A financial institution reversed years of IBM mainframe cost growth. The review rebuilt the MLC baseline, tested Tailored Fit Pricing honestly, and tuned the peaks that set the bill.
A financial institution running core banking on IBM Z asked us to review a mainframe software bill that grew every year while workloads stayed flat. Nobody inside the institution could explain the growth line by line.
The review rebuilt the cost baseline, found the growth drivers, and delivered a material, recurring reduction without touching application functionality.
The institution cut its recurring mainframe software cost materially by rebuilding the MLC baseline, retuning the peaks that set monthly charges, and renegotiating its pricing structure from a tuned position rather than an unmanaged one.
The estate ran z/OS with Db2, CICS, and MQ supporting core banking. Software charges followed the rolling four hour average peak, and those peaks had drifted upward for years as batch scheduling decayed.
Finance flagged that mainframe software was the fastest growing infrastructure line while transaction volumes were flat. That mismatch is almost always a licensing mechanics problem, not a demand problem.
The rebuilt baseline showed the bill was set by a few predictable peaks, carried legacy products nobody used, and included sub capacity reporting errors that had never been challenged.
Monthly license charge products bill on the peak rolling four hour average, reported through IBM's sub capacity reporting process. The institution generated reports faithfully and reconciled them never.
Peak analysis traced the charge setting intervals to specific workloads:
With the baseline tuned, alternative structures were modeled honestly, including IBM Tailored Fit Pricing against traditional sub capacity MLC under the Passport Advantage framework for the IPLA products. The right answer depended entirely on the tuned consumption profile.
Four levers produced the reduction: peak retuning, workload placement, portfolio cleanup, and a renegotiated pricing structure entered from strength.
Cost reduction by lever
| Lever | What changed | Effect on bill |
|---|---|---|
| Peak retuning | Batch spread out of charge setting windows | Monthly peaks dropped |
| Workload placement | Dev and reporting moved off production peaks | Peak contribution removed |
| zIIP offload | Eligible workload shifted to specialty engines | General processor charge fell |
| Portfolio cleanup | Unused legacy products terminated | Recurring charges eliminated |
| Pricing structure | Renegotiated from tuned baseline | Growth terms contained |
The common advice is to move to Tailored Fit Pricing because predictable consumption pricing beats chasing peaks. We disagree as a default. In roughly 15 to 25 mainframe reviews we advised across 2024 and 2025, customers who signed consumption deals against unmanaged baselines locked their waste into the new contract's base year, while customers who tuned first either negotiated far better TFP terms or found tuned MLC cheaper. The buyer side move is strict ordering: tune the estate, then model both structures, then choose. Pricing model changes reward whoever controls the baseline.
The tuning targets were set against the IBM Z platform documentation and the reporting reconciled with IBM Z software pricing tools, keeping every correction on IBM terms.
Source: Redress Compliance advisory engagement file, 2024 to 2025.
The work was operational discipline applied to licensing mechanics, not contract heroics.
More IBM cost analysis lives in the IBM knowledge hub and the IBM practice.
A material, recurring reduction in monthly software cost, achieved by retuning charge setting peaks, moving eligible workload to zIIP engines, terminating unused products, and renegotiating the pricing structure from a tuned baseline.
Monthly license charge products bill on the peak rolling four hour average of the month, reported through sub capacity reporting. A few peak intervals, often batch windows, set the bill for the entire month.
It depends on the consumption profile, and the comparison is only honest after the estate is tuned. Signing TFP against an unmanaged baseline locks existing waste into the contract base.
zIIP specialty engines run eligible workloads, such as much Db2 and Java processing, without driving the general processor charges that MLC bills on. Eligible workload left on general processors is avoidable cost.
Monthly. Reports that flow to billing unreconciled accumulate errors and stale mappings, and every unchallenged month becomes precedent for the charge.
The eleven move framework, the IBM mainframe framework, the IBM zSeries licensing framework, the IBM ELA framework, and the buyer side moves at every step of the IBM mainframe renewal cycle.
Used across more than five hundred enterprise software engagements. Independent. Buyer side.
IBM framed the IBM mainframe framework as the immediate IBM uplift at the renewal cycle. Redress reframed the framework around the customer's actual IBM mainframe utilization framework. IBM mainframe licensing review delivers material commercial saving.
Independent. Buyer side. The advisory firm enterprise software vendors do not want you to hire.
IBM mainframe framework signals, IBM zSeries licensing framework signals, IBM ELA framework signals, and the broader IBM software licensing leverage signals across the practice.
Once a month. Audit patterns, renewal benchmarks, vendor commercial signals across Oracle, Microsoft, SAP, Salesforce, IBM, Broadcom, AWS, Google Cloud, ServiceNow, Workday, Cisco, and the GenAI vendors. No follow up sales pressure.
Free providers (Gmail, Yahoo, Outlook) cannot subscribe. Work email only. Unsubscribe in one click.
