NOVEMBER 2017CIOAPPLICATIONSEUROPE.COM9HOW DO WE LEVERAGE THE PROMISE OF THIS HUGE CHANGE CALLED ARTIFICIAL INTELLIGENCE?a voucher for a child's meal or a reminder to book a table on Mother's day and so on. Communication quickly becomes personal, individualized, and more importantly highly relevant.Further through the process and to provide the customer with a welcoming experience, CRM data can be passed in an easy to understand visual way to the server's hand-held terminal. This will offer a server the opportunity to engage on a personal level ­ `Can I get your favourite drink while you look at the menu?' or treating you extra specially because last time you visited something was not quite right or simply because it's your wedding anniversary. Even suggesting a different drink based on the AI's analysis of your preferences, demography, occasion, and time of year. The possibilities are endless ­ surprise and delight a technique used increasingly in hospitality venues become almost a seamless experience. More importantly for the business, net promoter scores rise inexorably, and our engagement is highly personal.Live Wi-Fi connectivity data fed back into CRM for further analysis can be used to modify the digital signage programming on advertising screens allowing advertising to be targeted precisely at the type of people in the bar at exactly the right time of day. Coffee at lunchtime, Xbox and games media late at night, thereby maximising advertising revenues from venue displays.On a generic level ­ machine learning techniques can take big data from online review and social media platforms, match that with your own surveys and sales data to identify patterns and trends in consumer taste, allowing the business to prepare and produce a new product and product combinations before the customer knows they'll want them. As Jobs said, `Customer's don't know what they want until you tell them'!Sales history combined with event information, matched with on the hour weather predictions at postcode level allow managers to more accurately predict bar staff numbers by the hour and day, optimising both labor costs and consequently, the customer's experience. The same information coupled with immediate sales data feeds allows bar and restaurant pricing to be optimised dynamically and appropriately and in near real time to both supply and demand.IOT, or the `Internet of Things' potentially streamline stock to sales processes, minimising stock loss, and optimising sales revenues. Data from sensors monitoring outside and internal temperatures or fridges with knowledge of `Use by dates' provide data that AI can then utilize to create on the fly specific promotions advertised on wall boards to quickly sell near out of date products. Ice cream prices, dynamically increased on a hot summer day and promoted on screen squeezes extra revenue. Moreover, if you get your working premise, process design, and the parameters right ­ it's feasible without human intervention. There's so much more to go for, but to me, right now, it's not so much about the machines, and it's not really about the robots ­ it's about the data, understanding it, and knowing what to do with the conclusions. Can a machine do that? Quite possibly, of course but not quite yet, though it doesn't hurt to try. Identifying patterns, trends, and drawing simple conclusions are all possible today but deciding what to do about some situations can still require a level of lateral creativity not yet available to silicon.Conclusions? The Retail sector clearly has much to gain from leveraging A.I. We're only at the start of this next revolution, we can and should take advantage now so that what we learn helps refine further what shape A.I. takes in the future. That shape has enormous promise certainly but I'd also ask that, we remember we're all humans and we have a responsibility for those that come after us. Doing it just because we can, is no excuse for not using our intelligence. The real opportunity comes from AI's incredible ability to support the restoration of human to human interaction
< Page 8 | Page 10 >