AIBest PracticeDemand ForecastingInventory Planning


Apr 12 2023 ...

Advanced retail tech picks up where WSSI leaves off

3 ways AI enhances WSSI for better merchandising decision-making.

Merchandisers are responsible for making mission-critical business decisions every day. To do so, they must be able to monitor and control inventory across their networks. Likewise, senior management must also have visibility of the reality on the ground for high-level decision-making.  One industry-standard practice, “weekly sales, stock and intake” (WSSI), enables stock availability by analyzing sales, inventory, and forecasts to manage budgets, warehouse space, and margins on a weekly basis. It evolves as actions are reflected in subsequent reports. And for the large part, WSSI seems to have stood the test of time, its popularity due to its fairly simple and easy-to-use-and-understand premise.

But it’s that same simplicity that makes WSSI (“weekly sales, stock and intake”) too blunt of a tool on its own for today’s merchandising complexities. Ongoing disruption, an increasingly blended online and offline landscape, and constantly shifting customer demand mean that merchandising must be faster, more dynamic and more ongoing in order to walk the fine line between customer-centricity and cost-effectiveness.

Next-gen AI-driven merchandising solutions, however, bring the intelligence of machine learning, the scalability of cloud computing and the efficiency of automation to revolutionize how WSSIs are approached.

Here are 3 ways AI-driven merchandising solutions enhance WSSI for better continuous decision-making.

1. Drill down for enhanced visibility and better, ongoing decision making

Traditionally, merchandising teams and retail executives have used WSSI outputs to make critical inventory decisions based on insights into sales performance and stock levels. However, there are some limitations to doing so.

WSSIs are often simplified or aggregated to calculate at the category or class level to avoid over-complicating reporting. But by sacrificing granularity, a true vision of reality on the ground can be lost. Similarly, WSSI is often used to calculate by channel, isolating stock instead of unifying it. But how can you make the best inventory decisions at all times with an oversimplified and siloed view of inventory?

With AI-driven merchandising solutions, on the other hand, you don’t have to sacrifice granularity for the sake of ease. You can drill down to the SKU/point-of-sale level to bring you much closer to the reality of demand, capturing changes and allowing you to adapt to them faster.

And being able to analyze data from the bottom-up gives you the power to sense even subtle market changes in real time for much more accurate demand forecasting. It also allows you to take a unified view of your inventory, as opposed to having to silo it by channel, so that you’re best primed to tap into full-sales potential at all times.

Through AI-driven demand forecasting and optimization, you always know what the best next step to take is with your inventory, making it as “tight” as possible, and avoiding top-down misalignments that often lead to overcompensation in producing, buying, and allocating which inevitably leads to storage issues, discounting and waste.

Since AI merchandising solutions can deliver more accurate results , WSSI becomes a stronger monitoring tool rather than a decision making one.

Read about the 3 most important data sources for forecasting retail demand.

2. Automated merchandising processes for shorter decision cycles and fewer errors

Generally speaking, WSSIs have been deployed through spreadsheets like Excel or dedicated Merchandise Financial Planning tools. While these tools usually enable some degree of automatic data collection and calculation capabilities, some manual input and analysis is generally required, making the decision-making process a slow, time-consuming, and error-prone task.

And as retailers grow their businesses and move into new channels, their operations reach new levels of complexity in that not only will they simply have more data, they’ll need to tap into and process exponentially larger, newer data sets – a fact that will make any manual calculation, and resulting decision-making slower, and more error-prone, hampering the accuracy and reliability.

AI-driven merchandising solutions overcome these limitations by automating complex calculations and generating accurate demand, rather than sales, forecasts. What’s more, thanks to advanced statistical models, machine learning algorithms, and the scale provided by cloud-computing, next-gen AI merchandising solutions can even analyze vast amounts of internal and external data such as historical sales trends, customer behavior, market demand signals, and even real-time data, breathing life into your data with speed and precision.

The advantages of AI merchandising solutions mean that as retailers grow, they don’t have to worry about processes taking longer or getting more complicated, and can actually benefit from additional data to identify hidden patterns and correlations that humans or spreadsheets might miss or get wrong.

The result? More accurate and data-driven inventory decisions delivered in the fraction of a time.

Learn how the Nextail partnership with Snowflake, the world’s leading data cloud platform, is boosting processing power for calculating mission-critical merchandising decisions.

3. Missing or distorted retail data doesn’t have to set you back when forecasting demand

What would retail, especially in fashion, be without the psychological draw of “newness”? The introduction of new products is fundamental, but traditional methodologies like WSSI tend to rely solely on a limited set of an item’s past sales and performance data to determine how many units of a product you need or where to send it.

But how can you forecast for new products and set them up for success if performance data does not yet exist? The same predicament occurs when opening a new store or entering a new market. In the past, merchandisers only had their gut and past experience to work with, frequently leading to under- or over- forecasting items.

Modern AI merchandising solutions also use historical data to forecast for new products, but are also able to draw from a bigger data pool by extracting attributes (e.g. print, category, size, long-sleeved, pleated) and sales performance from currently selling or previously sold products. The demand forecasts that they are able to generate are highly granular, allowing retailers to define or redefine store assortments before sending products to points of sale. In situations where no direct comparisons exist within a specific point-of-sale, some solutions expand the analysis to consider comparable products at a regional or country level using “dynamic fallbacks”.

But even newer, state-of-the-art merchandising solutions can take forecasting a step further. Advanced machine learning can learn from vast amounts of time-series data, pulling from even broader and richer pools of data to dynamically understand different selling conditions, even without relying on pre-existing seasonality components. By uncovering patterns over time, these solutions achieve “context awareness,” resulting in highly accurate and robust forecasts that can adapt to new products with limited historical information, such as low-rotation luxury items.

Another advantage of newer best-of-breed merchandising solutions is their ability to protect forecasts from distortion and inaccuracies caused by unusual periods of demand in the past. Instead of using pre-existing seasonality components, as mentioned, these solutions generate seasonality patterns based on a richer pool of data, allowing them to recognize when demand fluctuations are caused by promotions or other events. This eliminates the need for manual tagging, or worse yet, having to rely solely on intuition to guide inventory decision-making.

Therefore, a lack of data, or distorted data are no longer issues that will hold you back from making the best-possible inventory decisions. WSSI outputs backed by AI merchandising solution outputs provide increased reliability and have a lower risk of being haunted by data gaps or inaccuracies. With the integration of AI-driven merchandising solutions, you can confidently navigate the challenges of forecasting for new products, store openings, and market expansions, ultimately improving their inventory management strategies.

The final verdict: AI merchandising solutions and WSSI can have a synergistic relationship

There’s room for a complementary relationship between WSSI and AI-driven merchandising solutions that holds tremendous potential for retailers in today’s fast-paced and complex industry.

While WSSI has served as a reliable tool in the past, it falls short in meeting the demands of modern retail. By integrating AI capabilities, retailers can unlock enhanced visibility, real-time data analysis, advanced forecasting, and adaptive decision-making.

The combination of WSSI’s foundation with AI-driven technologies amplifies the power of traditional merchandising tools, providing decision-makers with the agility and intelligence required to thrive in the competitive retail landscape.

With AI-driven merchandising solutions, retailers can unlock new levels of operational efficiency, optimize inventory management, and deliver exceptional customer experiences. Embracing this synergistic approach is not just a competitive advantage, but a necessity for retailers looking to stay ahead.

Read the Style Union case study to learn how the partnership between the growing retailer and Nextail led to 5.5K additional full-price sales (approximately €46K) just 4 months after Store Transfer go-live.