Supply Chain Analytics | Is it doomed to be a bridge to nowhere?

?

Medical science defines OCD as an often-repeated pattern, which ends up badly. So, when we see the same in corporate world, does it qualify as a "Corporate OCD"?

Let’s take an example, which has tried & troubled so many organizations.

?

Usually it starts something along these lines->

We are going to modernize our analytics and AI applications, so IT will extract raw tables data from ERP/ CRM / SCM / eCommerce / DB’s etc into the new cloud stores.

Then they will reverse engineer all the table relationships & rewrite the business logic (which a vendor like SAP / Oracle etc took decades to build & also, was fine-tuned by our business users over many years).

BTW, we will go live in 6 months’ time with our new analytics.

& Sure, to get started there are a variety of data extractors available to choose from, which gets the project off the blocks.

Sounds familiar so far?

However, soon realization kicks in (a typical SAP ERP system has >450,000 tables, S4HANA has >800,000 tables). They also have views of all kinds (DB views / CDS et al)+ reports (>90,000 of them). Most of them with German field names & documentation not exactly designed for reverse engineering.

Soon 6 months starts to spill to 9 months , to 1 year & so on... & neither reporting nor analytics is working. Some reports that do get delivered are overwhelming reams of information in a tightly controlled front-end and based of copies of data, which makes it even more difficult to rely on it for any decision making.

Wash. Rinse & repeat. ?Seen it being tried at so many places that have lost count by now, but then a new one comes who wants to try all over again.

To put it mildly: THAT DOES NOT WORK.

So, what are the successful organizations doing?

1.????They flip the analytics from an IT project, to being central to how business leaders & operators make decisions. Sure tech is needed, but they start with the business persona and what information/ insights are needed for that role to take the actions needed, in-time.

An insight in-time, saves nine tables of information 
(& incriminations) later.        

For example, in Inventory Controlling: they leverage SAP modeled reports / extractors for real-time inventory data, including material stock levels, stock values, and movement information from the SAP ERP / Warehouse Management System (WMS). It provides up-to-date visibility into inventory status, helping supply chain managers optimize stock levels and ensure timely order fulfillment (the underlying BW Extractor is: 0IC_C03) and can be provisioned for real time decisions.

2.????To streamline their operations, they first streamline the information flows. What does that mean?

They establish a very lean, time-sensitive flow of the events->insights->actions & continuously improve it by cutting any lags / friction (especially for supply chain).

Unsurprisingly lot of times they are applying Lean supply chain principles 
to their information supply chain, with good results.        

Hint: brittle data pipelines for copying over data don’t help. What does work is solid integration with real-time, reliable and relevant information. One manufacturing companies uses KE24 report directly in their PowerBI for reporting daily invoiced sales, as that is a trusted source everyone can rely upon.

?

3.????Enable Pull mode for information & insights: If 40+ years of mass-produced, feature driven analytics has proven one thing, it is: the Business does not like it. ?They are looking for the information & insights in a self-service tool, which allows them to analyze, enrich with their own datasets, in time & with reliable information. ?Once data is provisioned to the user, AI generated visualizations & insights accelerate this journey, with generative AI producing insights on-demand and within context, enabling the business user to self-service is the most impactful role IT can play.

?

?