Oracle PULA vs ULA: what the certification you give up is worth.

A PULA grants the same unlimited deployment right as a ULA, minus the certification that freezes your count and opens support to a reset. On our models, that trade only pays when deployment grows at double digit rates for a decade.

What is the difference between an Oracle PULA and a ULA?

An Oracle ULA gives you unlimited deployment of a named list of programs for a fixed term, commonly three years, and ends in certification. A PULA, or Perpetual Unlimited License Agreement, grants the same deployment right with no end date, so certification never happens. Until the last day of the ULA term, the two agreements behave identically.

Everything that separates them flows from that one event. A ULA includes the right to stop counting, freeze your entitlement and reset the support base once. A PULA hands that right back to Oracle for a larger entry fee which, if deployment keeps growing, costs less than the second and third ULA terms you would otherwise buy.

What happens on the day a ULA is certified?

At expiry you declare the quantities you have deployed, and those quantities become perpetual licenses. Three things change on that single day:

• The count freezes. Your entitlement becomes a number instead of an open right, and every later deployment is measured against it.
• The unlimited fee stops. You are no longer paying for a deployment right you may not be using.
• The support base becomes reviewable. You now hold discrete licenses, so you can stop paying support on programs you no longer run, within the limits of Oracle's support policies described further down.

A PULA removes all three. The rest of this page puts a price on that removal, so you can choose between the two instruments on their numbers.

How do you price the certification you give up with a PULA?

Treat the certification clause as an option, because it has a value whether or not anyone writes it down. When you sign a PULA you sell that option to Oracle, and the saving against future ULA terms is what Oracle pays you for it. Value it the way you would value an insurance policy.

Estimate the probability that your Oracle footprint is materially smaller in seven years, then multiply it by the support you would stop paying. Four inputs go into the calculation:

• Probability the footprint shrinks. Take it from the capacity plan and the application road map, and ignore the sales narrative.
• Support you could shed. The portion of the annual support stream attached to programs you would retire.
• Years of benefit. Every year from the reset date to the end of your planning horizon.
• The comparison. The perpetual premium over the ULA, expressed as a single number today.

Why is deployment free until the count freezes?

Inside the ULA term, every extra processor of a named program costs nothing. The first unit above the certified number is a new purchase at whatever discount you can win under pressure, with support attached for good. Oracle's price list of September 15, 2026 sets that at $47,500 per Database Enterprise Edition processor, plus $10,450 a year in support.

A ULA therefore rewards deploying hard before the freeze and discipline afterwards. A PULA removes both behaviors. That makes it simpler to run, and more expensive from the moment your growth stops.

How do you compare PULA and ULA costs over ten years?

Build the model over ten years, because the comparison Oracle presents almost always stops at the point where the perpetual route still looks cheaper. Four inputs decide the result:

• Deployment growth on the named programs. Compound annual growth in processors or users, taken from funded capacity plans.
• The support base and its escalator. The starting stream and the annual uplift, typically 3 to 4 percent, though Oracle's drafts often propose more.
• The reset you would achieve. The share of support you could stop paying after a disciplined certification.
• The probability of a structural event. A divestiture, a migration off Oracle, or a large acquisition inside the horizon.

A worked ten year example with flat deployment

Say Oracle quotes a three year ULA at $4,000,000 and, the same week, a PULA for the same programs at $6,400,000, or 1.6 times the ULA. Assume support at 22 percent of each fee, the price list ratio, rising 3 percent a year. Deployment stays flat, and at certification you drop programs carrying 25 percent of the ULA support.

The PULA costs $10,294,984 more over the decade. Most of the gap comes from support priced off a larger fee, and the reset adds $1,842,055 to it. Even if you shed nothing at certification, the ULA route stays $8,452,928 cheaper, because flat deployment never needs a second term.

Which instrument wins at each growth rate?

