Perhaps Mick Jagger and Keith Richards of the Rolling Stones, really were onto something in 1965 when they first sang “I Can’t Get No Satisfaction,” especially when one considers that the band’s lyrics still resonate with audiences some 55 years later.
A lot has changed in business-to-consumer organizations like retailers, restaurants, and hospitality organizations over those five-plus decades, especially when one considers the essential role that consumers now play in telling enterprises exactly what it is that they want, and when they want it. This shift in consumer behavior has not stopped these enterprises from struggling to accurately anticipate what to stock, despite the fact that inventory management software has been around for just about as long as that aforementioned Rolling Stones song. Consumers are still getting “No Satisfaction” since all too often what they want is not in stock.
But why is this the case? There is more timely information, better technology, and more data than ever before, which should make it easier to avoid overstocks or out-of-stock situations ever again, right?
So, why is there a renewed sense of frustration by consumers?
The answer is actually simple; e-commerce and digital transformation, has changed everything. It has changed consumer behavior forever; and retailers, restaurants, and hospitality organizations may be stuck with technologies and processes unfit to weather this transformation.
It is no surprise that the clear winners in today’s world of commerce have real-time perpetual inventory management systems capable of dynamically interacting with online and instore transaction processing systems. They are able to proactively avoid lost sales attributed to out-of-stock inventory while simultaneously reducing non-value-adding inventory that needlessly ties up working capital. Winners also have less shrinkage and spoilage of inventory as well.
One of the problems that exists with most legacy inventory management systems is that they are built on top of legacy demand forecasting algorithms that lack channel-specificity; failing to understand online shopper demand patterns. More clearly, existing demand forecasts show a bias towards in-store shopping and will be much less accurate for digital channels.
When the above occurs, consumers are easily frustrated because they do not actually end up getting everything that they ordered. In fact, in some retail channels particularly, consumers may only get ~80% of what they originally ordered online, resulting in the consumer being forced to accept substitute products.
The solution? Enable your data to harness the power of Artificial Intelligence (AI), Machine Learning (ML) and Data Science to autonomously revolutionize your demand forecasting and inventory so that out-of-stocks and overstocks become a thing of the past.
Demand forecasting in the 2020s must be AI-powered, autonomous, real time, provably accurate and channel specific. It must understand both in-store and online demand signals. Forecasts must be accurate by item, by store, by date, and by time of day. The demand forecasts must account for the effects of price, promotion, assortment, also accounting for seasonality, events, holidays, and weather.
With the proper autonomous demand forecasting in place, the ideal solution for most companies will be a real-time perpetual inventory that can support available to promise commitments for customers. Many business to consumer companies elect not to use a real time perpetual inventory because their receiving, inventory, and POS scan practices lack discipline and integrity.
An AI-powered autonomous inventory solution is a highly attractive alternative helping retailers, restaurants and hospitality organizations to auto-detect and proactively avoid inventory scarcity issues or resolve them in a timelier way without all of the disciplines and integrity required for a true perpetual inventory.
What does AI mean for you?
Few would disagree that there is a sort of AI fatigue and posturing permeating the world of inventory and forecasting across the supply chain. From movies to social media suggestions, video adverts to traditional coupons, it seems that the ad men of Mick Jagger’s days are still telling us what we want, but now with 1’s and 0’s.
At its core, AI is a powerful set of tools that can, with the right data, make quantitative analysis easier, faster, and more robust. This differs from many standard definitions that assume there is some sort of magic algorithm being leveraged by incredibly powerful computers that will increasingly replicate human behavior, facial recognition, speech patterns, and the like.
To be clear, there is a lot of great work going on in the data science fields of machine learning and deep learning that help non-technical decision-makers harness data-driven insights without becoming overly dependent upon IT or in-house data scientists. This is critical as in no way will data completely replace the strategic expertise and intimate knowledge decision makers have about their business, but rather, make more concrete their resolve going forward on new initiatives.
AI is not going anywhere, of course, but where specifically it goes within inventory is certainly up for debate.
