Algorithms are everywhere in our lives – whether you are using social media, checking train times or doing a Google search. Algorithms ensure that we have information available that matches our preferences. They make our lives more pleasant, more efficient and safer. Algorithms also ensure that we can automate certain steps within a process. The same applies to Optiply’s purchasing algorithm – which can automate parts of the purchasing process. In this blog, we explain how Optiply’s purchasing algorithm works, what makes it unique and how it helps e-commerce businesses. In addition, we explain the direction that the algorithm is developing in.

How does the purchasing algorithm of Optiply work?

The purchasing process consists of three layers: strategic, tactical and operational. Within the process, each layer manages specific purchasing tasks. These tasks can – without the help of automation – take a lot of time. Optiply’s purchasing algorithm is one of solutions to make your purchasing process less time-consuming and more efficient. The purchasing algorithm can be divided into the following three phases:

Optiply purchasing algorithm

1. Data – collect and clean

The algorithm runs on historical data. These are figures that provide insight into – for example – the number of sales, purchasing costs, delivery times and/or transport costs. In the first phase, the numbers are collected and cleaned for analysis.

2. Forecasting

The core of Optiply’s algorithm is formed by data-driven forecasting based on 40 different models. Future purchasing is forecasted on the basis of historical demand data (sales), seasonal influences and trends. However, supplier data such as delivery times, purchasing costs and delivery reliability are also included in the forecast.

3. Purchasing advice

In the final phase, the data and forecasts are converted into a purchasing advice. That means how many products you have to order, when is the best time to order them and from which supplier(s).

Do you want to know how your e-commerce company can operate flexible and scalable in a changed supply chain? Discover 6 tips for making purchasing decisions
and how forecasting can help.

What makes the purchasing algorithm of Optiply unique?

With virtually every type of purchasing optimisation software and/or smart Excel solution you can automate the operational tasks within your purchasing process. Automation enables you as a purchaser to focus on the tactical tasks – where human insight can make the difference – within the purchasing process. Optiply’s purchasing algorithm goes one step further and is unique because of the following components:


The fact that up to 40 forecasting models are used to recognize patterns is unique. The algorithm can combine these models and tailor them to specific situations. Because the algorithm has already made innumerable forecasts for almost every conceivable situation, it can also make predictions for products with a very unpredictable demand, such as luxury, non-seasonal items (e.g. luxury watches).

Purchasing advice

Where other software and/or Excel solutions are highly dependent on the ‘quality’ of the user, Optiply’s algorithm is unique in its level of detail and ensures that you have accurate insight into what you have to purchase, the quantities and from which supplier(s). This ensures that you receive purchasing advice per product line.

How does the purchasing algorithm help companies?

Optiply’s purchasing algorithm ensures that you can get a grip on a number of crucial matters. These have more impact than just making your purchasing more efficient.


Automation ensures that fewer people are needed to carry out the purchasing tasks. And, the purchasers, who are working can carry out their work efficiently. This can save your staff on average 80% time spent on purchasing tasks and saves your business overhead costs.


Due to the accurate purchasing advice you can bring your stock level down by – on average – 26%. You can keep your stock at the right level and reduce the chance of obsolete stock, which can lead to higher holding costs and write offs.

Customer satisfaction

When your stock is at the right level, it will also lead to less lost-sales. It means that you can meet customer demand and that will have a positive effect on customer satisfaction.

The impact of these three things is that it will improve your reputation and reliability – both internally and externally. Lower overhead and inventory holding costs and rising customer satisfaction will lead to a higher profit, which you can then use to reinvest in your company.

Do you want to know how your e-commerce company can operate flexible and scalable in a changed supply chain? Discover 6 tips for making purchasing decisions
and how forecasting can help.

Threats for the purchasing algorithm

The accuracy of Optiply’s algorithm also depends on a number of conditions. Threats to purchasing advice (the output) can occur in the 1st and 3rd phase.

Lack of quality data

The better your data is (input), the more accurate the purchasing advice will be (output). If there is little high-quality data available within the WMS, ERP or CMS system of your company, this will make the outcome less accurate. The collection and cleaning of data is therefore essential for the functioning of the algorithm and the outcome.

Human insight

If you do not let human influence play a role in the final stage of the process, you can overlook or misinterpret exceptions in the data. You may have to deal with outliers within a dataset and human insight will be required to correctly interpret these extreme values (high and/or low). The same applies if a supplier underperforms, but has a evident reason for this. You need human insight to give feeling to the data – and the purchasing advice.

How will the purchasing algorithm develop?

The purchasing algorithm is constantly evolving due to the increase of available data. It will become smarter over time and solve problems that were previously performed by humans – due to Artificial Intelligence (AI). The algorithm increases in power due to the increase of high-quality data and is able to recognize more patterns and process combinations. This makes the algorithms ‘predictions’ and purchasing recommendations increasingly accurate.

Already on the basis of two weeks of historical data, the AI purchasing algorithm of Optiply can give purchasing advice and thereby improve your inventory management and automate the purchasing process.


Also innovate with Optiply?

Do you want to know how your shop can apply innovative inventory optimisation? Book your demo now. Within 30 minutes you will know how AI and forecasting models play a role in an innovative way by creating more revenue and spending less time on purchasing.