① Journal · Strategy
A publisher prices against the alternatives it believes you have. Where a function has genuine competing products, keeping a credible second vendor in play is frequently the cheapest leverage available, not because you split the workload, but because the incumbent has to price as if you might. Here is where the strategy earns its keep and where it does not.
Leverage is not what you do. It is what the vendor believes you could do.
Mainframe software pricing is a function of captivity. A publisher that knows you cannot move charges as if you cannot move. A publisher that has to reckon with a credible alternative prices against it. The two vendor strategy is the deliberate preservation of that alternative. In categories where mature competing products exist, an enterprise keeps a second supplier viable, sometimes by running both, sometimes by maintaining the skills, the migration path, and the relationship that would let it move. The point is not to operate a costly duplicate of everything. The point is that the incumbent renews you knowing the door is real.
The mainframe has more of these categories than its reputation suggests. Job scheduling has Control-M from BMC and CA 7 from Broadcom (CA), with BMC running an active displacement practice. External security has RACF from IBM, Top Secret and ACF2 from Broadcom (CA), all built on the same SAF interface, which keeps a migration technically credible. Source control, debugging, output management, and tape management all have competing products with real switching paths. In each of these, a buyer who has done the work to make a move credible holds leverage at renewal that a single source buyer simply does not. The strategy is not free, and it is not right everywhere, which is the part that matters. Read this with our comparisons of Control-M vs CA 7 and Top Secret vs RACF.
Two vendor decision matrix · function by function, what we commonly observe
| Function | Competing products | Two vendor verdict |
|---|---|---|
| Job scheduling | Control-M, CA 7, IBM Workload Scheduler | Strong, active displacement and a real migration path |
| External security | RACF, Top Secret, ACF2 on common SAF | Strong as leverage, but switching is a major program |
| Output management | CA Deliver, CA View, IBM and ISV equivalents | Moderate, credible alternatives exist |
| Debugging and SCM | Xpediter, IBM z/OS Debugger, Endevor, ISPW | Moderate, switching cost real but bounded |
| Core subsystems | z/OS, Db2, CICS, IMS, MQ | Weak, effectively single source, lean on contract terms |
These are general patterns, not recommendations for a specific estate. The right call depends on your spend, switching cost, and risk tolerance per function; validate competing products and migration paths against your own environment before acting.
The strategy only works where a credible second product exists and a migration path is genuinely deliverable. Scheduling, security, and tooling categories qualify. Core subsystems where you are effectively single source do not, and pretending otherwise wastes effort the incumbent will see through. Concentrate the two vendor posture where the alternative is real, and lean on contractual protections everywhere else.
Two vendors where the door is real, contracts where it is not.
An alternative that exists only on a slide is not leverage. Keep the second vendor's proposal current, the migration scoped, and the switching cost quantified, so the option is deliverable on the timeline the renewal demands. A publisher prices against the alternatives it believes are real, and the costed, ready to execute option is the one that moves the number. See our note on building a walk away position.
A current, costed option. Not a slide.
Running two vendors carries duplicate license, training, and operational cost, and that overhead is real. The strategy pays off only where the preserved leverage and the renewal stakes outweigh the duplication. For low spend functions, or where switching cost dwarfs the savings, single vendor with strong caps and exit rights is the better economics. Do the arithmetic per function rather than adopting the posture across the board.
Leverage has a cost. Make sure it clears it.
④ Why the second vendor matters
A publisher prices against the alternatives it believes you have. The credible second vendor is the cheapest leverage there is. Pick the right functions, keep the option current, do the arithmetic.
Typical reduction negotiated on renewal spend
Mainframe spend negotiated on the buyer side
Engagements delivered since 2019
Keeping a credible second supplier in play for a function where competing products exist, such as scheduling, security, or output management, so no single publisher holds you captive. The goal is not to split the workload evenly. It is to preserve a real, costed alternative the incumbent has to price against at every renewal.
It can carry duplicate license, training, and operational overhead, so it is not free. It pays off where the function has genuine competing products and the renewal stakes are high enough that preserved leverage outweighs the duplication. For low spend or single source functions, single vendor with strong protections is frequently better economics. See the five levers that work.
In categories with mature competing products and a credible migration path: scheduling with Control-M and CA 7, security with RACF, Top Secret, and ACF2, and source control and debugging tools. It works least where a function is effectively single source or where switching cost dwarfs the spend at stake.
Keep the second vendor's proposal current, the migration scoped, and the switching cost quantified, so the option is deliverable on the renewal timeline. Our license negotiation service builds and prices the alternative, and our contract review service secures the caps and exit rights for the functions you keep single source.
Related: Control-M vs CA 7 · Top Secret vs RACF · building a walk away position · license negotiation · contract review
Audit notice or renewal under 18 months out? We mobilize within 48 hours.
Get expert help →