Heizen

Acme Industrial Systems

Industrial components manufacturing · 11 countries · 2 plants

Discovery conversationThursday 20 August 2026, 15:00 IST· 45 minutes

Gaps / pain points

What we suspect, and have not confirmed

3 hypotheses · 6 unknowns · 3 matched engagements

These are potential gaps. Not one of them is a client problem yet. Each carries the evidence it rests on, what we still do not know, the question that would settle it, and any Heizen work that matches.

HYPOTHESIS

Everything on this page is a belief we hold, not a finding

A gap becomes evidence only when the client confirms it in the room. Until then it is framed as a question. Taking a hypothesis into a meeting as a finding is the fastest way to lose the room.

Potential gaps · 3

Ordered for the CTO. The order changes with the person you are meeting; the hypotheses do not.

HYPOTHESISfrom a market pattern — unconfirmed herecritical if true58% confidence

Master data may not be harmonised ahead of the cutover

What we think may be happening — not what they have told us

The migration scope names master-data harmonisation across eleven entities. If regional material and vendor masters have diverged since the regions were stood up, the cutover date is exposed — and so is anything built on that data afterwards.

What normal looks like

Harmonisation named explicitly in a job scope usually means it is known to be a problem, not that it is finished.

Evidence we are reading it from

  • —ERP modernisation is live and mid-flight
  • —Regional expansion has outpaced the operating model

What we do not know

  • UNKNOWNWhich entity's material and vendor master is treated as authoritative today
  • UNKNOWNWhether a data-quality baseline has ever been measured

The question that would settle it

Which entity's material and vendor master is treated as authoritative today, and who signs off a change to it?

Seen beforeKestrel Logistics · S/4HANA master-data harmonisation across 11 entities

HYPOTHESISfrom their evidence and a market patternhigh if true64% confidence

Regional operating processes may be fragmented

What we think may be happening — not what they have told us

Regional teams may be coordinating operational decisions through people and spreadsheets rather than through a shared process. If so, each region added costs more to run than the one before it.

What normal looks like

Manufacturers that pass roughly eight operating countries usually either standardise the regional planning process or accept a permanent coordination tax. Acme is at eleven.

Evidence we are reading it from

  • —Regional expansion has outpaced the operating model
  • —Regional planning runs on spreadsheets outside the system
  • —The planning cycle is lengthening as regions are added

What we do not know

  • UNKNOWNHow regional teams settle it when their plan and the central plan disagree
  • UNKNOWNWhether any region has already built its own local tooling

The question that would settle it

How do the regional teams settle it when their plan and the central plan disagree?

Seen beforeNorthfield Foods · Multi-region demand planning coordination

HYPOTHESISfrom evidence about themhigh if true71% confidence

Manual reconciliation may be absorbing planner capacity

What we think may be happening — not what they have told us

Two open roles exist to consolidate and reconcile plans by hand, and a planner describes re-keying regional files weekly. The effort may be large enough to price — but no public source states the hours, so we do not know that yet.

Evidence we are reading it from

  • —Regional planning runs on spreadsheets outside the system
  • —Headcount is absorbing the coordination load
  • —The reconciliation work is visible to the people doing it

What we do not know

  • UNKNOWNHow many hours per cycle the regional reconciliation actually takes
  • UNKNOWNWhether the weekly variance pack has ever changed a decision

The question that would settle it

How many hours does the monthly regional reconciliation take, and who does them?

Seen beforeBeltra Components · Reconciliation and three-way match automation

Open unknowns · 6

Each is attached to a hypothesis and to the question that would close it. An unknown with no question against it is a hole in the preparation.

UNKNOWNunder master data may not be harmonised ahead of the cutover

Which entity's material and vendor master is treated as authoritative today

Without an answer there is no baseline for the harmonisation, and no way to size the cutover risk.

UNKNOWNunder master data may not be harmonised ahead of the cutover

Whether a data-quality baseline has ever been measured

Determines whether anything can be built on their data in the first two quarters, or only after cutover.

UNKNOWNunder regional operating processes may be fragmented

Whether any region has already built its own local tooling

A region that has already solved this locally is either the pilot site or the political obstacle.

UNKNOWNunder manual reconciliation may be absorbing planner capacity

How many hours per cycle the regional reconciliation actually takes

This is the number the business case is built from. Nothing outside the company states it.

UNKNOWNunder regional operating processes may be fragmented

How regional teams settle it when their plan and the central plan disagree

Decides whether this is a tooling problem or an accountability problem. The two need different proposals.

UNKNOWNunder manual reconciliation may be absorbing planner capacity

Whether the weekly variance pack has ever changed a decision

If it has not, the reporting is ceremony and the real decision happens somewhere we have not found yet.

Relevant Heizen experience · 3

Shown here rather than as its own area, because past work only earns a place when it matches a hypothesis on this page.

Kestrel Logistics

Logistics · 2025

S/4HANA master-data harmonisation across 11 entities

Brought up byMaster data may not be harmonised ahead of the cutover

Why this experience is relevant here

Identical entity count, and an identical harmonisation clause in the programme scope. The work there was almost entirely establishing which master was authoritative before anything could be merged — which is precisely the unknown sitting under this hypothesis.

What was measured there

  • —340k material records deduplicated to 190k
  • —Authoritative master agreed across 11 entities in 9 weeks
  • —Cutover held its original date
OPPORTUNITY

Pre-cutover master-data baseline

Sits inside a programme they are already funding, which is usually the cheapest way in.

Proven on Kestrel Logistics

Northfield Foods

Food manufacturing · 2025

Multi-region demand planning coordination

Brought up byRegional operating processes may be fragmented

Why this experience is relevant here

The same shape: nine countries planning locally in workbooks, and a central S&OP that could not see any of it until the monthly review. The coordination cost showed up as cycle time, exactly as it appears to here.

What was measured there

  • —Planning cycle 14 days → 6 days, held over two quarters
  • —220 manual hours per cycle → 40
  • —Regional plans written back to the central plan on a fixed daily interface
OPPORTUNITY

Regional planning coordination layer

A thin layer over the existing plan submissions rather than a replacement for them, so it can be built before an ERP cutover rather than after.

Proven on Northfield Foods

Beltra Components

Automotive components · 2024

Reconciliation and three-way match automation

Brought up byManual reconciliation may be absorbing planner capacity

Why this experience is relevant here

The reconciliation pattern is the same one described in the Acme planner review — regional files re-keyed against a central record every week — and Beltra is close enough in size that the measured numbers transfer as a reference, not as a promise.

What was measured there

  • —Exception rate 4.2% → 0.6%
  • —₹1.8 Cr recovered annually against a measured baseline
  • —Payback in 7 months
OPPORTUNITY

Automated regional reconciliation

Cannot be priced for Acme until the reconciliation hours are known. Positioning only until that question is answered.

Positioning only — no proof point until the number is known

Prepared for Technical Consultant meeting CTO. 0 questions on the agenda.