Key takeaways

• Accurate forecasting helps you keep stock and cash flow in balance.This article outlines 8 proven ways to improve forecasting in e-commerce, such as using multiple forecasting models, accounting for seasonality, and reacting fast to demand shifts.
• Spreadsheets can’t keep up as your business grows. They’re too manual to scale, can’t handle seasonality or sudden demand changes, and even small errors can lead to costly mistakes.
• Optiply automates forecasting up to 100%, so you can focus on growth instead of manual work.

What are the real costs of inaccurate forecasts in e-commerce

When forecasts are off, the effects add up quickly. If a popular product runs out, one missing item can cost you the entire order. Overestimate demand, and your cash gets stuck in slow-moving stock. Many retailers face this challenge, but Fitwinkel turned it around, lowering inventory levels by 30% and reducing lost sales by 80%.

The actual costs of inaccurate forecasting are hard to determine without a real business case. That’s why we came up with a concrete example where we show the difference in costs using Excel and a modern inventory optimisation software solution such as Optiply. 

‍

Want to know exactly what you can save on your inventory with optimised forecasting? Use the ROI calculator and you'll find out in less than a minute.

What are the risks of using spreadsheets for e-commerce forecasting 

Spreadsheets might work in the early days, but they quickly start to slow you down as your business grows. Here’s why spreadsheets fall short when it comes to forecasting:

Spreadsheets are too manual to scale

When your assortment grows, so does the time spent fixing data. Correcting stock levels, adjusting supplier lead times, it all adds up. This means that over time, it almost becomes impossible to keep your data accurate. 

No grip on seasonality or demand shifts 

Spreadsheets show what happened, not what’s coming. They can’t account for seasonal peaks or sudden changes in demand. This often means missed sales or cash tied up in stock you don’t need. 

Small errors, big consequences

One wrong cell or outdated formula can throw off your entire forecast. What looks like a small error in Excel can lead to stockouts and overstock. This creates unnecessary costs. 

Now you know exactly why forecasts can always be optimised, let’s have a look at how you can actually achieve it.

8 proven ways to improve forecasting accuracy in e-commerce

Once you move beyond spreadsheets, you can start to improve forecasting in ways that simply aren’t possible with manual work. These eight methods show how automation helps you plan and stay in control as your business grows.

1. Use multiple forecasting models

Forecasting models use real-time data to create accurate demand predictions. By combining multiple models, each product gets a forecast that truly reflects customer demand. Optiply does this by feeding its models with data on sales, ABC categorisation, seasons, trends, peaks, and actual delivery times.

2. Factor in seasonality and trends

Customer demand changes throughout the year. Recognizing seasonal peaks and long-term trends helps you order in time and avoid running out when demand is high. For instance, DealDonkey was already using Picqer (WMS), and after integrating it with inventory optimisation software from Optiply, purchasing became faster and more accurate. 

Inventory planning and purchasing is now automated based on real-time data, helping the team plan ahead for busy periods like Christmas.

3. Automate safety stock calculations

Safety stock is the extra inventory that helps you handle sudden changes in demand or supplier delays. Calculating it manually takes time and easily leads to mistakes. Automating it ensures every product has the right buffer, enough to prevent stockouts without overstocking. Smart inventory optimisation tools, such as Optiply, apply a safety stock formula automatically, using your sales data and lead times to keep stock levels balanced.Martijn from Fitwinkel explains how Optiply takes over calculations and analysis, saving his team hours of manual work:

“The fact that Optiply makes the calculations, analyses the sales of recent years and can therefore make a prediction – that saves so much work for me. That time-saving is really the trigger. And it is also a lot less error-prone than our old way of working.”

4. Flag and correct outliers automatically

An outlier is a data point that doesn’t follow the usual pattern, like a sudden sales spike or a one-off promotion. Even a few of these outliers can throw your forecasts off and make future predictions less reliable.

