Inventory Forecasting: Methods, Benefits, & Best Practices

See into the future with good inventory forecasting. Supply chains, consumer demand and market forces are constantly changing and you need the right tools to keep up.

To create an accurate forecast, you need data, strong analytics, an understanding of customer demand and the right software. That’s just to start but what exactly is inventory forecasting?

Let’s explore how inventory forecasting works for e-commerce and retail, the types of inventory forecasting available, its benefits, the best practices and how to choose the best possible inventory forecasting software for your business.

Table of Contents

What is Inventory Forecasting?

Sometimes known as demand planning in retail, inventory forecasting predicts inventory levels for a future period by analyzing past data, trends and upcoming events. Done well, inventory forecasting gives you the tools to consistently keep stock at optimum levels. Enough to meet customer orders but not so much that you’re incurring unnecessary inventory storage costs.

Get the All-in-One Inventory Planning Software

Forecast demand, issue and track POs, re-order on autopilot, and step up your reporting game across multiple channels and locations. Get in touch to see how Singuli can help you optimize your inventory.

Important Inventory Forecasting Models and Metrics

It’s crucial to familiarize yourself with some of the most important inventory forecasting metrics. Let’s review them:

Safety Stock

Safety stock is the minimum stock you can safely operate with. If you have too little, you’ll experience stockouts and won’t be able to serve your customers, or you’ll have to put their purchases on back order.

📚 Further reading: Weeks of Supply Formula: Is it Right For Your Business?

Lead Time Demand

Demand is the amount of units you sell over a set period and lead time is how long it takes for stock to arrive at your warehouse from the point of purchase. To work out your lead time demand, you multiply the lead time period by the demand per period. This tells you the demand you need until your expected delivery arrives.

For example, you sell 100 shirts per week and it takes five weeks for stock to arrive at your warehouse. This means you need 500 (100X5) shirts altogether to cover the stock sold until your expected delivery.

Reorder Point (ROP)

You need to order the right products at the right time. ROP tells you when the right time is to replenish your inventory. To work out ROP, add lead time demand to safety stock. Keep checking your lead time demand so you can adjust your ROP whenever there’s a change.

$$\text{Reorder Point} = \text{Lead Time Demand} + \text{Safety Stock}$$


Stockouts happen when you have no more stock of a product but there’s still customer demand. It’s extremely risky as customers will go elsewhere and may never come back.

Average Sales

Average sales take account of the inventory you’ve sold divided by the time it took to sell that stock. For example, let’s say you order 90 t-shirts between June 1st and August 29th and sell them all. So, over 90 days, you average one t-shirt sale per day.

The problem with average sales is that it doesn’t include days when you had no stock available. Let’s use the same example. This time, you notice there were 45 days without any t-shirt stock between June 1st and August 29th. Using the average sales calculation won’t give you an accurate picture, but the sales velocity formula will.

Sales Velocity

Sales velocity is a more accurate measure as it only takes account of the days in stock. For the t-shirts sales velocity, divide total sales by the number of days your t-shirts were in stock. This gives a daily adjusted velocity.

The sales velocity for this period would be (90/45) = 2.

$$\text{Sales velocity (daily)} = \frac{\text{Total Sales}}{\text{num. of days in stock}}$$

Cost of Inventory

How much does it cost to order, keep and move inventory? Let’s start with ordering costs. They include the wages for the procurement department, like payroll and taxes, as well as labor costs for the other staff members involved with the ordering process.

Holding onto stock isn’t free. You need storage space and insurance. You have to pay for utility bills, maintenance, warehouse staff and other related costs. You’re also tying up money in inventory and if your stock becomes obsolete, you have to sell it at a discounted rate or get rid of it for free.

Finally, you need to think about associated administrative costs. How much does it cost to pay accountants to work out your inventory costs? There’s auditing costs, warehouse staff recruitment costs and more.

Base Demand

Base demand uses pre-sales and accurate sales pipelines, based on historic sales to predict the demand you’re confident of meeting over a specific period. For example, let’s say you’ve sold 15 t-shirts in the first week of each of the past 12 periods to the same customer. They’ve indicated that they’ll buy 15 again in the next period. This sale is extremely likely so you’ll add it to your base demand.

