01 Overview

Order to cash at a major automotive parts manufacturer

Two days at one site. Sixteen at another.

The same configured order to cash process, running across every distribution centre in the network. Service metrics looked healthy in aggregate. Underneath them, identical orders were behaving nothing alike.

Of 2.73M order lines, a quarter sat in the bottom quartile19.84 days against 4.59 for the rest

682,752 lines at 19.84 days2.05M lines at 4.59 days

Bottom quartileEverything else

44.09Mrecords reconstructed

2.73Morder lines analysed

26activities across 3 functions

99.4%fill rate reported

6 monthsof production data

02 The problem

Healthy metrics, uneven execution

Fill rate above ninety-nine per cent. On-time shipment in the mid-eighties. Backorders at four per cent. By every standard measure the operation was performing, which is precisely why the variation underneath had never been addressed.

What the reporting could not show was how an order line actually travelled between booking, picking, packing, shipping and payment — or why the same order type took two days at one distribution centre and sixteen at another. Four things stayed invisible.

Site performance varied fivefold
The same process ran in days at one distribution centre and in weeks at another, with no visible cause.
Bottlenecks without causes
Delay appeared in totals but never against the individual activities producing it.
Execution was entirely manual
Every activity in scope was performed by hand, with no measure of where automation would pay.
Cash arrived late, predictably
Invoices were settled well past due date in a pattern nobody was tracking as a process.

03 How it worked

From Oracle EBS records to findings

The manufacturer provided read-only extracts covering six months of closed orders and a short period of subject-matter support. RE-ViVE handled the data modeling, execution reconstruction and analysis from records the business was already keeping.

Step 1

Read the source evidence as it existed

Sales and internal orders, delivery and pick events, billing documents, hold records and receivables were distributed across Oracle E-Business Suite. The evidence was not delivered as ready-made process flows organised by site or order type.

Step 2

Create a common execution foundation

RE-ViVE mapped those records into a common Execution Data Model, linking each order line to its deliveries, invoices and payments so distribution centres, warehouses, customers and items could be compared consistently.

Step 3

Reconstruct and analyse execution

RE-ViVE reconstructed the observed paths and measured cycle time, SLA deviation, delay contribution and automation potential across all 26 activities, each drillable to the evidence behind a single order line.

One order line, reconstructed from source evidenceOL-53235 — standard sales order

04 Jul 08:22	Order created
04 Jul 09:10	Order booked
06 Jul 13:45	Sales order pick
10 Jul 11:02	Packing workbench	4.9 days
12 Jul 07:55	Delivery created
13 Jul 16:30	Billing — invoice
22 Aug 09:12	Payment received	40 days

Neither highlighted row is an exception. The packing step averaged 5.4 days across the estate, and the wait for cash routinely exceeded the entire fulfilment process that preceded it.

04 What we found

What the evidence showed

Finding 1

A quarter of order lines took five times longer

Order-to-delivery averaged 8.4 days across 2.73M order lines. The spread inside that average was the real story.

682,752 lines in the bottom quartile ran to 19.84 days. Remove those variants and the remainder completed in 4.59 — without touching the three quarters that already performed.

19.84 days

bottom quartile — 682,752 order lines

8.40 days

the average across all 2.73M lines

3.96 days

potential target once low-performing variants are addressed

Finding 2

The difference between two sites was a single step

One distribution centre averaged 15.82 days across 621,510 order lines. Another ran the identical process in 2.32.

Comparing the two populations activity by activity isolated delivery creation as the difference — running roughly ten times longer at the slower site. Not workload, not order mix, not headcount.

2.32 daysThe fastest site — the proven path exists in-house

15.82 daysThe slowest site, and the one carrying the volume

The slower site handled roughly a quarter of everything analysed, so its performance is a large share of the customer experience rather than an edge case.

Finding 3

One custom step carried the fulfilment delay

The packing workbench averaged 5.4 days — the single largest activity-level contributor, and the one the automation ranking surfaced as an outlier across all 26 activities without anyone nominating it.

Sales order pick told a second story: 1.45 days normally, but 4.02 once the SLA was missed, across a known population of 219,610 order lines.

5.4 daysaverage time in a custom packing step that sits outside standard reporting

Finding 4

Collection ran longer than the process itself

Three of the five highest-volume customers averaged more than one hundred days to pay, the slowest at 117.6.

The lateness was consistent rather than erratic — no customer paid beyond thirty days past due — which makes it a collections discipline question rather than a credit risk one.

Slowest payingFastest of the five

Where an invoice had several payments, the first was measured against the due date — these figures describe when cash starts arriving, not when the invoice clears.

05 What we recommended

Where the time comes back

Five plays, in the order we would take them. Every figure is drawn from the manufacturer's own records and describes opportunity identified during the engagement.

Standardise onto the proven path

Retire the low-performing variants behind 682,752 order lines and hold new orders to the flows that already work.

Measured opportunity: move average cycle from 8.4 days toward 3.96

Automate the packing workbench step

The activity ranking surfaced it as the outlier across all 26 activities, and it sits outside standard reporting.

Measured opportunity: targets a 5.4-day average activity

Close the distribution-centre gap

Transfer the delivery creation practice from the fastest site to the slowest, where the same step runs ten times longer.

Measured opportunity: 15.82 days toward 2.32, proven in-house

Hold sales order pick to its SLA

Missing the SLA nearly triples the step, and the exposure sits in a known population of order lines.

Measured opportunity: 4.02 days toward 1.45 on 219,610 lines

Work the slowest-paying accounts

Lateness is consistent and bounded, which makes it a matter of collections discipline rather than credit risk.

Proposed action: target the 60–118 day payment averages

06 Next steps

The opportunity was already in the data

Every number on this page came from records the manufacturer had been keeping all along. RE-ViVE made them measurable, and then made them actionable.

This is Execution Intelligence: reconstructing how work actually executed from the evidence already produced by the enterprise, then exposing where time, effort and complexity accumulate.

Read the full case study Talk to us about your process