• Written by Azhar Aliyar, Director

The sales dashboard that the category manager uses to track monthly targets is only as good as the data it draws from, and the same is true for the AI that answers questions about sales: it needs a set of specific systems and fields to produce useful answers. For example, a shift supervisor checks the maintenance log to see when equipment was last serviced, but in sales, the equivalent check is on the historical sales data at planning grain, which is used to forecast future sales. This data is typically stored in a database such as the purchase order table or the sales ledger.

the systems

The systems that a sales deployment of AsscherAi depends on include the historical sales data at planning grain, the order book and pipeline, inventory positions, and promotional calendar. The historical sales data provides a baseline for forecasting, while the order book and pipeline give a view of current and future sales. Inventory positions are necessary to understand what products are available to sell, and the promotional calendar outlines the marketing activities that will drive sales. A sales team, such as those described in our sales teams article, will typically have access to these systems and use them to inform their sales strategy.

Each of these systems contains specific fields that are necessary for the AI to produce useful answers. For example, the historical sales data should include fields such as date, product, and sales amount, while the order book and pipeline should include fields such as customer, order date, and expected delivery date. The inventory positions should include fields such as product, quantity, and location, and the promotional calendar should include fields such as promotion type, start date, and end date.

what a missing system does to an answer

If one of these systems is missing, the AI will not be able to produce a complete or accurate answer. For example, if the inventory positions are not available, the AI will not be able to determine whether a product is in stock or not, and therefore will not be able to answer questions about availability. Similarly, if the promotional calendar is not available, the AI will not be able to take into account the impact of marketing activities on sales, and therefore will not be able to produce accurate forecasts.

A missing system can also lead to incorrect assumptions being made by the AI. For example, if the historical sales data is not available, the AI may assume that sales are consistent over time, when in fact they may be subject to seasonal fluctuations. This can lead to inaccurate forecasts and poor decision-making.

the definitions that must be agreed

In order for the AI to produce useful answers, it is necessary for the definitions of key terms to be agreed upon. For example, what constitutes a sale, and how is it different from an order or a shipment. These definitions should be consistent across all systems and fields, and should be understood by all users of the AI. This can be a challenging task, as different teams and departments may have different understandings of these terms.

It is also necessary to define the granularity of the data, such as the level of detail at which sales are recorded. For example, are sales recorded at the product level, or at the level of individual components. This will affect the accuracy and usefulness of the answers produced by the AI.

quality problems that only surface later

Quality problems with the data can be difficult to detect, and may only surface later when the AI is in use. For example, if the historical sales data is incomplete or inaccurate, the AI may produce forecasts that are consistently wrong. Similarly, if the inventory positions are not up-to-date, the AI may produce answers that are based on incorrect assumptions about availability.

These quality problems can be caused by a variety of factors, such as poor data entry, inadequate data validation, or insufficient data cleaning. They can also be caused by changes to the systems or fields over time, such as the introduction of new products or the reorganization of the sales team. It is therefore necessary to have processes in place to monitor and maintain the quality of the data, and to update the AI as necessary to reflect changes to the systems and fields.

when it is the wrong choice

AsscherAi is not the right choice for every sales team or organization. If you are looking for a simple way to automate routine sales tasks, such as data entry or lead qualification, then AsscherAi may not be the best option. However, if you are looking for a way to gain insights into your sales data and make better decisions, then AsscherAi may be a good fit. If you are unsure whether AsscherAi is right for you, please contact us to discuss your needs and determine whether our platform can help.