Average Inventory

Your average inventory is the stock you have on hand during a given period. Consistency is key to avoiding stockouts and overstocks. You work it out by adding the inventory at the beginning of a period to the inventory at the end of a period and dividing it by two.

$$\text{Average inventory} = \frac{\text{beginning inventory} + \text{end inventory}}{2}$$

Inventory Turnover

Inventory turnover is the number of times you’ve sold and replenished a product over a set period. The ratio is calculated by taking the cost of goods sold (COGS) and dividing it by the average inventory over the same period.

$$\text{Inventory turnover ratio} = \frac{\text{COGS}}{\text{average inventory}}$$

Economic Order Quantity (EOQ)

EOQ uses an economic standpoint to work out your ideal inventory quantity. To calculate it, you need to know your spend per order (order costs), your sales per season (demand rate) and the cost of storing your products (holding costs). The idea of EOQ is to minimize logistics and storage costs and to optimize sales.

$$\text{EOQ} = \sqrt{\frac{ 2\cdot \text{order costs} \cdot \text{demand rate}}{\text{holding costs}}}$$

How Does Inventory Forecasting Work For E-commerce and Retail?

Inventory forecasting uses past, present and predicted future data to make sure you have enough inventory to prevent stockouts without sitting on so much stock that your cash is tied up, causing an increase in the cost of inventory. A demand planner considers historical data, seasonality, trends, growth rates, marketing plans and more when forecasting inventory needs.

An inventory forecast looks at three key areas: the forecast period, the base demand and known variables. The forecast period can be anything you like but longer periods usually tend to be less accurate. Common forecast intervals are 30 days, 90 days or a year.

Variables and trends aren’t always easy to predict. The best software uses all relevant data to ensure accuracy and it’s clever enough to spot anomalies too.

💡Pro Tip: Get a holistic view of your inventory data with Singuli’s Reporting & Analytics. Track not only inventory data, but also marketing, promotions, site traffic, historical sales and customer behavior. You’ll have all of your information in one place, ready to use in inventory forecasting and planning.

Inventory Forecasting vs Replenishment

Inventory forecasting works out the different quantities and types of stock needed to cover expected demand over a set period.

Replenishment is the stock needed to meet your inventory forecasts. It’s based on forecasts and also considers current inventory levels, lead times, stock on order and returns.

What Types of Inventory Forecasting Exist?

Efficient forecasting is about predicting demand but there’s more than one way to do that. You can combine them or use them individually, depending on your needs. Let’s explore the different types of inventory forecasting that exist:


Trend forecasts predict changes over a set period by analyzing patterns and demand spikes. Although trends are a good measure, trend forecasting doesn’t take seasonal effects, external events or sales anomalies into account, so your forecast might not be as accurate as you’d like if this is the only inventory method you use.

Trend analysis can be done using top-down forecasting or bottom-up forecasting. Top-down forecasting looks at broader trends that affect the business as a whole and uses that information to predict trends for individual products and SKUs (stock keeping units).

Bottom-up forecasting does the opposite. Larger businesses often opt for this method as you can create a more granular forecast. This gives you a better chance of keeping optimum stock levels.


Seasonal forecasting predicts spikes and dips in demand based on events like festivals, sale seasons, weather, concerts and sports finals like The Super Bowl or The World Cup. It offers a wider view than trend forecasts alone as it takes external events into account.

Most businesses are affected by seasonal changes in one way or another. A clothes store might sell warmer clothes in cold weather months and more swimwear in the summer. A sports store might sell replica soccer balls or boots during The World Cup and a cosmetic store might sell more of the eyeliner worn by Taylor Swift during her sell out stadium tour.


Qualitative data can help to create more accurate forecasts. One-to-one interviews, focus groups, social media conversations, case studies and reviews offer a more detailed analysis than some forecasting measures. They’re proactive measures and can take time to complete.

Be specific with your questions. Ask about favorite products, preferred buying times, individual features and anything that you think will help to improve the customer experience. This data is harder to analyze but if you measure it well, you’ll find insights that the other inventory forecasting types don’t offer.