Growth changes the picture, because a growing footprint forces a second ULA term and a second fee. The table sets out how the two routes compare across a ten year horizon:

As a rule, the perpetual route needs compound growth of roughly 12 to 18 percent a year on the named programs, sustained for a decade, before it beats two disciplined ULA cycles. Few enterprise Oracle footprints grow that way, and those that do are usually midway through a database consolidation or carry a funded acquisition pipeline built on Oracle.

Is a PULA the right choice for your company?

Only if you can answer yes to five questions before you look at a price. Ask them of your own plans, with finance and corporate development in the room:

1. Is the growth funded? Point at approved capital, named projects and dates. An aspiration does not count as growth.
2. Is the growth on these programs? Database growth does not justify unlimited middleware, and the reverse holds just as firmly.
3. Could you certify if you had to? If your inventory cannot produce a count that would survive an Oracle review today, the ULA exit you are relying on is theoretical.
4. Is a divestiture plausible within seven years? Corporate development will not promise anything, but they will describe the shape of the portfolio.
5. Would you buy the perpetual right at twice the price? If not, you are buying comfort instead of economics.

Five clear yes answers is the only pattern that justifies a perpetual agreement. Four yes answers and one doubt point to a ULA with a well drafted certification clause, and three lean the same way, since any doubt favors the instrument you can reverse. Two or fewer point elsewhere, usually to a straight purchase of what you run.

That last case is the question our ULA decision guide settles before the choice of instrument arises at all.

How does the answer change with your situation?

• Flat or shrinking deployment. The ULA wins clearly. Support resets once and then runs flat on a fixed count, while a perpetual stream keeps compounding.
• A database consolidation under way. The growth the perpetual route needs can be real here, provided the program is funded, dated and runs on the named programs.
• Acquisitions with Oracle as the standard. This is the second profile where a PULA can win. Check how each acquired company joins the entity schedule before you compare fees, because an entity outside the schedule is outside the unlimited right.
• A likely divestiture. The ULA wins, because certified licenses can go with the business you sell. Our note on the Oracle assignment clause covers the mechanics.
• A migration off Oracle. The perpetual support stream keeps running on programs you are leaving, while a certified ULA allows you to drop them.

Which contract clauses carry the risk on a PULA and on a ULA?

Different clauses carry the risk on each instrument, and negotiating a perpetual agreement as though it were a ULA is the most common drafting error we correct. On a ULA the certification clause is the product. A PULA has no certification to protect, so its negotiable value sits in the support base and the assignment clause.

Read the assignment language against your own corporate development pipeline. The conversion of cloud deployments into processor counts follows Oracle's authorized cloud policy, so name the averaging period in the clause itself.

Both instruments leave the named program list, the entity and territory schedule, the virtualization counting rules and Oracle's audit rights fully in force. Scope discipline matters equally on either route. The counting standard for certification is set out in our complete ULA guide.

What do Oracle's support policies do to the reset?

They limit it, so plan the reset around whole programs. Oracle's Software Technical Support Policies state that if a subset of licenses on a single order is terminated, support on the remaining licenses is repriced at Oracle's list support price minus the standard discount. On a heavily discounted order, that repricing can erase the saving.

The same policies require every license in a license set, meaning all your licenses of a program with its options, to sit at one support level. Retiring a whole program meets that rule, but repricing still applies to whatever remains on the same order.

A certified ULA usually lands on one order. Before you certify, ask how the certified quantities will be split across orders, and model the repricing before you count the saving.

Is ULA certification a risk worth paying Oracle to remove?

No. Certification is the single moment in the Oracle relationship when the customer decides what the number will be. Selling that moment for a lower price on perpetual deployment is the most expensive trade in Oracle licensing, and the only one in the deal that no proposal puts a price on.

Why we reject the advice to buy your way out of certification

The common advice, from Oracle and from most resellers, is that a PULA is simply a better ULA. It removes certification risk and ends counting for good, so any large and growing Oracle customer should prefer it. The reasoning fails at its first step, because it treats certification as a hazard when it is an asset you own.

