They already had SAP. The operation was still running on spreadsheets.

The manufacturer had a structured ERP environment built on SAP, with several other systems and data sources feeding the operation. Most of the information already existed, somewhere.

Forecasting was the problem. Nobody had a dependable view of raw materials coming in, sales going out, or goods sitting between the two. Planning happened without it.

So the team built their own tool: a lightweight forecasting and planning sheet in Google Sheets.

To be fair to the sheet, it did what it was meant to do. It filled a gap SAP was not filling, and it gave the planners something to work from.

It also asked for hours back every week: collecting figures from different systems, pasting them in, rechecking formulas and references after every change, and keeping planning data in step with everything else. The numbers still had to be read by hand before they meant anything.

The tool that was supposed to make planning easier had turned into a process of its own - one that needed attention whether or not anything had changed.

We came in as an operations technology consulting and systems engineering engagement, not a software pitch. The ERP was producing value. It was not producing answers.

So we asked a narrower question: where does operational data have to travel before a person can use it to make a decision today?

Two systems came out of that. An AI-powered operational knowledge and intelligence platform, and an AI-powered transportation and stock visibility platform. One works on understanding the data. The other follows the goods.

The first system connects to the company's ERP and other operational sources, and keeps a live knowledge layer across all of them. Nobody has to know which system holds which number before they can start.

They ask in plain language. The platform reads from the connected data, answers the question, and any answer can be saved as a dashboard the team reuses next week.

The second gap was physical. Inventory was not only sitting on shelves. It was loaded on trucks, moving between sites and customers, and the operational picture stopped at the warehouse door.

A driver could be two hours down the road with a full load while planning still looked at yesterday's figures. That is not a reporting problem. It is a missing system.

So we built a transportation visibility application, with a mobile app installed on the drivers' devices. Drivers feed it from the road. The operations team sees the result on one map.

The knowledge platform tells the company what its systems know. The transportation app tells it where the physical stock is. On their own, each is half a picture.

Together they close the distance between what the records say and what is happening on the ground - and that distance is where most planning arguments come from.