Qualitative data asks why. It’s the motivation behind a buyer’s habits. Quantitative data asks what, how, where, when and who. What’s being bought, from which location, how often, at what times of year and by what type of customer? It’s easier to analyze this type of data, but if your business is small, you might struggle to get a big enough sample size to make a statistical impact.


Graphical forecasting isn’t really a forecasting type, it’s more of a communication tool. Graphs created to display trend forecasts make them much easier to read and analyze.

How to Choose the Best Inventory Forecasting Method For Your Business

You’ve learned from the forecasting types above that some data is easier to collect than others. For smaller businesses, it can be hard to get enough quantitative data to make an analysis so qualitative forecasting might be your starting point. If your retail or e-commerce business is more established, it’s easier to perform inventory analysis using historical data, so quantitative forecasting is a useful option.

If you have the resources and the right inventory planning software, the best forecasting combines each data type. Quantitative data is your starting point, qualitative data adds specificity. Trend forecasting and seasonal forecasting help you understand what’s going on inside and outside of your organization. One important note is that there will always be unexpected events, like a global pandemic, for example, so you need to be ready to adapt.

6 Inventory Forecasting Benefits

Good inventory forecasting creates multiple benefits so let’s see what they are:

1. Manage Costs

Inventory forecasting tells you what you need to buy and when so you’re never second guessing. You can take advantage of bulk buying discounts without overstocking, tying up cash flow and paying for holding fees.

2. Streamline Back-End Operations

The more you can automate, the more time you’ll have to spend on growing your business. An efficiently managed back end reduces the amount of data you need to manually input. With accurate forecasts, you can set up reorder points, update inventory logs and manage purchase orders using automated software.

💡Pro Tip: Use Singuli to automate your inventory planning. Automatically sync your expected receipts with the warehouse, keep track of your landed costs and share ETA dates with your freight forwarder with Purchase Orders & Shipments.

3. Keep Customers and Suppliers Satisfied

Customers are happy when they can buy with convenience. If inventory isn’t available, they have to place a backorder and wait for something they want or need. Many customers just won’t wait, they’ll find what they want elsewhere.

Suppliers don’t want to have to fulfill emergency orders. They have processes in place to make sure they’re serving all their customers well and difficult customers can make it harder to help customers who have forecasted well. Good inventory forecasting makes it easier to keep your customers and suppliers satisfied.

4. Reach Goals with Better Strategic Insights

Company goals are meant to be challenging, but good inventory forecasting can make them easier to reach. If everyone across the business understands the inventory forecast, you can make sure you have enough product to meet realistic sales targets, adjust inventory to match increased demand after marketing campaigns and cut costs. This all helps to meet financial, marketing and sales goals, which in turn help your overall strategy.

💡Pro Tip: Take advantage of Singuli’s detailed Reporting & Analytics to create a forecast that meets the strategic needs of your business.

5. Avoid Stockouts

Stockouts are bad for business. You lose potential sales and worse, you could lose customers for life. Good inventory forecasting lets you better plan your inventory so stockouts never happen.

6. Improve Cash Flow

If you’re overstocked, your cash flow is tied up in unsold inventory. You can’t use that cash until the stock is sold so you miss out on marketing opportunities, investment opportunities and the possibility of buying the right kind of stock. Plus, you’ll incur holding fees and extra costs of inventory.

📌Get Started: Optimize your inventory levels and reduce overstocks. Use Singuli to replenish the right amount of stock at the right time so your inventory costs stay low.

16 Steps to Inventory Forecasting Success

1. Determine the Forecast Period

30 days, 60 days, 90 days, a year, it’s up to you. Decide the period you want to forecast before you do anything else.

2. Work Out the Base Demand

What pre-sales have been made? Use these with an accurate pipeline based on past sales of the same or similar products to work out your base demand.

Look back at historical data to see the trends and variables that could affect your sales demand. Think about promotions and marketing activity that could have affected the base demand in those scenarios and use it to forecast future demand.

4. Calculate the Sales Velocity

Remember sales velocity is based on the products you have in stock, it’s not the average sales. This is important for an accurate forecast.

