① Comparison · IBM pricing models
Neither model wins universally. Sub-capacity rewards an estate that already shapes its R4HA peak. Tailored Fit Pricing removes the peak and prices full consumption with growth discounts and operational freedom. The right answer is the one your workload profile and growth trajectory point to, and you only know it by modeling both.
Move to Tailored Fit Pricing (TFP) if you have measured MSU growth ahead or your operations are constrained by soft capping. Stay on sub-capacity if your estate runs flat and you have already optimized the Rolling 4-Hour Average hard. The trap is assuming TFP is automatically cheaper because it is newer and cloud like. For an estate that has spent years suppressing its R4HA peak, TFP can price the consumption that peak management was keeping off the bill, and cost more. The decision is an arithmetic one, not a fashion one.
The two models measure different things and reward different behavior. The contrast that matters at renewal:
| Dimension | Sub-capacity (R4HA) | Tailored Fit Pricing |
|---|---|---|
| Pricing basis | Rolling 4-Hour Average monthly peak | Total measured consumption over a baseline |
| What lowers the bill | Shaping and capping the peak | Negotiating the baseline and growth rate |
| Growth treatment | Full rate on a rising peak | Discounted rate on measured growth |
| Soft capping incentive | Strong, capping suppresses cost | Removed, run without capping constraint |
| Predictability | Varies with monthly peak | More predictable, baseline plus growth |
| Best for | Flat estates already optimized on R4HA | Growing or capping constrained estates |
| Main watch out | Peak management overhead and risk | Baseline set too high; consumption now fully priced |
Directional and pattern level. IBM still offers MLC, One Time Charge, Value Unit Pricing, and Country Multiplex Pricing alongside TFP, so the choice is not binary. Model your own SCRT data against each before committing.
Map your estate to the model whose incentive matches how you actually run:
Lean Tailored Fit Pricing if
Stay on sub-capacity if
The single most important move either way is to validate the baseline. A TFP deal is only as good as the consumption baseline it is built on, and that figure, like an R4HA measurement, must be measured and challenged independently before you sign.
The newer model is not automatically the cheaper one.
Explainers: the Rolling 4-Hour Average explained, hardware model capacity ratings and software cost, and zIIP engines and software cost offload. Other comparisons: annual vs multi year terms and in house vs outsourced mainframe. Hub and commercial: the IBM buyer side guide, mainframe cost optimization, and mainframe contract review.
Audit notice or renewal under 18 months out? We mobilize within 48 hours.