Before you start that fight, here is the uncomfortable read. The quote is probably right, in the sense that it is a faithful price of exactly what you handed over, counted honestly and priced by the rules. Quotes can carry real errors, wrong assumptions, stale weights, or a mix-up between SaaS and customer-managed terms, so rule those out first. But once you do, the high number is not padding. It is your current access, sized and totaled. Which means arguing the quote is the wrong fight, and you will lose it, because the math is on the vendor's side. The savings you are actually after are upstream, in the model you fed them, and unlike a one-time discount they repeat every year of the term.
A quote prices your inputs, not your intentions
A reseller cannot design your future role model for you. They do not know which of your three hundred users only check status, or which planners could sit a tier lower, or which inspectors log in twice a month. They price what you give them: a user list, the apps those users can open, and the tiers that implies. Hand them the accumulated state of your system and they price the accumulated state, accurately and in good faith.
So when the number lands high, the honest reading is usually "we gave them a high model," not "they inflated it." I want to be fair to the vendor here, and not only because it is true. Misdirecting the blame points you at the one thing you cannot change, the price book, and away from the one thing you can, the model. Spend your energy on the quote and you are negotiating rounding error. Spend it on the model and you are changing the number that gets multiplied.
Why the model you handed over is almost always bloated
The input is built from current access, and current access is years of "just give them access so they stop calling me." Every grant that was easier to add than to scope is still there, and every one of them is now a line on your bill. That is not a knock on your team. It is what every long-lived Maximo system looks like before someone sizes it for a move.
The usual suspects are easy to name once you go looking. Tiers handed out by title or seniority instead of by the work the person actually does. Everyone who touches Maximo carrying a consuming license, including the people whose whole job is requesting and checking status. Authorized access by default, even for people who log in once a week. Dev and test sized as if each were a second full production. Dormant accounts still holding live licenses. None of that is the vendor's doing. All of it is the input.
| The input problem | What it does to the number | The fix |
|---|---|---|
| Tier by title, not work | Over-tiers a whole population | Assign tiers by what the work requires |
| No self-service | Licenses pure requesters who only submit and check | Move them to Self-Service at zero |
| Authorized by default | Pays for seats nobody occupies | Concurrent for low-peak roles |
| Environments double-counted | Prices dev and test as production | Model each environment on its own terms |
| Dormant accounts | Pays for people who left | Reclaim them to No-Access |
The math you are actually choosing between
Picture the two paths side by side.
On the first path you argue the quote. Best case, you win a single-digit discount on a number that is structurally too big, and it is a one-time concession against a multi-year commitment. You signed up to overpay, just slightly less.
On the second path you fix the model and re-quote. The reduction is not a few percent, it is the difference between sizing your future bill off your present mess and sizing it off what your roles actually need. A well-run design often lands at or below the licenses you already hold. And because AppPoints are committed for the term, that reduction repeats every year you are under contract, not once.
That is the whole case. An afternoon of honest role design is worth more than any concession a reseller can hand you, because the concession happens once and the design saving happens every year.
Use the vendor, do not fight them
The move is not to dispute the number. It is to change the input and ask for a fresh quote against the designed model. Rebuild the roles first: lowest tier that covers the work, pure requesters to Self-Service at zero, intermittent low-peak roles to Concurrent sized on real peak rather than headcount, dormant accounts reclaimed, each environment priced on its own terms. The full method is in how to size without overpaying.
Then hand the clean model back and ask them to price that. A good reseller will do it happily, because a clean model is a faster and cleaner deal for them too. While you are at it, get a measured peak instead of an assumed one. IBM ships a small usage tool you install into 7.6 that captures actual connection counts and peak concurrency, which is exactly what a Concurrent pool has to cover, so your concurrent roles are sized on data and not on a guess. Concurrent is a peak-usage bet, not a default discount, so the measured peak matters.
The difference between the first quote and the second is the real prize. The discount you might have negotiated is rounding error next to it.
What to do before you sign or push back
Rebuild the input, role by role. Assign tiers by what the work requires, move the requesters to Self-Service, decide Authorized or Concurrent by how each role actually works, reclaim what nobody uses, and price your non-production environments the way your deployment model prices them. Then re-quote against that, and compare the two numbers.
Do not let the clock stampede you into signing the bloated quote just to beat a deadline. The migration timeline is real and it is pushing you toward a decision, as the two clocks lay out, but for a small clean shop the first-pass cleanup is an afternoon, and for a larger or more political environment it is a few weeks of structured work, still far cheaper than locking in a bloated model for the term. Rushing to sign a number you have not cleaned up is how you lock in years of overpayment to save a few days now.
I built a free AppPoints Estimator for exactly this. Put your designed role model in and it gives you a number you can hold up next to the vendor quote, so you walk into the conversation knowing what a clean model should cost.
Frequently Asked Questions
Why is my MAS AppPoint quote so high?
Usually because the model you handed over is high, not because the vendor padded it. A reseller prices the user list, apps, and tiers you give them. Hand over a decade of accumulated access and they price the accumulated access, accurately. Rule out real errors first, then attack the model, not the quote.
Should I negotiate the MAS quote or fix the model?
Fix the model. Arguing the quote wins at best a one-time single-digit discount on a number that is structurally too big. Fixing the model, with lower tiers, Self-Service, Concurrent for low-peak roles, and reclaimed dormant accounts, produces a reduction that repeats every year of the committed term.
What inflates a MAS quote?
Tiers assigned by title instead of work, no Self-Service for pure requesters, Authorized by default for people who log in rarely, dev and test sized as full production, and dormant accounts still holding licenses. All of it is the input you provided, not the vendor's doing.
How do I lower the quote without just negotiating price?
Rebuild the input role by role: assign tiers by work, move requesters to Self-Service, decide Authorized or Concurrent on real peak, reclaim dormant accounts, and price non-production on its own terms. Then hand the clean model back and ask for a fresh quote. The gap between the two quotes is the real prize.
Quick Maximo questions are always free. Reach out on LinkedIn. I never charge for chatting.