5. Consider Upcoming Marketing Campaigns

You’ll already have marketing plans worked out so make sure you understand your figures. What are you expecting to bring in with these campaigns and how much stock will you need to cover the difference in predicted sales?

6. Review Industry Changes

What’s new? Are there new competitors on the market? Has something political changed the perception of what you sell? Are there cheaper products coming onto the market? Keep on top of industry changes and add them into your calculations.

7. Break Down Seasonality By Product

Every product will have its own cycle. Assess the seasonal changes for each one. You can do this by spreading the data over 12 months, calculating a seasonal index for each month or using more complex statistical methods. The best forecasting software will calculate this for you.

💡Pro Tip: Avoid unexpected changes in demand and automatically account for seasonality in your inventory forecasts with Singuli’s Advanced Forecasting tools.

8. Account For Unexpected Spikes

What happens when someone talks about your product positively on TV or a celebrity makes something you sell trendy? Even if you’re not directly involved in the marketing, you’ll have to deal with the consequences. It’s hard to predict these spikes so cross-functional communication is vital. Keep in touch with your marketing team, make sure you have people analyzing the market and be ready to adapt if you need to.

9. Create Models

You’ve gathered a lot of information so it’s time to put together statistical models. These will give you a better insight of what’s ahead.

10. Remove Anomalies

Check the data for random events that don’t seem to be connected to anything. Make sure they’re not aligned to a seasonal change or a trend and remove them from your data. Anomalies can completely skew your forecasts.

11. Set Parameters Where Necessary

To perform some data analysis, you’ll need to set parameters like how many days of data your using, which categories to include or exclude and what “weights” to assign. These parameters enforce certain assumptions to confine the data and can be set for relevance.

12. Load Data into Inventory Forecasting Software

You have the data so now it’s time to put it into your inventory forecasting software so your inputs can turn into useful outputs.

13. Plan for the Best and Worst Case

Your forecasting data will not return one specific point so use the outputs you have to plan for the best and worst case scenarios.

14. Validate the Model

Use alternative data and check back to make sure the model is accurate.

15. Adjust, Adjust Again

Changes happen in the world all the time. When changes occur that affect your data, you need to adjust your inputs.

16. Reforecast

It’s time to start everything again. Learn from any mistakes you made the first time and repeat the whole process.

How to Deal with Events, Trends and Unknown Variables

What happens when there’s a global pandemic? If there are strikes or extreme weather, what do you do? It’s much easier to plan for the aftermath of a celebrity using one of your products than it is to plan extreme and completely unpredictable events.

It’s important to have great supplier relationships so you can work with them on managing any upturn or downturn in stock. Use inventory forecasting software that can take new information and add it to its algorithm to predict new results.

We’d also suggest looking into past events. What happened before, during and after? What were the effects and how can you use that data to make the most of future possibilities?

📌Get Started: Every product has its own story. However your products are growing, whenever there’s a change, Singuli’s forecasting models will identify it and adapt accordingly.

Planning with Inventory Forecasting Boundaries

Forecasting for seasonal products, new products and replenishments can be tricky. How do you know how many new necklaces you’ll sell or what difference a season makes to buying habits? You set boundaries. This ensures you’re using reasonable and probable logic, instead of guessing and hoping for the best.

Seasonal Products

One method of planning for seasonality is to use the seasonal index formula. It compares variances in a season to its average. There are many ways to calculate the seasonal average but let’s explore a simple averages method:

  1. Arrange the data in periods — most businesses use a three-month or quarterly timeframe (see an example below)

  2. Work out the season totals and the season averages

  3. Calculate the grand average, which is the average of the season averages

  4. Use your data to work out the season index as a %

$$\text{Seasonal index (\%)} = 100 \cdot \frac{\text{seasonal average}}{\text{grand average}}$$

Q1 Q2 Q3 Q4
2019 $155,300 $452,000 $85,400 $927,000
2020 $175,800 $485,700 $73,200 $955,000
2021 $182,000 $475,600 $101,000 $985,000
2022 $225,000 $454,600 $103,000 $999,000
Season Totals $738,100 $1,867,900 $362,600 $3,866,000
Season Average $184,525 $466,975 $90,650 $966,500
Season Index 43% 109% 21% 226%
Seasonal Sales Analysis

📌Get Started: Seasonal trends change all the time so make sure you’re planning for them with Singuli’s Advanced Forecasting.

