Static reports are not sufficient

Information systems such as SAP ERP, BI, BW, etc. provide employees with data that is linked to relevant and analyzable information through defined KPIs.

They help us to channel the daily flood of information and thus make decisions based on facts instead of emotions. Another advantage is that they are later comprehensible to third parties. Each level of the company needs specific information.

When we look at the structure of a company, we see that different levels of the company need different information.

At a higher management level, target or plan data is compared with actual or historical data in aggregations. These information requirements are usually covered by classic BI implementations.

At the operational level, on the other hand, there is often a whole zoo of Excel files, Access databases or specific reports. Both the classic BI implementations and the mosaic of partial solutions at the operational level usually measure past performance.

This is no longer adequate for today's fast-moving times, when we need a forecast of future events rather than a look in the rearview mirror.

If we take a closer look at the information needs of middle management or those at the operational level, we find that these tend to run horizontally across the entire value chain.

They are also extremely variable and often located at the lowest granular level - for example, at the item or even schedule line level in the SAP system.

These complex information requirements are rarely covered by today's systems, and if they are, then only with extremely high effort. The difficulty here lies, among other things, in the complexity of the supply chain in conjunction with constantly changing requirements.

Static reports are not sufficient, because no sooner is a piece of information determined than the next question arises from it. Therefore, such an information system must be extremely fast and flexible.

The devil is in the details

Analyses on historical data allow a look back. Near-real-time data at the lowest granular level additionally enables proactive ad-hoc decisions that also incorporate data horizontally across the value chain.

However, in order to exploit the full potential of the data, it must be intelligently linked so that it can subsequently be used for predictions.

An example: Every day there are new unforeseen events in the company, such as delayed deliveries due to the next rail strike. The question is whether and what impact the delayed delivery will have on the value chain (manufacturing, sales).

To do this, you have to be able to determine the corresponding bottlenecks directly, for example, on the basis of the shipping notification and the purchase order number; be it production orders that rely on the delayed material, sales orders or similar. And this takes into account the possibly multi-level supply chain with all the intercompany processes.

Correlation is not enough, causality is required

The above example clearly shows that looking in the rear-view mirror is no longer enough these days. However, this also means that all documents must be brought into a causal relationship with each other at the lowest granular level (in SAP: schedule line level).

It may still be possible to predict an event by correlation, but at the latest when analyzing the causes and predicting the impact across the entire value chain, you need the causal relationship.

For end-to-end process analysis of this complexity and with this volume of data, an "in-memory" snapshot is essential.

IT often does not have the capacity to quickly meet the information needs of the business units. There is often not the corresponding know-how or even the authorization to obtain this information from the systems.

And even if they do, 90 percent of the time is then spent on information gathering and only ten percent on problem solving.

What is needed, then, is true self-service for the business department, where a user needs only ten percent of the time to gather information and can spend 90 percent of his time on solving the problem. Two things are necessary for this:

1. that such an information system pre-creates classic KPIs (such as service quality, delivery reliability, OTIF) as well as hundreds of integrated, calculated calculations (e.g. the demand in/for the last week/month/quarter or the value of overstock) on an up-to-date data basis.

2. that business users can quickly and easily customize these analyses and perform complex calculations across the entire value chain (e.g., the percentage of assigned or dependent orders).

Conclusion

Operational business analyses are aimed at business departments that want to use ad hoc analyses to reduce inventories, for example, and at the same time increase the service level to the end customer.

The name reveals that the operational level receives decision-making bases on the basis of which it can immediately intervene in any process. Bottlenecks in the value chain are identified before they arise in practice, paralyze processes or cause delays in deliveries.

Near-real-time analytics empower employees to make decisions quickly at the most granular level within their sphere of influence and prevent process bottlenecks.

This makes it easy to search for a needle in a haystack: Among thousands of data records, employees can find the data records relevant to their area of responsibility in a matter of seconds.

Every Angle offers a plug-and-play solution for customers using SAP. Because the solution interprets the customizing of the SAP system, no modeling is required and the system is ready for use in two days.