All analyses
AN-2026-06Sample analysis · synthetic data

The €500 approval limit nobody has touched since

A plastics processor has every purchase above €500 signed by the managing director. The limit is old, the prices are not. Four out of five orders pass through a gate that opens twice a week.

In short

The managing director signs off 79 % of all purchases. Not out of distrust, but because an approval limit dates from a different price era. That makes it the bottleneck in 78 % of runs, on 3 to 12 minutes of work per signature.

How long it takes

3.8working days

In 8 of 10 cases between 1.2 and 5.5 days.

Where it sticks

Freigabe Geschäftsführung

In 78 of 100 simulated runs this step was the hold-up.

What it costs

€49,000

Per year. Estimated between €43,000 and €55,000.

What the fix would return

€40,000

Per year. Estimated between €35,000 and €45,000.

01Starting point

Purchase approval

Industry
Plastics processor
Size
95 Mitarbeitende
Cases per year
3,100
Fully loaded rate
€70 / h
Simulated runs
120,000
Date
August 2026

Purchasing complains about waiting, the departments about express surcharges, the managing director about the volume of signatures. All three are right, and all three are describing the same number.

3,100 purchase requisitions a year, six steps, four roles, a €70 fully loaded rate. Order values are right-skewed: a median of €1,150 with a long tail upwards. Modelled and simulated over 120,000 runs.

02The model

Six steps, four roles, one measurable path

The process modelled in FlowVisual. ↯ marks a media break — the point where data is retyped from one system into the next.

Every value in the model, as a table.
StepRoleSystemDuration P10–P90Bottleneck
01Bedarf meldenFachabteilungPapier / E-Mail514 min0 %
02Prüfung & LieferantenwahlEinkaufERP1049 min3 %
03Freigabe AbteilungsleitungAbteilungsleitungERP39 min19 %
04Freigabe GeschäftsführungBottleneckGeschäftsführungPapiermappe312 min78 %
05Bestellung auslösenEinkaufERP618 min0 %
06Auftragsbestätigung prüfenEinkaufERP / E-Mail412 min0 %
03The measurement

120,000 runs, one clear answer

Every chart below shows before against after. Values are labelled directly — the colour is a second signal, never the only one.

Abb. 1Probability of each step being the bottleneck in a run. Before against after.
Abb. 2Lead time as a P10–P90 range. The light tick marks the median (P50).
Abb. 3Utilisation per role. Everything right of the red line is structural overload.
Throughput

77

Bestellungen/Woche

Days over capacity

53 %

Range P10–P90

1.2–5.5working days

04The finding

Freigabe Geschäftsführung — 78 %

In 78 % of runs the bottleneck is sign-off by the managing director. The work involved takes 3 to 12 minutes. The gate opens on Tuesdays and Thursdays, and one folder in five still sits there. Meetings, travel, sickness.

The approval limit is €500. With a median order value of €1,150 that means 79 % of all orders, 2,437 a year, need that signature. On paper that is 284 hours of managing-director time for a step that decides nothing the department head has not already decided.

Utilisation explains nothing here. No role sits above 80 %. What creates the backlog is not volume but rhythm: on 53 % of working days more is left open at close of business than can be cleared in a day. Almost always at the managing director, because up to four days pass between two signing sessions.

At the end of it, money leaves the building: 7.6 % of orders are placed later than they were needed. Purchasing chases, and in some of those cases an express surcharge applies, around €6,700 a year.

05The intervention

The approval limit rises from €500 to €2,500.

No tool, no role, no interface. One number in a policy.

The new limit follows the measured distribution rather than a gut feeling: at €2,500, 23 % of orders sit above it instead of 79 %. Department-head sign-off would remain in place for every single order. What disappears is a second signature, not a control.

What that costs belongs on the same line as the return: €2.1 m of annual order volume the managing director would no longer see case by case. Whether that is acceptable is not for a simulation to decide.

06The re-measurement

Median lead time down 66 %

How long it takes
3.81.3 working days
Throughput
7778
Days over capacity
53 % → 1 %
Saving per year
€40,000

Median lead time would fall from 3.8 to 1.3 working days, the P90 from 5.5 to 3.7. The share of orders placed too late would fall from 7.6 % to 1.6 %, express surcharges from around €6,700 to €2,400 a year.

The managing director would sign 711 orders instead of 2,437, 83 hours a year instead of 284. Utilisation in this process would fall from 76 % to 22 %, and days with a backlog from 53 % to 1 %.

The bottleneck moves. And again onto a gate, not onto capacity. The new most likely bottleneck, at 72 %, would be department-head sign-off, which happens once a day at close of business. Going further would mean changing that rhythm, not adding people.

What the intervention deliberately does not touch: one requisition in seven is too vague to price without a query. That loop stays in the model, keeps purchasing at 70 % utilisation, and is the next candidate once there is a fresh measurement.

07Limits of this analysis

What this analysis cannot tell you

Every measurement has limits. A measurement that hides them is advertising.

  1. 01

    This is a sample analysis. Process, roles, order values and timings are constructed, not collected at a client.

  2. 02

    An approval limit is a control decision, not a lead-time decision. This analysis calculates the time gained and quantifies the volume that would pass without individual sign-off. It does not evaluate the risk. A monthly report on every order between €500 and €2,500 restores visibility, but it does not replace an approval.

  3. 03

    The managing-director time freed up is valued at the fully loaded rate. That does not turn it into cash, it turns it into capacity. Whether value follows depends on how it is used.

  4. 04

    Express surcharges are modelled at 4 % of order value, capped at €200, applying to 35 % of late orders. All three are assumptions; the costing carries them with a range of 0.6 to 1.4.

  5. 05

    “Days over capacity” here means: on that share of working days, at least one role ends the day with more work outstanding than it can clear in a day.

  6. 06

    “Throughput” is the ceiling the scarcest role allows, not the actual volume. That is 62 orders a week. The ceiling barely moving is not an error: afterwards it depends on the requesting department, no longer on the approval.

  7. 07

    P10–P90 is not a worst case. In 10 % of cases it takes longer than the P90 value.

Your process will look different.

This analysis is a sample. Your numbers are not. With FlowVisual you model your own process and get the same evaluation — on your machine, with your values.