The system algorithm is designed to calculate baseline sales (expected without promotion campaigns) and effects (sales growth) for both past and upcoming promotions. The algorithm operates across different sections (SKUs, product types, geography, format). Input parameters are sales and balances history, period for baseline calculation, history of all past promotions, sales days to be excluded from the calculation (holidays, events, or events that resulted in a significant change in demand), seasonal factors and overlapping of similar items in promotions. As an output, the system provides the user with data on the actual/planned change in demand due to the promotion campaign.
The system features tools that allow analyzing automatic calculations of promotion campaigns and based on the factors that influenced the recommended effect, evaluating effects and inputting final values to be used later in the analysis of actual sales. It is possible to estimate different types of effect: absolute (total volume of sell-out during the promotion period), additive (sales growth during the promotion period, in pieces / kilograms / liters (decaliters), multiplicative (sales growth during the promotion period, as %). The system also enables to flexibly evaluate the actual effects of completed campaigns to be used in the future as analogs of the planned ones: changes over time in and growth of sales during promotional campaigns, occurrence and magnitude of shortages, the distribution of sales by locations. If the effect of a promotion campaign has been significantly influenced by external factors, it can be checked and it will not be used as the most similar peer during future promotions of similar products in similar periods.
In the system interface, users can manage all the parameters for calculating promotion campaign effects: base levels, selection of similar campaigns and forecasting of growths. Customization control is available both on a point-by-point basis (for individual SKUs, locations and mechanisms) and via flexible group operations (defining single calculation parameters for multiple combinations).
The system features a convenient campaign log and a form for managing the composition and parameters of the campaign, allowing to automate the process of planning, filling and coordinating promotions both for individual categories of goods and locations and in general for the entire campaign while the advanced logging of actions in the system ensures the maximum transparency of processes.
The system enables to forecast the effects of promotions for SKUs being introduced into the product mix as well as for stores that have no sales history and were opened recently or to be opened during the promotion. Users of the system assign new products and new stores to analogs that have sales history to be used by the system in calculations for the period designated by users.
The system enables users to assess the accuracy, economic parameters and shortages that occurred in stores during the promotional period using reports that can be generated both for the entire promotion campaign and by SKUs and locations. Also the need to extend the promotion to reduce excess inventory is assessed based on the profile of regular sales and product balance.
Predicting the effects of a planned campaign based on “similar” campaigns in the past. Analyzing and evaluating the effects of planned and actual campaigns. Flexible management of calculation settings.
Automating and streamlining the process of planning, stocking and approving promotions. Flexible customization for business processes. Mechanism for approval statuses and stages.
Predicting the effects of promotions for SKUs being introduced into the product mix as well as for stores that have no sales history through the designation of peers.
The module enables to connect and use external data such as competitors' promotion campaigns, FDO data, weather data, etc. for more accurate forecasting. Along with the use of data on the history of own promotional activities, onboarding data from open sources enables to generate more accurate forecasts of the effect of planned promotions.