Combination forecasts
The primary forecast of this subteam is the mixture degree. gross sales forecast. With this challenge, we forecast gross sales for the subsequent X weeks, each on a weekly and day by day foundation. To offer some context on aggregation, one doable degree of aggregation might be gross sales for the corporate as an entire. Such a forecast may also help make business-level selections and work to set objectives and expectations. One other doable degree could be gross sales coming by bowl warehouses, that are essential to operations and workforce allocation.
An essential frequent characteristic of most of our crew’s aggregate-level forecasts is that additionally they rely upon the gross sales forecast (making them later forecasts), since gross sales is usually the first driver of many different metrics we forecast.
This brings us to a different essential prediction, which is the customer support interplay forecast. With this challenge, we’re offering an estimate of what number of interactions our customer support brokers can count on within the coming weeks. This forecast is essential for the enterprise, as we don’t wish to make too many forecasts, which might result in overstaffing in customer support. Alternatively, we additionally don’t wish to underestimate the forecasts, since this could trigger lengthy ready occasions for our clients.
To make sure that our providers (webstore, app) scale nicely through the peak interval (November and December), we additionally supply a request forecastthat’s, what number of requests providers can count on throughout peak intervals.
Lastly, we provide a sequence of forecasts associated to logistics. Bol has a number of warehouses the place we retailer each our personal objects and the objects of our companions who wish to use bol’s logistics capabilities to maintain their enterprise operating easily. As such, we provide just a few totally different forecasts associated to logistics.
The primary is outbound logistics forecastsThat’s, a forecast that signifies what number of objects will depart our warehouses within the coming weeks. Likewise, we offer a incoming forecastwhich focuses on the objects that arrive at our warehouses. Moreover, we additionally present a extra specialised inbound forecast that additional breaks down incoming objects based mostly on the kind of bundle they arrive in (for instance, a pallet vs. a field). That is essential as these several types of packages are processed at totally different stations inside the warehouses and we have to guarantee they’re appropriately staffed.
Merchandise-level forecasts
The second subteam focuses on item-level forecasts. Bol provides round 36 million distinctive objects on the platform and for many of them, we have to present demand forecasts. These predictions are used for storage functions. On this method, we attempt to anticipate the wants of our clients and order the objects they might want nicely upfront to ship them as quickly as doable.
Moreover, the crew supplies a devoted forecast that may deal with newly launched objects and pre-orders. With this forecast, stakeholders can anticipate what number of objects can be offered sooner or later earlier than the launch and inside the month following the launch. This fashion we will make sure that we have now sufficient copies of FIFA or Stephen King’s newest novel.
Lastly, our crew additionally developed a promotional enhance forecastwhich helps consider the rise in gross sales of a given merchandise based mostly on the value low cost and the length of the promotion. Our specialists use this forecast to make higher data-driven selections when designing promotions.
