Table of Contents
Introduction Jennifer Sherman Kibo Rob Garf Salesforce Commerce Cloud Brian Rigney Zmags
Omer Artun AgilOne
Peter Sheldon Magento
Juliana Pereira Smartling
About Retail TouchPoints
Bryan Chagoly Bazaarvoice
James Green Magnetic
Jim Davidson TurnTo
Jared Blank Bluecore
Maribeth Ross Monetate
Mihir Kittur Ugam
Allen Nance Emarsys
Oscar Sachs Salesfloor
Pete Olanday Vertex2018 E-COMMERCE TECHNOLOGY PREVIEW
RETAIL TOUCHPOINTS
2018 E-COMMERCE TECHNOLOGY PREVIEW
Retail TouchPoints is proud to introduce the third annual E-Commerce Technology Preview, featuring insights from 15 e-Commerce industry experts. This guide offers an exclusive and unique look at how retailers are gearing up for e-Commerce and omnichannel success in 2018 and beyond. This comprehensive collection of e-Commerce thought leadership will help retailers determine the most effective go-forward business strategies. Key topics include:
Artificial To
We hope you find a significant takeaway from each contributed article that you can share with your team to help make 2018 a most successful year!
Debbie Hauss Editor-In-Chief Retail TouchPoints
Intelligence (AI);
Personalization;
Beat Or Join Amazon; Science; and Strategies.
Data
Mobile-First2018 E-COMMERCE
TECHNOLOGY PREVIEW
OMER ARTUN CEO AGILONE
THE TRIFECTA: RETAILERS, CONSUMERS AND MACHINES
We have entered a time when technology advances have brought us to a place where computers are able to augment human understanding and behavior. While I don't think we have gotten to the point when artificial intelligence will replace human work entirely, there is a lot for retailers, in particular, to gain embracing the power of machine learning.
CLUSTERS & DECISIONS
Machine learning works best for marketing situations where the marketing mechanics, the automated response to customers, require human judgment and decisioning. A perfect example is clustering, where one of the key heuristics is representativeness. Representativeness is when humans naturally group things together to reduce complexity. When machines do this with unsupervised learning algorithms, marketers gain meaningful customer segments without the work it would take to develop those segments manually. Clustering also helps marketers recognize future customers earlier, predict whether someone likes a product or message, and so on. Machine learning manages these variances exactly as a human would: calculating and decisioning without explicit programming. Machine learning can accommodate for a wide range of human idiosyncrasies, such as: Complexity -- when tasks are non-linear, high dimensional, and go beyond the surface Low SNR (needle in the haystack) -- when events are few and far between, and filled with noise in the meantime Incomplete information -- when only some of the necessary information is present Probabilistic events -- when events following other events are likely, or not Memory -- when problems, conditions, and data change, the system learns and forgets
A MARKETER'S DREAM COME TRUE
By harnessing the power of machine learning, human neural pathways are replicated by machines. For marketing, the result can be the ability to personalize the consumer experience for a lot more people than could fit into a corner market. The experience feels authentic to the consumer and cultivates a real brand-consumer relationship, while creating lifetime value. So, what are the human intelligence heuristics that we can mimic with machine learning to create these kinds of results Here are some of the important ones: Anchoring -- Comparison of past and future Availability -- Calculation of probabilities Representativeness -- Grouping things together for patterns Gains and losses -- Providing opportunities to win and avoid losing (Note: People don't want to lose, they want to win. Loss aversion creates inertia to stick to your current status. People are twice as miserable for a loss than for a gain of the same item.) Status quo -- Providing a desirable default state (Note: People tend to stick to what they do. Trial subscriptions exploit this. This heuristic also highlights importance of defaults in a system.) Framing -- How the information is presented against an alternative Even last year, these sophisticated advances had marketers scrambling to harness all the data that was being generated by consumers opinions expressed on social media, having the insight into the products that consumers most recently purchased and what they were most likely to buy next. For the first time ever, we have the computing power to scale combining the structured data from CRM databases to unstructured data from social networks and free flowing real time data from devices and the Internet of Things (IoT).
THE TRIFECTA: MARKETERS, HUMANS, AND MACHINES
So, how does machine learning help relationships between brands and customers It takes the 1:1 dynamic and propels that dynamic to a massive scale. Building high quality customer relationships with high lifetime value can't be done simply from machine-human (the old spray and pray dynamic), and it can't be human-human (this doesn't scale). The answer instead is in the human+machine-human dynamic. For marketers, humans AND machines are better than humans OR machines. We've come a long way in harnessing machine learning for effective marketing. It will be exciting to see where machine learning continues to take us.2018 E-COMMERCE
TECHNOLOGY PREVIEW
BRYAN CHAGOLY VICE PRESIDENT OF TECHNOLOGY BAZA ARVOICE
AUTOMATION, AI AND MACHINE LEARNING WILL IMPROVE 1-TO-1 PERSONALIZATION IN E-COMMERCE
To stand out in today's noisy e-Commerce environment, retailers are working toward making the online shopping experience as convenient, relevant and enjoyable as possible for consumers. From a technology standpoint, one area where we're seeing retailers focus and innovate is improving the customer experience through personalization -- recognizing individual shoppers, recommending relevant products for them, and directly marketing to them based on their past shopping activity. With the combination of new technology and customer data, retailers are getting closer to providing a 1-to-1 shopping experience for consumers -- an experience where the retailer can communicate, recommend products, and provide relevant offers as if the shopper were interacting with a live customer service representative or sales associate at a physical store.
USING SEARCH, BROWSING AND BUYING DATA TO IMPROVE PERSONALIZATION
Even with technology and automation, true 1-to-1 people-based personalization is incredibly hard to achieve. For retailers to be successful, they must focus on building comprehensive people-based models that include more robust signals than what can be seen from just one e-Commerce site. Consumers tell us what they are in-market for every day, but retailers must pay attention. Based on the websites they read, retailers they shop at, reviews they leave, and places they go, consumers are signaling intent everywhere -- it's up to retailers and their technology partners to synthesize this information to find these consumers and provide relevant content and enjoyable shopping experiences for them. Retailers should leverage systems that can combine, analyze, model and automate these data signals to get closer to true 1-to-1 customer interactions. In addition to collecting a rich set of shopper data, retailers must also understand what qualifies as relevant shopper data because high-quality and fresh real-time signals of intent are essential for personalizing the shopping experience and providing timely recommendations. Marketers must evaluate the freshness of their first-party data to remain relevant and ensure that their digital efforts are identifying consumers with the greatest propensity to buy. Retailers must better defin