Consumers continue to change the way they shop (buy anywhere, receive anywhere, and return anywhere), which has had a profound impact on how inventory decisions
are made. Get it right, you sell more. Get it wrong, you lose sales.
As such, many retailers are leveraging new technologies, like AI, to completely transform how they address inventory decisions today to help overcome the challenges of omni-channel retailing. We recently brought together a few retailers and Celect (a SaaS solution provider focused on retail inventory optimization) for a discussion around inventory strategies that businesses are using to mitigate risks and leverage their stores more effectively to meet the ever-growing changes in customer preference – along with how this affects existing processes from an organizational and technological standpoint.
Here are four key takeaways from our discussion with these retailers around addressing some of these challenges:
- Compensation and Measurement Needs to Change
In order to really drive the flexibility consumers expect, retailers need to change their current approach to compensation and measurement. This starts at the corporate level and goes right down to the store level. Those retailers who attended shared their experience with how business results actually improved when the compensation moved from “compensation by channel” to a focus on progressing a sale and completion. While it may require a shift in culture, it will help in the long run for retailers to really use inventory more efficiently across channels.
- Data Science is Key to Inventory Success, Yet Questions Around Organization Persist
Everyone agreed that data science is key to transforming the economics of the balance sheet. However, the challenge lies in how to best infuse science into existing inventory decision-making processes. One retailer in particular discussed their experience in transitioning the organization of their data science team from centralized to decentralized (where the science resources actually sat within the functions of each business unit). This was an important move because regardless of whether a retailer has an in-house data science team or not, what matters is being able to provide the business functions (or users) with science-based tools to deliver actionable insights. This is especially true when it comes to adopting AI technology to overcome many of the traditional inventory forecasting challenges retailers face. A prime example of why this is important can be attributed to Amazon’s success, where AI is built into the backbone of their business and is embedded in everything they do.
- You Can Get Started, even if Your Data Isn’t Perfect
The consensus among retailers was this: your data does not have to be perfect to get started with AI. This was particularly important to note because most retailers do not have all the data they want or need. The difficulty is finding which solutions will allow you to start with imperfect data, get value based on the data you currently have, all while still being flexible enough to add new data sets as they become available over time. One of the retailers in attendance shared their excitement about how their allocation optimization tool will change their day-to-day job because of this ability – and will only continue to deliver more and more value as the AI system continues to learn from new information added over time.
- Returns Drive In-Store Sales – A Win/Win
Even though the percentage of returns has continued to grow due to e-commerce shopping, everyone agreed that stores can be used as a competitive advantage. Consumers like to return goods to a store because it’s easy and often more convenient. Once a customer is in the store to return an item, the goal is to get another sale. Many retailers today are using the “returns to store” as an option to consider for store fulfillment, making the return to store option a win/win scenario for both the consumer and the retailer.
Do you agree with these challenges or are there other, more pressing challenges within your organization? We invite you to comment below.