- 5 Key statistics about the retail industry to start
- Why is demand forecasting important in retail — especially in 2021
- How is demand forecasting done accurately?
- Common demand forecasting pitfalls and how to avoid them
- Off you go
Demand forecasting is used in many industries, from small businesses to massive franchises, for tasks such as financial planning, customer success management, and supply chain control. Forecasting allows them to take preventive measures to avoid loss in clients or sales while enabling their businesses to become more agile as they adapt to new and sudden changes. In addition, new opportunities to increase revenue can be identified using forecasting methods.
From there, companies can formulate strategies to optimize their operations and spending patterns. Whether they need to allocate more funds to certain products in the inventory, intensify their marketing and promotional efforts, or halt the hiring process of new employees, sales and demand forecasting is necessary to predict future financial situations and make more informed decisions.
In this article, we take a look at the importance of demand forecasting in the retail space and how businesses can perfect their forecasting methods, regardless of their organization size.
US retail sales dropped 0.7 percent in December 2020.
Retail sales are projected to amount to around $26.69 trillion by 2022.
3 out of 5 consumers say retail’s investment in technology is improving their experience both online and in-store.
65 percent of retailers plan to invest in new products.
Small retailers hire 39.8 percent of all retail employees.
With the worldwide outbreak of the COVID-19 pandemic in 2020, entire countries came under quarantine orders and almost completely shunned the physical interaction between consumers and retailers. The situation has left many retailers scrambling to adapt to new consumer demands and forced many organizations to make structural changes faster than they had previously planned. One of the key weaknesses the pandemic has exposed is the lack of proper demand forecasting in the retail space, leaving 29 major retailers across the U.S. no choice but to file for bankruptcy.
The retail industry has shown its vulnerability to sudden disruptions. No matter what niche you are into, it is essential for your retail business to forecast consumer demand if you are to stay in the game. Yes, it may not be easy to accurately forecast what’s next, given the current context. Yet, smart retail businesses should start thinking of all possible different scenarios and plan their next move accordingly.
If you follow the Competera blog, you are probably familiar with the concept of demand forecasting in price optimization. However, this phenomenon is much broader and here are the 4 main advantages of properly forecasting demand:
1. Allows smarter financial planning — Information derived from accurate forecasts allows retail companies to better plan financially for the future. Projections can clearly outline peak periods, seasonality, and demand trends by months, days, or even by the time of the day. By planning accordingly, companies can ensure that they have sufficient cash flow to prepare for upcoming peaks, orders, and unexpected expenses.
2. Helps prevent staffing problems — Forecasting can help correct staffing problems that retail companies may encounter, especially during peak seasons. By estimating the rise and fall of demand each month, day, or hour, businesses no longer have to rely on naive guesswork to schedule retail workers. Instead, managers will have a better idea of which shifts will require the most manpower and be able to plan ahead.
3. Enhance the development of marketing plans — Forecasting is also beneficial for developing effective marketing plans. For example, when companies can predict a dip in sales, it is especially important to increase their marketing efforts. Marketing plans developed according to demand forecasts allows you to prepare tailored promotions to fill that gap and even adjust your pricing strategy according to customer behavior.
4. Improved inventory management — Forecasting is also useful to companies looking to improve their inventory and production management. The last thing owners want is an inventory overflow when sales are down or popular products going out of stock during peak periods. Accurate demand forecasting gives them the foresight to adapt the supply chain to meet the demands of specific products and services while ensuring that the company has the shipping, materials, and labor available.
There are 3 models of demand forecasting commonly used in the retail space. Keep in mind that each model has its flaws but can still give you an edge when it comes to predicting consumer demand. The most accurate way to forecast demand is by using both internal and external data.
Internal KPIs (key performance indicators) involve the historical number of sales, the amount spent on ads, and store traffic (website or foot). On the other hand, external metrics take into consideration emerging customer trends, industry changes, and even your competitors' doings.
