Measure Your Holiday Forecasting Accuracy in the New Year
January 5, 2024
by Carrie Bradfield
With the retail holiday season behind us and the new year laid out ahead, it is a great time to reflect on the accuracy of your forecasts from the holidays and where you might have opportunities for change and improvement.
The Impact of the Holidays
It’s easy to put a huge focus on the impact of the holiday season in the retail industry. The mythology around the origins of the term “Black Friday” plays into the importance that the holidays have in determining the success of a retail business. However, while that story has proven false, the holidays still have a large impact on the retail industry.
When we look at the U.S. Census data¹ on retail sales, we see that in some sectors, the season does impact the year as a whole. If we think of the year as four quarters, you’d expect each one to result in roughly 25% of the year’s revenue, but for segments like toys, department stores, and even clothing, the last quarter of the year comes in at over 30% of the year’s revenue. The holiday season is important, but it’s not the only impactful time of the year for overall sales.
You will find more infographics at Statista¹
Forecasting Accuracy All Year Long
How did your holiday actuals perform against your forecasts? The busiest time of the year can also be one of the most volatile. The early sales results released by Mastercard showed an overall trend of spending up 3.1% in the retail sector. Still, your sales performance versus your demand forecast has implications reaching far beyond the overall market view. The accuracy of your demand forecast plays directly into your labor spend, customer service levels, and ability to convert traffic to sales. Every hour of overstaffing or understaffing can directly impact your bottom line, employee satisfaction, or even customer service reputation.
The Economic Impact
Looking at the most obvious effect on the business, labor spending is an easy target. In a Forrester study on the economic impact of using Legion’s AI-powered demand forecasting and schedule optimization, organizations could see the elimination of 10 hours of overstaffing each week for a single store. As the average hourly rate of all retail workers in the U.S. hit an all-time high of around $24 in 2023, according to the Federal Reserve’s numbers, that 10 hours a week starts to add up quickly. Over a year, a retailer with 1000 stores would see a savings of nearly $12.5M—a compelling reason to scrutinize how your demand and labor forecasts are stacking up against your actuals.
Legion’s 2023 State of the Hourly Workforce report shows that 62% of hourly employees said they plan to leave their jobs in the next 12 months, and 48% cite undesirable working conditions as a reason for leaving. Having to take on additional tasks due to understaffing leads to burnout, which makes employees consider leaving. The cost of turnover is high, even for these hourly employees, coming in at around $3500 for recruiting, training, and other indirect costs.
Since some of the largest spending growth in 2023 is in the online sector (6.3% compared to 2.2 for in-person stores), retailers need to consider ways to protect their brick-and-mortar stores. One of those ways can be through good customer service. Marshall Fisher shared a case study in the Harvard Business Review showing that a focus on accurately predicting your demand and fine-tuning your labor model to match that demand, combined with targeted employee training, can increase in-person and online sales, with the expertise and availability of staff providing a better experience for the customers.
Have you taken a look at your forecasting numbers? Is it time to see if you can do better? This is a great time to start looking at your forecasting and labor demand generation tools. Here are a few things to consider when you’re looking to improve your forecasting accuracy:
- Do you have predictable data? Your demand forecasting tool should be able to tell you how predictable your data is. Predictability analysis, like the tools available in Legion’s Demand Forecasting, gives you confidence in the results of your forecasts. If your data set scores low on predictability, maybe it’s time to look at other ways to drive workload. Sometimes, it just takes combining two similar data feeds to increase predictability. Don’t just guess whether you’ll get reliable results—you should know up front and make an informed decision about how to drive your labor model.
- External factors like weather and local events like graduations, sporting events, concerts, etc, may impact your forecasts. Look for a demand forecasting tool like Legion’s that automatically considers these events when forecasting, finding any correlations between your data set and these events automatically.
- Look for forecasting tools that can learn from your data and update the methods used for forecasting each week and each store. Every single one of your stores is unique, so you can’t assume what works for one will work for the other. Use the power of AI to decide each week, for each store, what the most accurate forecasting model will be, and let that process continue through the year so it can easily react to seasonal changes or even sudden world events that may impact forecasts.
Considering these factors, consider checking out Legion’s product information for Demand Forecasting and maybe even try the Demand Forecasting Accuracy Challenge. Make 2024 the best year for your forecasting accuracy—the benefits are clear.
To learn more, contact us today for a demo of Legion WFM.