Skip to content
AnrilX

Procurement

Which suppliers quietly became more expensive

Price drift does not announce itself. It arrives one purchase order at a time, spread across plants and materials, and by the time it is visible in a spend report the contract is already up for renewal. AnrilX finds it by comparing windows, not by waiting for a threshold.

What gets in the way

Spend is on the item, not the header

The purchase order header does not carry an order value on many S/4HANA surfaces; the money lives on the line. Every “spend by supplier” report has to know that, and every ad-hoc extract that does not quietly reports the wrong grain.

A supplier is called something different in four places

It is the supplier on a purchase order, the invoicing party on an invoice, the vendor in conversation and a number in the data. Asking one question across those means knowing which field each document uses.

You find price drift after the renewal

Nothing flags a 6% move spread over eleven months and three plants. It is only visible if somebody thinks to compare the same material across two windows, which is exactly the query nobody has time to write.

Where you start, and what you get

Questions you can ask on day one

  • “Which suppliers have quietly become more expensive this year?”
  • “What did we spend with each supplier last quarter?”
  • “Which materials move most, by plant?”
  • “Where are goods receipts running late against the purchase order date?”
  • “Which purchase orders are still open past their delivery date?”
  • “What is our average price per unit for material TG11 this year against last?”

What you get back

  • Spend read at the grain where the money actually sits
  • Supplier resolved across documents that name the role differently
  • Price movement computed as a comparison, not eyeballed from two reports
  • A saved view that re-reads live SAP every time it is opened

Procurement questions are comparisons too

The useful procurement question is almost never what did we spend. It is what changed: which supplier moved, which material moved, and whether the movement is real or a mix effect from buying different things.

AnrilX answers those by running the same read against two windows and computing the difference in code. The currency is never mixed: a figure carries the currency field the document was written in, so a total across three currencies is refused rather than silently summed into a number that means nothing.

Where the money actually sits

On this SAP surface the purchase order header carries no order value at all, only a down payment and a release-time total. Spend is an item-level question. The header answers how many orders; the line answers how much.

That is not a detail you should have to know to ask a question, and with AnrilX you do not: the platform knows which entity carries which figure and reads the one that can actually answer. What it will not do is read the header, find something numeric, and present it as spend.

Processes that are not running

Some procurement processes are configured and empty: an RFQ flow nobody uses, a supplier evaluation that was never switched on. AnrilX reports that as what it is: this process holds no records here, rather than reaching for the nearest populated neighbour and returning a real figure for the wrong thing.

One of these questions, answered

Where are goods receipts running late against the purchase order date?

Four supplier and plant pairs are drifting. Tabuk Fabrication into plant 1000 is the worst: agreed lead time is 21 days, the trailing average is now 34, and it has moved in one direction every month since April. Nothing here is flagged late by SAP, because each receipt landed inside its own confirmed date; it is the confirmations that have been moving.

Supplier · plant Agreed · actual lead time
Tabuk Fabrication · 1000 21 · 34 days
Al-Majd Components · 1000 14 · 21 days
Hail Industrial Supply · 1200 30 · 38 days
Eastern Steel · 1400 10 · 14 days

Source: zsd_i_purchaseorder / I_PurchaseOrderHistory · lead time = goods receipt PostingDate minus purchase order CreationDate · per Supplier and Plant · trailing 90 days against the agreed term

Assumed: against the purchase order date, not against the confirmed delivery date. Measuring against confirmations would show almost nothing wrong, which is the point: the confirmations moved.

Before a pilot

What it will not do for this role

The limits that bear on these questions specifically.

  • Supplier identity is whatever your master data says it is. Where one vendor exists under several numbers, they are reported separately rather than merged on a name that looks the same.

  • A price it can see is a document price. Rebates, retrospective credits and anything settled outside the purchase order are not in that figure, and the answer says which documents it read.

  • Price movement is reported as it appears in the data. Whether a rise is drift, mix or an agreed change is a judgement the platform leaves to you.

Questions people ask us

Can it show price drift before a contract renewal?

Yes. That is the shape it is best at. The same read runs against two windows and the difference is computed in code, so a six per cent move spread over eleven months is visible as a figure rather than as a hunch.

Why do spend figures and order counts read different things?

They live at different grains. The count is a header question; the value is an item question, because on this SAP surface the header carries no order value. The platform reads whichever entity can answer and names the one it read.

We have suppliers named inconsistently across documents. Does that break it?

The supplier is the supplier on a purchase order and the invoicing party on an invoice: different fields for the same business role. The platform knows that, so you do not have to.

Bring the question your reports cannot answer

Thirty minutes against a live SAP system we provide: no access to yours, nothing to set up. If it cannot answer, you find that out in half an hour rather than three months into a pilot.