How to Mitigate Stock Discrepancies

In 2023 the UK retail market was shaken by unprecedented news. Pedal Group, who has just acquired 99 Bikes (ex Hargroves Cycles) fell right into a stock "hangover" and had to cancel a year-long line of orders. Why? They were overstocked. How could this happen? Obviously, there are known reasons for stock discrepancies that include overstocking and out-of-stocks: warehouse receiving errors, numerous returns, misplaced or lost inventory, inaccurate records of returns, outdated warehouse technology, poorly trained employees. But even all these factors combined could not lead to the devastating results a retailer had to face. You don’t cancel the year-long line of orders due to the inaccurate returns, do you? 

‍

IHL Group estimates that inventory imbalances cost retailers nearly $1.1 trillion globally. That's a trillion with a "t". It seems like the tech we've been relying on to avoid these pitfalls might be falling short.

‍

Pedal Group's story isn’t a one-off. IHL Group estimates that inventory imbalances cost retailers nearly $1.1 trillion globally. That's a trillion with a "t". It seems like the tech we've been relying on to avoid these pitfalls might be falling short. We'd argue, though, that there’s an overlooked piece of the puzzle that retailers have been missing: accurate demand prediction.

‍

The Shortcomings of Traditional Techniques and Business Intelligence for demand prediction

Back in the day, retailers leaned hard on Business Intelligence (BI) tools and standard forecasting methods for inventory management. But here's the thing: BI tools mainly forecast sales, not demand. The two aren't interchangeable. Sales forecasts look at historical sales data, market context, and how well you can sell a certain number of items within a specific timeline. It doesn't give you the skinny on customers' desire to buy or potential demand.

What's more, BI tools are usually the domain of Category Managers which means that they don’t impact purchasing decisions at the SKU level. The sheer amount of data and variables to deal with makes it a colossal task. Let's face it, even the big eCommerce players are still wrangling with Excel spreadsheets. Not exactly a scalable solution when you're dealing with hundreds of thousands of SKUs.

Let's face it, even the big eCommerce players are still wrangling with Excel spreadsheets. Not exactly a scalable solution when you're dealing with hundreds of thousands of SKUs.

The result? Purchasing decisions aren't jiving with demand at the product level. This disconnect leads to piles of unwanted stock and sold-out popular items.

‍

‍

Harnessing the Power of Predictive Analytics and Price Optimisation: The New Era of Inventory Management

There's light at the end of the tunnel. Emerging tech like predictive analytics and price optimisation is stepping in to pick up the slack, minimise human error, and streamline inventory management.

‍

Predictive Analytics

Predictive analytics uses a blend of statistics, data mining, machine learning, and artificial intelligence to evaluate current and historical data, facilitating accurate future predictions. This technology enables retailers to make forecasts at a granular level, thus aligning purchasing decisions more accurately with demand and preventing misplaced inventory.

Predictive analytics can generate optimal quantities for purchasing, allocation, safety stock, replenishment, balancing, and incremental inventory needed to conduct successful promotions. By integrating real-time data from various sources along the supply chain, predictive analytics can adjust to market or consumer behaviour shifts promptly, empowering retailers to react efficiently.

‍

Price Optimisation

Price optimisation, coupled with predictive analytics, determines the most advantageous price point for each product that will maximize demand and profitability. By adjusting prices dynamically based on real-time demand, competition, and other market factors, retailers can stimulate demand for overstocked items and maximise revenue from in-demand products.

Predictive analytics and price optimisation form a powerful partnership. Predictive analytics can inform on the expected demand at different price points, and price optimisation can leverage this information to set prices that accomplish the retailer's objectives, be it maximizing profit, reducing overstock, or avoiding out-of-stocks.

‍

Data is not an end but a means to an end 

It's true, modern inventory management systems offering real-time tracking of inventory levels and smooth supply chain management can be a lifesaver. They give us heaps of data that category managers can use to make informed decisions. But there's a catch. The sheer volume of data is overwhelming, and we risk drowning in it. The human brain just can't handle this much info. This is where AI for retail comes to the rescue. AI excels at storing, handling, and crunching big data. Plus, it doesn’t get tired or emotional. Pair AI-driven predictive analytics and price optimisation tools, and you've got a powerful combo that can help retailers predict demand and manage sales velocity at the SKU level. It seems we've got the answer we've been searching for all along.

‍