New Products

New product launches have no history so how do you work out the stock you need? Start by looking at similar data on existing products and qualitative data to tailor models and draw assumptions. The best software does a lot of this for you.

💡Pro Tip: Use your past sales and product attributes to create accurate initial inventory orders with Singuli’s new product forecasts.

Manual vs Automatic Planning and Replenishment

Strong forecasts give you the information you need to replenish your inventory at the right time. Choosing to place orders manually or to use an automated system can be a difficult decision. You want to make sure there aren’t miscalculations with automated systems, but manual replenishment has a lot of room for error too.

If you decide to replenish manually, you need to have a clear understanding of who takes responsibility and you need to have back-up procedures in place. If you decide to use automated replenishment, you need software you can trust.

📌Get Started: Use Singuli to make sure your inventory levels are just right. With reorders on autopilot, you’ll always purchase the right quantities to the SKU level and you’ll never miss out on a potential sale.

Inventory Forecasting Best Practices

Inventory forecasting changes depending on your business and product types but there are some best practices that apply to everyone:

Set Up a Cross Functional Team

Inventory forecasting is not down to just one person or even an inventory forecasting team. A good forecast helps every business department and the best forecasts are made with everyone’s inputs.

For example, let’s say you have a furniture company. The logistics staff can offer insights on challenges with storing some of the larger products in a way that makes them easy to access when a sale is made. Marketing can talk about upcoming campaigns and make you aware of the impact of similar historical campaigns. Customer services can alert you to complaints and positive reviews of the latest blue velvet armchair.

Regularly Review Inventory

It’s important to keep track of your inventory, even if it’s sometimes easier said than done. Shutting your whole store to focus on a count takes time and resources, which is why many businesses only do it once or twice per year. A better way is to use cycle counts.

Cycle counts allow you to check the inventory of a small group of products without shutting down your regular operations. The next cycle count will look at a different set of products and you’ll continue doing this regularly until you’ve counted everything. It makes it much easier to manage discrepancies and greatly increases inventory accuracy.

Update Data in Real Time

The more up to date your data is, the more accurate your forecasts will be. Changes in demand, supplier lead times, unexpected events and many other factors can affect your forecasts but if you keep your data current, you can react and adjust in real time.

Let’s go back to the furniture store. Imagine Beyonce had seen the blue velvet armchair one day and decided to record a short Tik Tok video. She sits in it and says “Finally, I can truly relax”. As soon as your team notices this, they can enter the data into your inventory forecasting software, ready for an uptick in sales.

📌Get Started: Never miss a potential sale. Use Singuli to send reorder alerts straight to your inbox so you’re always ready to meet customer demand.

Make Sure Data is Relevant

It’s important to keep data current but it’s just as important to make sure it’s relevant. Comb through your data or even better, find software that will do it for you. Check for duplicates, irrelevant or incorrect data, structural errors, outliers and missing data to get the best results.

What’s working well and what’s falling short? Evaluating sales trends is a good way to stay on top of your data and helps you be as precise and accurate as possible. Keep on top of your customers’ needs, the market direction, seasonal trends and product popularity.

Eventually, Beyonce will move onto a new love and the blue velvet chair will go out of fashion. The market is moving towards chaise lounges so it’s a good thing your furniture company has been adjusting their stock portfolio to match their customers’ needs.

Examine Lead Times

Supplier lead times will vary and are often adjusted for changes in circumstances so it’s important to keep track. If your furniture store’s new chaise lounge arrives a few days late, you could miss out on crucial sales and lose customers. Make sure you have provisions in place for when a supplier can’t meet your demand.

💡Pro Tip: Maintain optimum stock levels and adjust to sudden changes by using Singuli to manage multiple vendors and products with different lead times.

Work Out Safety Stock

Safety stock is a lifeline. It’s the difference between being able to serve customers and not. You need to make sure you have just enough inventory, down to the SKU level, to keep your customers happy without tying up cash flow and incurring holding costs with overstocks.

