Estates licensed on the wrong measurement basis paid 20 to 40 percent more than the workload required, because the default basis is not always the cheaper one
Front end counts what you protect and back end counts what lands in storage. Encrypted and pre compressed workloads barely reduce, which reverses the usual answer.
Prepared by Redress Compliance · August 18, 2026 · IBM storage licensing reviews. 20 to 30 estates reviewed, 2024 to 2025.
Executive summary
Estates licensed on the wrong measurement basis paid 20 to 40 percent more than the workload required. The metric definition decides the bill before any discount is discussed.
Workloads were licensed as individual components long after the suite crossover point had passed. The suite wins once two or more products run against the same data.
Legacy processor based deployments without current metric reporting carried full capacity exposure into every audit conversation, whatever the deployment actually looked like.
Workload mix has shifted toward encrypted databases and pre compressed media that barely reduce, which quietly made the back end default more expensive than it used to be.
What does the measurement basis actually decide?
The bill. The suite licenses on managed capacity in terabytes, and front end licensing counts the data protected at source while back end counts what sits in storage after deduplication and compression.
The same workload can price very differently on each basis. Entitlement definitions sit on the product page and the commercial terms in Passport Advantage.
Which basis wins depends on how well your data reduces
- High reduction environments favour back end measurement, because stored capacity is a fraction of protected capacity.
- Low reduction workloads can favour front end, which is the case for encrypted or pre compressed data.
- Mixed estates need both bases modelled on real telemetry, because the wrong default cost 20 to 40 percent.
When does the suite beat individual components?
Once two or more components run against the same data. Estates adding a second product, or bringing space management into scope, almost always price better on suite terabytes than on stacked component licences.
| Dimension | Suite per terabyte | Component per terabyte | Legacy processor |
|---|---|---|---|
| Metric | Managed capacity, one pool | Capacity per product | Processor value units |
| Bundling | Full family included | Single product | Single product |
| Best fit | Multi component estates | A single workload | Stable legacy hosts |
| Audit exposure | Capacity measurement basis | Per product reconciliation | Sub capacity reporting rules |
| Cost trend | Predictable with growth | Stacks with each addition | Spikes on hardware refresh |
The mispricing found in the reviews was not estates on the wrong product. It was estates that passed the crossover point years ago and never revisited the structure.
The IBM audit defence guide
Capacity metrics, suite economics and the audit posture across the storage portfolio.
Get the brief →What 20 to 30 IBM storage estates showed
Across roughly 20 to 30 IBM estates with Storage Protect deployments reviewed between 2024 and 2025, the capacity model was the recurring cost surprise. Three patterns recur.
- Front end against back end confusion: estates licensed on the wrong measurement basis paid 20 to 40 percent more than the workload required.
- Suite against component mispricing: workloads licensed as individual components when the crossover point had long been passed.
- Sub capacity discipline missing: legacy processor deployments without current metric reporting carried full capacity exposure into every audit conversation.
Workload mix has shifted toward encrypted databases and pre compressed media that barely reduce, so the back end default quietly stopped being the safe answer.
- Your agreements decoded into plain English before the auditor interprets them for you
- Coverage grid: liability caps, intellectual property protections and service levels checked in one pass
- A defensible position paper generated in minutes rather than weeks
How do you stay audit safe on this portfolio?
By keeping the sub capacity evidence current. Legacy processor based hosts without current metric reporting default to full capacity, which is the most expensive reading of the same hardware and the easiest to avoid.
Sub capacity entitlement is conditional on evidence rather than on deployment. A missing report converts a sub capacity estate into a full capacity bill, and it arrives on the vendor's calendar rather than yours.
The capacity basis is itself an audit surface
On the suite the measurement basis is what gets reconciled, so the definition you signed and the telemetry you can produce have to agree. The reporting controls sit in the sub capacity compliance guide.
Watch the briefing · 5:44The IBM Audit Is the Sales CallSub capacity entitlement is conditional on evidence, not on deployment.
What should be retested at every renewal?
The measurement basis, against twelve months of telemetry. Reduction ratios move as the workload mix moves, and the basis that was correct three years ago is not automatically correct now.
The definition is the price
Negotiate tier placement on committed growth rather than on current capacity, and cap the support and subscription uplift before signature rather than after. The other metrics in the family are compared in the storage licensing models guide.
Reading the metric definitions before sizing is not optional. On a capacity metric the definition is the price, and it is the one part of the deal nobody negotiates because it does not look like a commercial term.
Where the storage line sits inside a wider agreement is worked through in the enterprise agreement renewal guide.
What the reviews measured, 2024 to 2025
Two cuts of the engagement file describe where the surprise came from.
Where the measurement basis did not match how the workload actually reduces, measured against what the estate required.
The point at which suite terabytes start to beat stacked component licences on the same protected data.
Neither is a discount question. Both are structure questions, settled before the quote and rarely reopened afterwards.
Your first five moves
- Model both measurement bases on twelve months of real telemetry, because the wrong default cost 20 to 40 percent and the right answer depends on how your data actually reduces.
- Check whether you passed the suite crossover point, which arrives as soon as two or more components run against the same protected data.
- Restore current metric reporting on every legacy processor host, since without it the entitlement defaults to full capacity in any audit conversation.
- Retest the basis at every renewal rather than carrying it forward, because reduction ratios move with the workload mix toward encrypted and pre compressed data.
- Negotiate tier placement on committed growth and cap the uplift before signature. The IBM practice models both bases before the sizing conversation starts.
Frequently asked questions
How does the suite bill?
On managed capacity measured in terabytes, with the measurement basis deciding the number. Front end counts data protected at source and back end counts what sits in storage after reduction.
Which basis is cheaper?
It depends on how well the data reduces. High reduction environments favour back end, while encrypted or pre compressed workloads that barely reduce can favour front end.
How much does the wrong basis cost?
Between 20 and 40 percent more than the workload required, across the estates reviewed. It is a definition problem rather than a pricing one.
When does the suite beat components?
Once two or more components run against the same data. Estates adding a second product or bringing space management into scope almost always price better on suite terabytes.
Why has the back end default weakened?
Because workload mix has shifted toward encrypted databases and pre compressed media that barely reduce, so stored capacity is no longer a small fraction of protected capacity.
What happens to legacy processor hosts?
Without current metric reporting they carry full capacity exposure into every audit conversation, regardless of what is actually deployed on them.
Is sub capacity automatic?
No. It is conditional on evidence rather than on deployment, so a missing report converts a sub capacity estate into a full capacity bill on the vendor's calendar.
How often should the basis be retested?
At every renewal, against twelve months of telemetry. Reduction ratios move with the workload mix, so a basis that was correct three years ago may not be now.
What should tier placement be negotiated on?
Committed growth rather than current capacity, with the support and subscription uplift capped before signature rather than argued afterwards.
What is the most overlooked term?
The metric definition. On a capacity model the definition is the price, and it is rarely negotiated because it does not read like a commercial term.