Oracle support fees are fixed at the ULA level and do not rise with the number you certify. Because there is also no fee to certify, every defensible processor you add to the count converts to permanent entitlement at zero marginal cost, which makes a higher count free value rather than added expense.
By the Meridian advisory team, former Oracle LMS and GLAS licensing analysts. Updated 4 June 2026.
No. This is the single most important fact in a ULA exit, and it is the one most often misunderstood. Your Oracle support fee was set when the ULA was signed and it continues at that level after certification regardless of the count you declare. Certifying 800 processors and certifying 1,600 processors produce exactly the same support invoice. There is no per license uplift, no recalculation, no penalty for a larger number. The fee is anchored to the agreement, not to the deployment you certify out of it.
That single fact reframes the whole exercise. If support does not move and there is no charge to certify, then the marginal cost of certifying one more defensible processor is zero. You are not buying those licenses. You already paid for unlimited deployment when you signed the ULA. Certification is simply the moment you convert that unlimited right into a fixed, permanent entitlement, and the size of that entitlement is yours to claim up to whatever you can legitimately defend.
There is no fee for certification and support stays flat whatever you declare. So each additional processor you can defend is permanent entitlement at no extra cost. The count is not a bill to minimise. It is value to maximise.
The belief that a higher count costs more usually comes from instinct rather than the contract. Teams reason that more licenses must mean more support, because that is how a normal purchase works. In a standard license buy, quantity and support are linked. In a ULA exit they are not, because the support level was already fixed and the unlimited right was already paid for. Carrying the normal purchasing instinct into a certification leads teams to hold the number down to avoid a cost increase that cannot occur.
The result is a quiet, expensive form of caution. A team certifies a conservative number, leaves real deployments out, and feels prudent for doing so. In reality it has surrendered permanent entitlement for no benefit at all. We treat this directly in the maximization mistakes that backfire, because timid counting is the most common of them.
Once you accept that the marginal cost is zero, the logic of the exit becomes one directional. Certifying a deployment you are entitled to count is free. Failing to certify it costs the full purchase price of buying that capacity again later, if you ever need it. There is no symmetric downside to a larger defensible count and a large, real downside to a smaller one. The discipline that follows is simple: count everything you can defend, build the evidence to defend it, and never hold a number down out of misplaced thrift.
Take an indicative case. An organisation runs a database ULA and expects to certify around 1,000 processor licenses. Independent work across virtualization, disaster recovery, non production, and a qualifying cloud estate brings the defensible count to 1,500. The support fee is identical at both numbers. Two years after the exit the organisation grows into 300 of those extra processors. Having certified 1,500, that growth is already licensed and costs nothing. Had it certified 1,000, it would now buy 300 processors at full price. The figures are indicative, but they show the shape: the larger count was free on the way out and saved a real purchase on the way back.
| Decision at certification | Support fee | Cost of later growth into 300 proc |
|---|---|---|
| Certify 1,000 (conservative) | Flat at ULA level | Buy 300 proc at full price |
| Certify 1,500 (defensible max) | Identical, no change | Already entitled, nothing to buy |
The exact numbers depend on your estate and your products, so treat this as the pattern rather than a forecast for your case.
A higher count is only free value when every processor in it is defensible. Maximisation is not padding the number with deployments you cannot evidence. A count you cannot stand behind becomes audit exposure in the years after the exit, when the evidence file is your defence. The work is to find every legitimate deployment and prove it, not to invent volume. Done properly, the larger number is both bigger and safer, because each line is supported by discovery output, server lists, and documented methodology. We explain how to build that runway in pre building the next three years of growth.
If the marginal cost of a defensible processor is zero, then the certification window is the cheapest capacity you will ever acquire, and the planning should reflect that. Start early enough to discover the whole estate, read the cloud and partitioning clauses to know what qualifies, deploy what you are entitled to deploy before the clock stops, and document as you go. The structured version of that work sits in the maximization project plan, and the full method lives in our pillar, the ULA deployment maximization guide.
A higher certified count does not raise your support fee, because the fee is fixed at the ULA level and certification carries no charge. That makes every defensible processor free perpetual value, claimed once, at the moment you certify. The mistake is to treat the count as a cost to minimise. The discipline is to treat it as value to maximise, backed by evidence, because the only entitlement you keep is the one you defend on the way out.
No. Oracle support fees are fixed at the level set in the ULA and do not move with the number you certify. Whether you declare a smaller count or a larger one, the annual support invoice is the same, which means every additional processor you can defend is free perpetual entitlement.
Because there is no fee to certify and support stays flat, the marginal cost of each defensible processor you add to the count is zero. The licenses become permanent entitlement at no extra charge, while failing to count them forces you to buy the same capacity later at full price.
Book a confidential assessment and we will measure the defensible count you are entitled to certify before the window closes.