SAP, Master Data and AI: A Practical Guide to Modernising Enterprise IT

title: SAP, Master Data and AI: A Real-World Approach to Modernising Enterprise IT

description: A no-nonsense guide to SAP support, management of master data, AI readiness, and software development for growing businesses.

Most businesses aim to transform digitally, with a focus on AI readiness. In reality, most enterprise IT issues stem from poor data management and insufficient support. Lack of clarity on priorities creates its own set of problems.

In my 25+ years of experience in enterprise IT, I've learned the most important lesson: if the foundations aren't established properly, everything built on top becomes problematic.

The following outlines the most important lessons learned in an easy-to-understand format.

1. Support your ERP systems: SAP and Oracle

Your ERP systems (SAP or Oracle E-Business Suite or similar) are the backbone of your finance, supply chain, and operations. Poor support can cause issues such as wasted reporting time, workarounds, and a lack of faith in the system among end users.

An effective support system usually includes:

• A clear plan on how your system will be

• Support staff that are onsite or nearby to quickly resolve issues

• Continuous improvements and upgrades to the system

• A support system that is flexible and can scale with the business

Organisations also use a mix of onshore consultants and offshore support centres to keep costs reasonable while still providing 24/7 support. Infoplus Technologies UK uses this model, having SAP-certified support staff in offshore centres and leadership staff in the UK.

1. Treat master data as an asset.

Master data refers to essential business data such as materials, suppliers, customers, and assets. Data duplication, inconsistency, and incompleteness present risks at multiple levels of a business. For instance, the reports and automation tools built on master data carry the same defects.

This is particularly problematic in asset-heavy industries such as manufacturing, utilities and energy, where MRO (maintenance, repair and operations) master data impacts purchase and maintenance decisions. Good data practice includes:

• Data cleansing

• Standardised cataloguing

• Data governance

• Inspection of plant and equipment to ensure records are up to date.

Better data helps eliminate stockouts, duplicative purchasing, and improves reporting.

1. Build custom applications only when required.

There are many packaged applications, and it is worth evaluating off-the-shelf applications for the processes you want to develop a custom application for. Custom application development is required when:

-There are no packaged applications to suit your business processes

• Integrating multiple applications is the real problem

-You want to differentiate your business from competitors

Custom applications should be small in scope. Testing and documentation should occur early and continuously and support for the application should be planned for. Outsourcing product development and testing can help deliver applications faster.

3. Get AI-ready (without the hype)

AIs are only as powerful as the data and systems that back them. Before investing in AI, you should consider:

1. Is our data accurate and structured?

2. Are our core systems integrated?

3. What specific problem are we solving with AI?

If the answer to the first two is “no”, then the best AI investment is to improve the basics from sections 1 and 2.