The pitch persuades because the risk is immediate and the value is deferred. Certification is hard work: the count has to hold up, virtualized environments are hard to measure, and a weak inventory invites an argument. That is a project management problem with a known solution, while a surrendered option cannot be bought back at any price you would accept.

Why does reversibility favor the ULA?

Because the conversion runs one way. A ULA can become a PULA at any renewal, and Oracle will welcome that conversation. Across the perpetual agreements we have reviewed since 2020, no customer moved back to a timed agreement without paying for the privilege.

When two options are close on the numbers, the one you can reverse is worth more than the model shows.

A reversible choice keeps your ability to be wrong cheaply, and that should settle most borderline cases. The mechanics of leaving a perpetual agreement are covered in our PULA exit guide.

When does Oracle put a PULA on the table?

The perpetual instrument tends to arrive at three predictable moments:

• After an audit finding. The PULA is offered as a way to close the finding and the counting in one signature.
• Ahead of a known acquisition. Unlimited deployment for good sounds like the answer to an uncertain integration.
• At a support renewal the customer has already called unaffordable. The perpetual deal is presented as the way to stabilize cost.

Each is a moment when you feel exposed and Oracle is offering certainty, which is exactly when a ten year model is worth building. The instrument itself is described on our Oracle PULA page, and a four axis comparison in our guide to the two unlimited license models.

What will the Oracle account team say about a PULA, and how should you answer?

Expect the PULA to be sold as relief from a problem. These are the lines we hear most often, with replies that keep the discussion on your numbers:

• "A PULA removes certification risk for good." Ask Oracle to show the certification right you would give up as a priced line in the proposal. If it will not, value it yourself with the option calculation above.
• "Your growth makes unlimited the obvious choice." Ask which funded projects and dates the growth figure comes from. If the answer is a sales forecast, model flat deployment alongside it.
• "This price is only available this quarter." Ask for a three year ULA quote for the same programs in the same week. A PULA price means little without the ULA price it replaces.
• "You can always certify out of a PULA later." Ask for that route in the contract at entry, with its price. Without a clause, leaving later is a negotiation Oracle controls.
• "The PULA settles the audit finding as well." Ask for the finding to be settled on its own numbers first, then judge the instrument on your growth plans.

What have we seen in recent Oracle PULA and ULA decisions?

In 2024 and 2025 we advised on roughly 35 to 45 Oracle unlimited agreement decisions, around 40 in total. Oracle led with the perpetual instrument in two of every three proposals, yet it suited the customer in fewer than one in three.

• Flat plans under a growth pitch. In the majority of files, the customer's own capacity plan showed flat or declining deployment within five years.
• Short horizons. Oracle's comparisons covered three to five years. On a ten year view the ranking reversed in most of the models we rebuilt.
• Growth taken from ambition. Growth rates came from targets with no funded plan behind them.
• Option value left out. Not one proposal priced the certification event as something the customer was giving up.

The lesson from these files is to build the ten year model yourself before Oracle prices either instrument. More Oracle analysis sits in our Oracle knowledge hub.

When should you start the PULA or ULA decision?

Two years before your current ULA expires, because certification preparation needs that long and the decision needs a count you can trust. Work to this timeline:

What to do next

1. Write a ten year deployment trajectory. Cover the named programs only, and source every figure from funded capacity plans.
2. Price both instruments across the full horizon. Include support escalation, and give the certification option its own line in the model.
3. Answer the five fit questions. Treat any doubt as a vote for the reversible instrument.
4. Negotiate the clauses that matter for your route. Certification terms on a ULA; the support base, uplift cap and assignment on a perpetual agreement.
5. Protect the exit you chose. If a ULA fits, schedule certification preparation two years before expiry. If a PULA fits, negotiate a defined conversion path at entry.
6. Get a second view before you sign. Our Oracle advisory team builds the ten year comparison with you, for a fixed fee.

Frequently asked questions