An industrial manufacturer ran a structured USMM cleanup ahead of submission, retired dormant users, and avoided a projected six point four million dollar named-user shortfall.
Every result on this site is anonymised at the client's request. Specific figures are real and verifiable through a confidentiality-protected reference call arranged on request.
The client is an industrial manufacturer with plants in the United States, United Kingdom, and Czech Republic, running ECC 6.0 with eight engine licences and a named-user population of approximately twenty-one thousand three hundred users. The annual USMM submission window was opening sixty days from the date of engagement.
The engagement was triggered by an internal review of the prior year's USMM submission, which had identified anomalies in the Professional-band classification but had been submitted without correction due to time pressure. The procurement and SAM leadership had committed to running a structured cleanup in advance of the current submission window.
What was at stake was a projected named-user shortfall of approximately six point four million dollars based on the prior year's submission methodology, which would have crystallised on submission and almost certainly triggered a Global License Audit and Compliance notification within the following quarter.
There was no formal claim at the point of engagement because no submission had yet been made. The projected exposure was modelled from the prior year's classification output applied to the current named-user population. The projection produced approximately seven thousand one hundred Professional-band users against contractual entitlement of five thousand seven hundred.
The projected shortfall of one thousand four hundred Professional users, valued at list rate, came to approximately six point four million dollars. The figure represented the worst-case crystallisation that would have followed from a re-run of the prior year's methodology without cleanup.
The internal SAM team had documented twelve specific classification anomalies during their review, including a parent role-collection assignment that propagated Professional classification to operational users in two plants, and a developer-band assignment that had remained active for a project decommissioned eighteen months earlier.
The cleanup opened with a configuration-history audit of the role-collection table over the prior twenty-four months. The audit identified seventeen changes that had affected named-user classification, of which six were undocumented and four had propagated unintended classification across operational user populations. Each change was reviewed and corrected against the original design intent.
The team then ran a transaction-history rebuild over twelve continuous months, with activity-frequency thresholds aligned with the contract definitions for each named-user band. The rebuild reclassified approximately two thousand four hundred users out of the Professional band, principally into Limited Professional and Employee Self-Service categories.
A dormant-user review identified approximately one thousand one hundred users with no transaction activity at all over the twelve-month window. These users were retired from the active population under the contract definition that permitted retirement of dormant accounts. The retirements were documented and indexed for the submission file.
A separate workstream addressed engine measurements on the eight engine licences. Each measurement was rebuilt against contractual definitions, with no measurement found to be outside contracted entitlement after correction.
The cleaned submission was made within the standard submission window. The classification output sat within contractual entitlement on every named-user band and every engine licence. No shortfall was reported, no true-up was triggered, and no Global License Audit and Compliance notification followed in the subsequent quarter.
Three internal protocols were adopted as part of the cleanup outcome. A documented-change-control protocol was introduced for the role-collection table with sign-off required for any change inside the measurement window. A quarterly dormant-user review was added to the SAM operating model. A configuration-history audit was added to the annual pre-submission cycle.
Total elapsed time from engagement to submission was ten weeks. The submission was made within the standard window with no extension required.
Across the matters the firm closes each year, the same defensible procedures recur. The following observations apply directly to other SAP estates of comparable scope. A reading of the SAP ECC topic page and the USMM and LAW Manual white paper expands the underlying framework.
Further analysis on this defence pattern is collected in the USMM & LAW reading room.
Submitting before the cleanup would have been the most expensive ten weeks we never spent. We avoided the audit entirely.
We rebuild the measurement run from configuration up, validate every classification against transaction evidence, and present a defensible submission.
Read the brief →A pre-audit baseline review that establishes the defensible position before any data is shared with SAP.
Read the brief →The reading room cluster covering field notes, defence sequences, and contract levers for usmm & law matters.
A configuration-up rebuild of an inherited USMM measurement that turned a contested submission into a defensible record.
A multinational insurer rebuilds a sixty-thousand-user USMM submission and recovers eight figures of over-classification.
An audit notification, a renewal quote, or a USMM cycle in flight — the first conversation is at no cost and under privilege. $180M+ in client savings across 500+ engagements, with a sixty-eight per cent average claim reduction. Twenty years on this work.
Contact Us →Every Wednesday. Field reports from active matters, decoded SAP communications, and what to look for in the next audit cycle. Work email only.