re using AR responsibly. Personal privacy threats, identity theft and even physical threats (including users of VR helmets who are unaware of their surroundings), while rare, are all real potential drawbacks of AR. Be mindful when implementing AR in your own commerce experiences, listen to your customers’ concerns and put policies in place from the start.
Consider: What safety and privacy concerns does your AR plan need to include?
Predictive preferences: Tell me what you want… or don’t
People will often say they want something, when in fact, they don’t know, or they want something else. Future consumer behavior tracking will analyze both declared and undeclared user preferences through clicks, mouse movement, inactivity and time spent per page.
This goes beyond cookie-based advertising. Technology will become better at predicting future behavior because what a customer needed four weeks ago may no longer be important today. For example, a dad who buys a baby carrier on Amazon may continue to receive newborn suggestions several years later, but modern AI will infer that he’s now shopping for a toddler.
Current best use: The nine phases of Amazon’s on-site recommendations
Amazon’s success in the retail marketplace is great news for the retail giant, but it’s also good for any company that sells its products through Amazon, including the small business-focused Amazon Storefronts. Why? Thirty-five percent of Amazon’s total sales are driven by its recommendation bots.
Amazon’s access to massive amounts of customer data not only allows for more accurate predictions about what people will buy, it also delivers those predictions in a variety of locations and formats across the site.
The nine phases of Amazon’s on-site recommendations
Here are nine ways Amazon delivers recommendations to shoppers throughout its site:
- Recommended for You, X: A full page of recommendations based on the categories a shopper has been browsing. Not just products they have already clicked on.
- Frequently Bought Together: You just added a new pair of boots to your cart — why not pick up a three-pack of socks, too?
- Your Recently Viewed Items and Featured Recommendations: Similar to the first category, these recommendations are variations on products you’ve already browsed. Maybe you’d be more likely to buy that watering can in a different color?
- Your Browsing History: Did you see an interesting book while shopping a couple of days ago, but can’t remember the title? Amazon can. Browsing history collects products you considered but didn’t add to your cart.
- Related to Items You’ve Viewed: Essentially the same as the third point, but with a different label.
- Customers Who Bought This Item Also Bought: Similar to the second point above, but again, with a different label. You can see Amazon’s theory here: The same thing worded a different way might resonate.
- There is a Newer Version of This Item: Upgrading an item a customer already owns or upselling an item they’ve added to their cart.
- Recommended for You based on…: This list pops up after checkout. Did you buy a tablet, but not a case? Amazon has plenty of ideas about what you might like.
- Best Selling…: These suggestions aren’t always related to a shopper’s own searches, but rather popular or trendy sellers. Categories like “emerging technology” can send a shopper down a whole new rabbit hole.
Actionable takeaways and opportunities
Helping buyers derive order from chaos. Consumers are becoming ever-more comfortable with AI’s ability to make suggestions, and soon will come to view it in the same vein as a personal shopper. Picture a shopper who’s looking to buy a pair of red shoes. AI will allow commerce platforms and tools to predict that person’s price range, preferred heel height and material — as well as the next things they’ll want to buy.
How much do you want to pay? Dynamic pricing, or pricing based on supply, interest or competitor’s availability, is an aging game. Instead, both B2B and B2C sellers have a real opportunity with personalized pricing. What is the buyer doing in real time? Are they a loyal customer? Are they highly engaged, and have they added a product to their cart without buying? Personalized pricing looks at these factors and more to find a price that’s most attractive to an individual buyer at a specific moment.
Potential pitfalls
It’s good to always remember there’s a fine line between advertising and other targeted content that helps a shopper move toward something that interests them, and content that annoys or creeps them out.
Consider: When it comes to putting predictive AI to use, how much one-sided contact is too much?
Predictive technologies, like all AI, rely on data. The more data, the better the predictions. As such, insufficient data will lead to inaccurate predictions. Even with a mound of data, predictive technologies can’t always consider variables like weather, relationships, moods and health. Hiring the right programmers to interact with AI is a great first step but be aware that predictive technologies are still an evolving arena.
Consider: How heavily do you want to lean on predictive AI for suggestions? What variables can you integrate to ensure better results for your customers?
How are you using AI right now?
This doesn’t necessarily mean customer-facing applications. Some of the same AI processes you’re using internally can be applied to any new technology you’re considering. You may already have internal experts ready and willing to take on a new or expanded role.
Which AI applications will most benefit your business?
Not every use is right for every retailer — for example, plenty of customers wouldn’t benefit from AR just yet. Consider surveying your employees and existing customers to see which forms of AI they most frequently use, and which they most enjoy interacting with. You’ve got a whole host of educated consumers at your disposal; why not use them as a resource?
Do you know how you’ll educate your employees and customers on AI? How will you gather feedback from them?
Start your AI journey with a partner who knows the technology inside and out and make training part of your contract. Just like the value proposition created between you and your customers, Elastic Path offers you complete value beyond the sales decision.
True success comes from the blend of artificial and organic intelligence. We’ll help you get employees and customers on board with these new technologies, calm any fears and pivot based on real-time feedback.
And, as the pioneer of headless commerce, our platform is purpose-built to meet the customer where they are, no matter the digital environment. We engage them using AR and live streaming, talk to them through voice technology and listen carefully to them with AI. Let us show you how Elastic Path can help you understand and deliver truly differentiating AI-driven commerce experiences.
Conclusion: Assessing your state of AI-readiness
By now, we hope you’re excited to see the differences AI can make in moving your sales future forward — and that getting started isn’t quite as daunting as it seems.
But before you jump in for the first time or make big changes to your existing AI instances, you’ll want to assess your internal AI readiness.
Source: elasticpath
