Sales predictions

What are estimations in the Nethansa Platform?

Our analytics system uses advanced machine learning techniques to process and analyze large data sets. After careful examination of the data, the model is trained and tested to provide accurate predictions. The results are then evaluated using rigorous statistical methods, increasing their reliability. The entire process is characterized by a high degree of complexity and accuracy, so we are able to provide extremely precise predictions. When selecting the appropriate model, we take into account key parameters in our evaluation, such as the length of time the offer has been available on the market, sales history or the number of competitors. All forecasts cover the next 30 days.

When using estimation data, it should be remembered that they are made for a given moment, in which the current situation of the offer is taken into account (price in relation to the competition, number of competitors, inventory, sales account balance, etc.). If there is a significant change in any of the parameters during the period for which the forecast was generated, the accuracy of the forecast may change.

The forecasting process produces two sales values in terms of volume: prediction and sales potential. Prediction is sales potential limited to available inventory. Potential is the projected sales volume with unlimited inventory and maintaining the current parameters of the offer and the surrounding market conditions (price relative to competitors, number of competitors, seller's account statistics, etc.).

Sales forecasts are generated per offer, and the trigger for their calculation is an event of a change in the offer parameters, such as a change in price, a change in sales ranking (BSR) or currency exchange rate. It can take up to several hours from the time one of the parameters changes to the generation of a new forecast, as this is a process that requires the collection of a lot of additional information. It usually takes about 1h.

How to use sales potential?

Sales potential is used to determine the quantity range an offer can sell. It allows you to determine the likelihood of sales or the sales appeal of individual offers. Potential can be used:

• To select products worth restocking with a limited purchasing budget - in the Shopping basket function.
• For prioritizing offers with a "Ready to send" status.
• For verifying which products or listings need to be authorized on specific Amazon marketplaces to get the best sales results.

Where can I find information about sales potential?

Sales potential is, by default, a hidden column in the Offer tab. You can add it by clicking on "Columns" and then selecting the Sales Potential column. Use the search box to find the required column more easily.

After closing the add column window, the newly added column will appear on the right side of the offer grid. If this column is used often, it is worth moving it closer to the left edge of the grid and saving the view so that you can quickly return to it when logging in again.

The screenshot below illustrates an example of the use of sorted offers according to their sales potential with the "Ready to send" offer status filter overlaid. When sending offers that are ready to send or when managing offers, it is a good idea to focus on those with the highest potential because these are the offers that will contribute to the greatest increase in sales.