The renewal quote averaged 1.5 to 2.2 times the fee against 5 to 15 percent growth

An Oracle ULA ends one of two ways. You renew and pay again, or you certify your usage and walk away with a fixed perpetual count. The technology rarely forces the answer, and the renewal models rarely match the deployment curve they claim to price.

Prepared by Redress Compliance · August 15, 2026 · Oracle advisory. 30 to 40 ULA certifications advised, 2024 to 2026.

• Model both paths against real demand before replying to any renewal proposal, because a quote answered without a model is answered on the vendor's growth assumptions.
• Read the certification clause before anything else, since it defines the metric, the date, and what cloud capacity counts, per the cloud counting brief.
• Start at least nine months out so the measurement, the freeze, and the letter all happen on your calendar rather than Oracle's.
• Count everything genuinely deployed, because undercounting cost 10 to 20 percent of real need and cannot be revised after filing.
• Freeze deployment near the count date so the estate you measure is the estate you certify, without late changes muddying the evidence.
• Expect retention activity in the final two quarters and treat it as commercial rather than informational, as the certification pressure brief sets out.

A ULA renewal proposal is a forecast wearing the clothes of an invoice. Its price implies a rate of deployment growth, because that is the only thing that makes unlimited deployment worth paying for twice, and in the exits we advised that implied rate sat well above what the estates were actually doing. Quotes averaged 1.5 to 2.2 times the original fee against final year growth of 5 to 15 percent. Both numbers are knowable in advance. Only one of them is usually in the room.

This is why the modeling exercise, rather than the negotiation, is the decisive act. Seven in ten clients who priced both paths against real demand chose to certify and walk, and then held flat or grew on perpetual licenses without incident. They did not out argue anyone. They simply put the renewal fee plus its support next to the licenses they would actually need at their real discount, and the comparison answered itself. The clients who did not model it were left assessing a quote against a feeling about growth, which is a contest the seller wins by default.

The exit signals are worth reading as a pattern rather than a checklist, because they describe the same underlying situation from three angles. A footprint that grew sharply and then flattened has already extracted the value unlimited deployment offers. A quote that outruns forecast demand is charging for headroom that will sit unused. A cloud or engine migration is quietly shrinking the base the next term would cover. Any one of them makes the second term a purchase of optionality nobody will exercise.

Where buyers do lose money at the exit, it is almost never in the decision and almost always in the count. Undercounting left entitlements 10 to 20 percent below real need in our engagements, and unlike the renewal decision that error is irreversible: the missing capacity is repurchased later at full price with no leverage. So the sequence is model first, then measure exhaustively, then freeze, then file. Decide commercially and count carefully. What life looks like afterwards is in the exit strategy guide, the second term arithmetic in the renewal tactics, and the wider library in the Oracle practice.

Watch the briefing · 4:30How to Negotiate an Oracle ULA: No Price List, Just Your Business CaseModel the breakeven yourself, and treat the renewal trap as the decision that determines whether the ULA was worth signing.

Across 30 to 40 Oracle ULA certifications advised, the renewal case and the deployment data rarely agreed:

The patterns: renewal models assuming double digit growth against actual final year growth of 5 to 15 percent, retention pressure concentrated in the last two quarters, and undercounting as the single largest error made by buyers who had already made the right decision.

The buyer side move is to model before you reply. The wider library sits in the Oracle practice.

When should you walk away from a ULA instead of renewing?

When deployment has stabilized and the renewal price outruns real demand for the next three years. Three signals point to the exit: a footprint that grew sharply early and then flattened, a quote that assumes growth your forecast does not contain, and a cloud or platform shift that is shrinking the on premise Oracle estate.

How large are ULA renewal quotes?

In the exits we advised, renewal quotes averaged 1.5 to 2.2 times the original ULA fee, while real deployment growth over the term was far smaller. Deployment growth in the final ULA year ran 5 to 15 percent in most estates, not the double digit growth the renewal models assumed.

How often does modeling both paths lead to an exit?

Roughly seven in ten clients who modeled both paths certified and walked away, then held flat or grew on their perpetual licenses. The modeling itself is the intervention: the decision is commercial rather than technical, and it changes once both numbers exist side by side.

What is the biggest error buyers make at exit?

Undercounting eligible deployments, which left perpetual entitlements 10 to 20 percent below real need in our engagements. That gap has to be repurchased later at full price, which is why the count deserves more preparation than the decision itself.

What does a clean exit require?

An accurate measurement, a deployment freeze near the count date, and a defensible certification letter, all run on your calendar rather than Oracle's. Start at least nine months before the certification date, and read the certification clause before anything else.

Why does Oracle apply pressure in the final two quarters?

Because Oracle has a direct commercial interest in the renewal: it extends and grows the support annuity, while certification caps it. Expect retention activity late in the term, and treat it as a predictable commercial move rather than new information about your estate.

Is the exit decision technical or commercial?

Commercial. The technology rarely forces the answer; the numbers and the calendar decide it. Model both paths against real demand before replying to any renewal proposal, because a quote answered without a model is answered on the vendor's assumptions.