Sadly, many smaller retail businesses or start-ups ignore demand forecasting because they have limited resources and very little information to work with, unlike the big brands that have accumulated years of data. Nevertheless, it is quite possible to produce reliable projections to guide the company to success. Here’s how:
1. Qualitative demand forecasting
This type of demand based on qualitative data. Some examples of sources of qualitative data include industry authorities or experts, focus consumer groups or even competitive analysis. This data is mostly based on gut-feeling or intuition instead of researched statistics or hard facts. This model is generally recommended for retail businesses that do not have enough historical data to analyse.
2. Causal model
Forecasting demand with the causal model takes into account factors that can change the initially predicted demand. In this model, the data is split into controllable factors, such as product pricing, marketing efforts and location, and uncontrollable factors like trends, competition, political reforms and even natural catastrophes. The causal model provides demand forecasts by combining data and intuition and is mainly used by data-driven retailers who have accumulated a lot of metrics over time.
3. Time series analysis
The time series approach is more inclined towards quantitative data. This model leaves all the ‘guesswork’ out and focuses on hard facts and statistics. This model relies heavily on a mathematical approach and is often considered rigid. Ideally, to use this model to forecast demand, retailers need to have previous data at hand, such as sales numbers for the past years, seasons, best selling items and changes in pricing.
At Competera, we use a complex approach to forecasting based on the client’s particular needs. Of course, in some cases, either casual or time series models are enough but a more sophisticated solution might also be needed. For example, Competera data science team has recently used a recurrent forecasting approach to design a model for a high-seasonality business. Eventually, an average of 96+% accuracy of forecasts was gained.
When not done accurately, demand forecasting can end up sinking your business productivity and potentially lead you to spend more (or less) than you should actually do. Whether it's a small business or a large corporation, relying only on the day-to-day management of operations is very often considered poor planning.
Here are some of the most common mistakes in retail demand forecasting and budget preparation:
Overestimating sales — It can be easy to overestimate sales figures. In fact, it may even seem like a safer option, since it could prevent underspending. However, overspending could result in unnecessary capital losses, which will hinder long-term growth potential.
Ignoring historical data — It is essential to keep past sales data in mind when building plans and projections. If you have past data at hand, use them for your future forecasts and base your predictions on your previous results.
Relying only on gut-feeling — Taking decisions based purely on gut-feeling is never a reasonable call for a business. Although sales forecasts are estimates, they should at the very least be based on data or facts. This way, they are always more likely to produce more achievable results than simple assumptions.
Lack of flexibility — It is also possible to rely too much on historical sales data. Good sales plans and budgets should be prepared to adapt and take into account the potential for abrupt changes. Of course, this is not always easy to achieve, but allowing some flexibility into your forecasting gives you a chance to recover from unforeseen circumstances or uncertain times.
Using multiple spreadsheets — Relying on outdated spreadsheets is not ideal. Manual calculation and data entry can leave a lot of room for human error. The key to smarter forecasting and budgeting is to keep it simple. Try automating this process with a budgeting software to save you time and potentially create more reliable forecasts with less effort.
Not updating forecasts regularly — Even if managers are confident in the data they have produced, it is important to update and refine this information as often as possible. Due to seasonal factors and unforeseen events, such as the COVID-19 pandemic, business projections will need to be re-evaluated frequently to obtain the most accurate and up-to-date information.
Forecasting is essential for your retail business if you are to play the long-term game. The 2020 COVID-19 episode has shown how important it is to forecast and adapt if you are to survive uncertain times. Developing a sustainable retail business starts by properly forecasting all key aspects to ensure your organization is running as close as possible to its optimal condition on a daily basis.
The article is co-authored with Derek Jones, a spearhead of key initiatives at Deputy, a global workforce management platform for employee scheduling, timesheets and communication. With a focus on Healthcare, Derek helps business owners and workforce leaders simplify employment law compliance, keep labor cost in line and build award-winning workplaces. Derek has over 16 years’ experience in delivering data-driven sales and marketing strategies to SaaS companies like MarketSource and Griswold Home Care.
The lack of data or its poor quality as well as wrong results' interpretability represent the two major challenges associated with demand forecasting.
The forecast of 95%+ accuracy is generally considered to be a good one. But the quality criteria may also vary depending on the area and complexity of the forecast.