Use the Best Forecasting Software For Your Business

Everytime a new product is added to your website or sales floor, everytime a supplier changes their T’s and C’s, everytime someone makes a change that nobody else can see, your workload increases and your availability to concentrate on customer satisfaction and business growth decreases. With the right software, you’ll save time and increase accuracy so you can concentrate on the things that make your business thrive.

How to Choose the Best Inventory Forecasting Software

That leads to an obvious question: What is the best inventory forecasting software for your business and what features should you look out for?

Support for Current and Future Products

Both your current and your future products need to be considered when forecasting and you’ll want to make sure your software can forecast demand and inventory for replenishments and brand new items.

Ability to Differentiate Product SKUs

Size does sometimes matter and it certainly matters when it comes to forecasting. You need a forecasting software that tells you what you need down to the SKU level. Make sure any differentials including size and color are accounted for.

💡Pro Tip: Automatically compute the right size curve for every sized-based product in your assortment with Singuli’s size-based forecasting and planning tools.

Flexibility to Support Your Specific Business Needs

It’s important that the software you choose works for your business and you don’t have to change the way you operate. For example, can the system you’re considering handle purchase orders (POs) with split POs and shipments, or inventory at multiple locations? This is especially important when you work across several selling channels like wholesale, retail or direct-to-consumer (DTC).

💡Pro Tip: Use Singuli’s location-level forecasts and replenishment alerts to optimize your inventory. If you have a mixture of retail and e-commerce channels, Singuli bases your forecasts on distinct models to reflect the variations in inventory patterns.

Capacity to Handle Multiple Data Inputs

What other inputs does your software need to handle? You’ll be planning marketing campaigns, you’ll have suppliers with different lead times, you’ll need to think about holding costs and you’ll have other inputs that are specific to your business. Make sure your software can handle your data and knows how to use it to create the most accurate inventory forecasts.

Seamless Reaction to Events

Professional boxer, Joe Louis, once said “Everyone has a plan until they get hit”. Good inventory forecasting software will be ready for that hit. Events like a pandemic, war, or inflation aren’t easy to deal with for any business but the right software expects the unexpected. It’s agile, it’s prepared and it’s tailored to your needs.

Makes Automation Easy

You don’t need to do everything manually. The best software automates repetitive processes so you can concentrate on making your business succeed. Think about what you’d like to automate. Would you like to track landed costs down to the item level? Could you sync your freight forwarder data? Would you work better if seasonality was automatically considered where applicable?

📌Get Started: Track exactly who did what and when they did it with Singuli’s Purchase Orders & Shipments. Every change is linked to a user and logged automatically.

Offers Robust Analytics and Reports

Reporting and analytics are a key consideration when it comes to forecasting and therefore, when it comes to picking the right software for your business. The right data, displayed in an easy to read and work with manner makes a big difference to accuracy. Think about the reports that matter the most to you and make sure your software supports your needs.

📌Get Started: With Singuli’s Advanced Forecasts, it’s easy to analyze historical data, market trends and customer inputs, which saves you or your inventory planner time and increases forecast accuracy.

Compare Inventory Forecasting Software

You have the information you need to find the best inventory forecasting software for your business so now it’s time to compare. Discover the options available on the market and use each point above to see which one meets your needs.

Automate Your Inventory Forecasting with Singuli

Inventory forecasting is vital to running your business efficiently so make sure you’re set up for success. Work out the most important factors for your e-commerce and/or retail store, follow the best practices and automate your inventory forecasting with Singuli.

Get the All-in-One Inventory Planning Software

Forecast demand, issue and track POs, reorder on autopilot, and step up your reporting game across multiple channels and locations. Get in touch to see how Singuli can help you optimize your inventory.

Your Message is Sent

Thanks for reaching out. We'll be in touch soon!

Contact Us

We'd love to show you how inventory optimization could transform your business.

* Required
Who is Singuli
We’re a multidisciplinary team of engineers, ph.d. researchers and data scientists with decades of retail experience.
Benjamin Kelly, Ph.D
CEO & Co-Founder
Thierry Bertin-Mahieux, Ph.D
CTO & Co-Founder