How AI is being used to understand inventory today
At present, AI has an unclear, customer-facing role. There are frustrating bots and chats on websites as well as less-than satisfying emails alerting consumers of when an out-of-stock item might be back in inventory. AI, ML, and other data science models in these roles is a recipe for disaster, especially as its strengths could be better leveraged in the deep, dark confines of back offices.
Consider, despite a near infinite number of inventory management software programs and inventory software solutions, that the demand planner and forecaster at even the world’s largest retail chains are still perplexed by the simple question, how much inventory should be on hand on any day and at any one time?
This is all in spite of the fact that AI, ML, and other data science models can deployed to help make sense of reams of data from legacy and disparate inventory management systems and, in so doing, make use of cloud software that makes sense of historic consumer buying patterns. In other words, and when the data is validated and verified as accurate, AI can aggregate and analyze consumer demand, supplier orders, production orders, re-order points, economic order quantities, and much more.
Inventory optimization, however, and as briefly alluded to at the start, requires more than smart machines and historical data in order to avoid future stockouts. Understanding why certain processes exist in the first place and how they no longer serve as a means to effectively minimize the risk of out-of-stock situations as well as prevent safety stock from becoming deadwood is imperative. Ensuring compliance with standards as well as implementing an appropriate inventory management system means that inventory planners will be able to better anticipate “how much” to order, and just as importantly, when to make those orders.
To be clear, while AI has made significant advances over the past few years, it is not a cure for poor company disciplines or a legacy inventory management system. In other words, AI can be used to transition from instinctive inventory management practices to those based on data science and statistics. By aiding retailers, restaurants, hospitality organizations and CPGs in their identification of those processes and practices that are resulting in unreliable inventory levels, AI can do most of the heavy lifting of removing uncertainty and variation from decision making.
Even though AI cannot predict the future with 100% certainty, a robust and intelligent platform can run far more combinations than any statistician and calculate a near infinite number of permutations that can, at the very least, guide planners and schedulers on what items are at risk of a stockout. Once a probability is known, decision makers can identify an appropriate amount of safety stock in which to have on hand and begin to prepare accordingly.
Assuming, of course, that there is no risk of a black swan event waiting in the wings (sorry, bad pun—we know!).
To say there is absolutely zero risk of some earthshattering event disrupting supply chains the world over is to laugh in the face of COVID-19 as well as other momentous events like it.
Unlike other events of a similar magnitude in the past, however, AI is yet a new capability in which retailers, restaurants, and hospitality organizations can now leverage to put their businesses back together. In short, all businesses were thrown a curveball, but AI, ML, and data science can react to today’s challenges with greater ease and accuracy.
That said, and in partnering with Hypersonix, entire fulfilment systems and logistics pipelines benefit to the satisfaction of customers around the globe.
This venture-backed company out of San Jose, California, has a proven track record of making sense out of disparate data sources in order to provide decision makers in retail, hospitality, e-commerce, and the like with actionable insights across their inventory. What’s more, and has been discussed above, is that Hypersonix goes beyond the “silver bullet” cure of AI and offers enterprises a more robust platform in which to leverage their data for internal decision-making purposes. The Hypersonix solution is both, at once, an AI-driven platform that automates certain processes while simultaneously looking for the root cause behind an issue so as to ensure satisfying customer-based outcomes.
In summary, and in looking to make consumerism less fraught with frustration on behalf of both the supplier and consumer, AI should be playing a much greater back-of-house role. Artificial Intelligence is making significant inroads into pricing and margin management, personalization, and loyalty program management to the great consternation of many, while missing out on the opportunity to improve, once and for all, real-time inventory visibility, perpetual inventory processing, omnichannel order fulfilment, and inventory optimization.
If your enterprise is ready to move beyond the guesswork of inventory management, then request a demo with Hypersonix, so as to not only automate your inventory intelligence, but also go above and beyond for your consumers.
Hmmm, maybe there is, after nearly 55 years, some hope of getting satisfaction out of all that underutilized information.