Guides · Software AG audit defense

Software AG Adabas and Natural Audit Defense.

A Software AG (formerly Software GmbH) audit does not live in your current usage. It lives in your oldest contracts: capacity definitions written for a machine you retired, entity clauses that predate three reorganizations, and a stack of amendments nobody has read end to end. The defense is to control the timeline, measure capacity yourself, and answer every question on the boundary the contract actually specifies.

The risk is in the paper, not the workload.

With its broader portfolio sold off and the business narrowed to Adabas & Natural and ARIS under Silver Lake ownership, pressure on the installed base is now central to how Software AG realizes revenue. Adabas and Natural estates are sticky and long lived, and that is exactly why the audit exposure is structural rather than behavioral. The licensing rarely sits in one clean contract; it is a stack of amendments, side letters and acquired entity agreements built up over decades, and the effective terms are whatever that stack adds up to.

Commonly observed audit themes follow from that. Capacity measured on the wrong boundary, the whole machine rather than the licensed LPAR, inflates the apparent usage. Acquired or divested entities run on licenses that never formally transferred during an M&A event. Disaster recovery arrangements were stood up operationally and never papered. None of these is necessarily real overuse. Most are definitional disputes, and definitional disputes are won by whoever measures first and reads the contract most carefully.

The defense playbook · five moves

1 Move 01

Control the timeline

Acknowledge the notice, then set the pace. Audits run on momentum the vendor manufactures with deadlines. Agree a reasonable schedule in writing, route all communication through a single point, and do not let an open ended data request set the clock. Within the 48 hours after a notice lands, the priority is containment, not answers.

2 Move 02

Read the actual contract stack

Assemble every amendment, side letter and acquired entity agreement and reconstruct what you are actually licensed for: which products, which entities, which capacity definition, measured on which boundary. The audit will be argued on these clauses. You cannot defend a position you have not reconstructed, and the vendor is counting on the stack being unread.

3 Move 03

Measure capacity independently first

Establish what Adabas and Natural actually consume, on which LPARs, against which contracted tier, before any number leaves the building. Data shared early on the vendor's definition of the boundary frames the whole audit on their terms. Your own measurement, on the boundary the contract specifies, is the position everything else is argued against.

4 Move 04

Answer on the contracted boundary

Respond to each finding on the measurement the contract actually requires, not the one the auditor prefers. Whole machine versus LPAR, production versus DR, the entity that holds the license versus the one running the workload: these distinctions are where inflated findings collapse. Every claim gets matched to a clause, and every measurement gets matched to its defined boundary.

5 Move 05

Convert the settlement into a clean contract

Resolve the audit and use the same reconciliation to fix the underlying problem: consolidate decades of amendments into one negotiated, right sized agreement with clear capacity definitions, entity transfer language that survives the next reorganization, and DR terms in writing. A settlement that leaves the ambiguous stack in place simply schedules the next audit.

What changes with us on the file.

Software AG audits are won on preparation and contract literacy, not on goodwill. The vendor relies on the customer never having reconstructed the stack or measured its own consumption on the correct boundary. We do both before responding. Across 500+ engagements and $180M+ of negotiated mainframe spend, the pattern on this publisher holds: the findings that look largest at the opening meeting are most often definitional, and they shrink fastest against a buyer who measured first and reads every clause.

What you get

  • A controlled audit timeline with communication routed through a single point
  • A reconstructed contract stack: what you are actually licensed for, by entity, product and capacity definition
  • An independent measurement of Adabas and Natural consumption on the contracted boundary
  • Each finding answered against the clause and boundary that actually apply
  • A settlement that consolidates the paper into one clean, right sized agreement

Frequently asked

Q1

Why do these audits hit old contracts?

Because Adabas and Natural estates have run for decades, and the paper has accumulated as a stack of amendments, side letters and acquired entity agreements rather than one clean contract. Capacity definitions, named entity clauses and usage restrictions written for a different datacenter are where exposure hides, and that ambiguity is what an audit feeds on.

Q2

What are the most common findings?

Capacity measured on the wrong boundary, the whole machine instead of the licensed LPAR; acquired or divested entities running on licenses that never transferred; and DR arrangements that were never papered. None is necessarily real overuse. Most are definitional disputes that turn on which clause and which measurement applies.

Q3

Should we send our own data immediately?

No. Measure independently first and understand your own position before any number leaves the building. Data sent early on the vendor's definition of the boundary frames the audit on their terms. Establish what Adabas and Natural actually consume, on which LPARs, against which tier, and only then decide what is shared and how.

Q4

Can an audit become a better renewal?

Frequently. Once the stack is reconciled and real consumption is documented, the same work exposes where the contracted tier sits above actual use. A settlement that also consolidates decades of amendments into one clean, right sized agreement removes the ambiguity the next audit would feed on.

Related

The Software AG buyer side guide →
01 Software AG audit defense service

Run this with usThe service that executes the playbook on this page if a notice has landed. Software AG mainframe audit defense →

02 Software AG MSU optimization

After the auditReset the contracted tier the reconciliation exposed. Software AG MSU optimization →

03 Contractual vs consumed capacity

The gap audits exploitWhy licensed capacity drifts from real use, with a worked example. Contractual vs consumed MSU →

A notice landed. Measure before you answer.

Get expert help

Audit notice or renewal under 18 months out? We mobilize within 48 hours.

Get expert help