Amazon Effect: Driving New Opportunities and Challenges in Demand Forecasting

The ‘Amazon effect’ has already changed the landscape of retail marketplace but the disruption goes much deeper. As a result, demand planning and forecasting is very different to how it was just a few years ago. With e-commerce comes e-planning and if you are already not forecasting and demand planning differently you will be.

Where there once was a corner department store now consumers shop online and want a shopping experience customized to their needs. The effects of this new omni-channel world are not only the need to plan faster but it is impacting what we forecast, how we forecast it, and when we forecast it. All of this creates challenges but there are also key opportunities presented by the new e-planning environment as well. To exploit these opportunities, we need to be aware of this different type of planning and of the value of the different types of information and then subsequent exploitation of it.

Brave New World

Traditionally, forecasting techniques were based on historical demand and the assumption that history repeats itself. While these methods and principles generally hold true, with e-planning many traditional forecasting models and the technology of the past struggle to keep up. It is a brave new world for demand planners where predictive analytics encompasses a variety of new statistical techniques like probabilistic modeling, machine learning (ML), and data mining  that analyze current and historical facts to make predictions about the future.

In today’s business environment, changes in the marketplace are swift, sudden, and may not follow the historical pattern. Just looking at historic shipments will not give you what you need and tell the whole picture. Instead we look at patterns of consumer behaviors and other attributes to try to not only predict the sell but understand why they purchased it in the first place. The new e-planning environment is dynamic and it operates on the power and speed of technology and innovation. Prediction is becoming more about behavior than history. This is powerful because once you understand the drivers, you can influence demand like never before.

With new modeling, comes new inputs and the means of collecting the data you need. Third-party syndicated data from Nielsen and others could help you better understand markets and competitors in traditional retail stores but now we have web crawlers that traverse multiple sites and bring us relevant data whenever we want. Instead of looking at just shipments or sales history, we have access to website clicks, rankings, and the number and sentiment of customer reviews. We have a new wealth of information all of which needs to be understood and modeled and translated into real-time forecasts.                                               

Planning in this e-commerce environment now means you collect data, plan demand, and micro-target at much lower levels of aggregation and time. This means we may no longer have weeks to put together the next demand plan and are dealing with changing prices or impact of new reviews hourly. With e-commerce, you are competing almost in real time with price, features, and delivery promises and feedback comes just as quickly in the form of reviews and competitive responses. To be more agile, companies are looking at technology and demand sensing techniques to translate the drivers into rules based or machine-learned responses.

This brings us closer not only to the level of demand, but also closer to demand intent. Where traditional demand sensing focuses purely on information from a customer relation management (CRM) or point of sale (POS) data from retailers aggregated weekly, you are now absorbing sales directly on an hourly basis or even quicker. At the same time, for those of us who aren’t Google, Amazon, Facebook, or Starbucks, many companies  struggle to take advantage of consumer insights and this digital revolution. Forecasting and demand planning is still only a supply chain problem to generate a discrete demand signals to assist operations and they miss opportunities this new world is offering. While artificial intelligence (AI) ML are buzzwords many technology providers are stuck in their old methods and still struggle to adapt to this new e-planning environment. In addition, for some companies, big data is as much a problem as it is an asset.

We need to better understand this new environment and we need to forecast and plan differently the winners in this new era will be the ones that can see, interpret, and act most efficiently.

A New Experience

By understanding what drives the consumer and forecasting e-commerce faster and better, companies can plan their business strategy to take advantage of e-commerce’s significant impact. Working with demand planners and focusing on new models, sources of data, and technology we can learn how to provide our customers with what they want a personalized shopping experience, affordable price, and a wide variety of available products.

Amazon has totally revolutionized the marketplace, and with it, demand forecasting and demand planning. The question may not be if the Amazon effect has in filtered your planning but when. Keep in mind though that this just happens to be the newest disruption that is impacting more than just retail sales but changing the way demand planning is forecasting and doing business. Research from the Institute of Business Forecasting (IBF) shows that this is just the beginning and the demand planners’ role will continue to transform. It goes to reason that with the changing landscape of the internet of things, AI, and unstructured data that things will continue to change, and we will need to innovate and adapt. Those left in its wake have no choice but to embrace change, technology, innovation, and find new ways to forecast and plan.