With automated outlier detection, your system can:

• Flag unusual data points (outliers)
• Adjust forecasts automatically to keep them accurate
• Save your team hours of manual checks and corrections

This way, your forecasts stay reliable, even when your data shows unexpected changes.

5. Integrate supplier performance data

Forecasts don’t just depend on demand but also on how well your suppliers deliver. Some demand planning or inventory optimisation tools work with supplier data. This way, you can ensure that buyers stay in control by knowing which suppliers deliver on time and which ones might cause delays. Optiply, for example, includes data on lead times, delivery reliability, and minimum order quantities. 

6. React fast to demand shifts

Customer demand can change overnight; a viral post, sudden weather change, or promotions can quickly impact sales. Automated forecasting updates your data in real time, helping you respond before stock runs out or piles up. This keeps your stock aligned with what customers actually buy.

7. Free up buyers for strategic work

When repetitive tasks like updating forecasts or creating purchase orders run automatically, your team can shift their focus to work that really drives value, such as:

• Strengthening supplier relationships
• Improving product and pricing strategy
• Identifying new opportunities for growth

Automation doesn’t replace buyers. It gives them the time and insights to make better decisions and work smarter, as Koopjesdrogisterij’s team discovered. Their purchasing team became 94% automated, with repetitive work reduced to a minimum.

8. Use AI-driven demand planning 

AI-driven demand planning takes forecasting one step further. It doesn’t just predict demand, it learns from new data every day and adjusts purchasing advice automatically. This way, your forecasts remain accurate, even when conditions change. It helps you order the right products at the right time without doing manual work. 

Frequently asked questions about forecasting accuracy 

How do you measure forecasting accuracy?

Forecast accuracy shows how close your forecasted demand is to the actual sales. It’s usually expressed as a percentage and calculated with the formula:

Forecast accuracy = 100% − (Forecast − Actual demand / Actual demand × 100 %)

The higher the percentage, the closer your forecast matches reality. Tracking this helps you understand how well your demand planning is functioning and where adjustments are needed.

What is considered a good forecasting accuracy in e-commerce?

In e-commerce, a forecast accuracy of 90% or higher is generally considered good, depending on the sector. Products with stable demand are easier to predict, while new or seasonal items tend to fluctuate more. What matters most is building consistent, reliable forecasts that keep your stock and cash flow in balance. 

How does AI improve forecasting accuracy?

AI improves forecasting accuracy because it learns from patterns in your data and adapts as conditions change. Unlike spreadsheets, it combines information from sales, suppliers, and seasonality with multiple forecasting models. This helps it make smarter predictions that stay accurate over time.

What should you look for in a tool to improve demand forecasting?

Look for a tool that makes your forecasting process easier and more effective, and that:

• Automates purchasing decisions, helping your business move toward 100% buying automation.
• Integrates with your existing systems, such as your ERP, marketplaces, and e-commerce platforms, bringing all data together for faster and smarter purchasing decisions
• Updates forecasts automatically without the need for manual input
• Refreshes forecasts continuously and adapts to real-time and unexpected events 

How Optiply makes accurate forecasting effortless 

Optiply was founded with a simple question that every buyer faces: what, when, and how much should I order? As former e-commerce entrepreneurs, the founders experienced first-hand how difficult it is to keep stock balanced when relying on spreadsheets. Every change in demand, every delay from a supplier, and every unexpected event adds complexity that’s nearly impossible to manage manually.

With these insights, Optiply was built to make purchasing simple and move businesses toward 100% automated replenishment. The AI-powered platform connects external data from sales, suppliers, and seasonality to forecast demand accurately and support smarter purchasing decisions.

The results? Optimal stock levels, a healthy cash flow, and more time for teams to focus on strategy and growth. Today, more than 500 e-commerce, wholesale, and retail companies use Optiply to achieve accurate forecasting and fully automate their replenishment process, without relying on spreadsheets.

Ready to see it in action? Schedule a free demo and discover the difference Optiply can make.