The State of Enterprise Software Discounts 2026: the discount ritual is ending, and 800 contracts show how fast
For two decades enterprise software pricing followed a stable ritual, high list prices, deep negotiated discounts, and a final number that rewarded customers who pushed: that ritual is ending, and the benchmark of 800 contracts, 200 per vendor in comparable deal size bands, shows average achieved discounts falling at Oracle, SAP, Microsoft, and Salesforce every year since 2019. Four different shapes, one direction: down, and the decline is structural rather than cyclical.
Prepared by Redress Compliance · August 8, 2026 · Cross vendor advisory. Based on 800 contracts benchmarked 2019 to 2026, 200 per vendor in comparable bands.
Benchmark at a glance. Across 800 benchmarked contracts, 200 per vendor, average achieved discounts have fallen since 2019: Oracle from 57 to 44 percent, Microsoft down 13 points, SAP near zero on perpetual S/4HANA, and the 2019 discount is not coming back.
Check your own quote against these bands on VendorBenchmark (free, no signup), see the discount benchmarks for all eleven vendors, or bring the number to the negotiation team.
Executive summary
Four curves, four shapes, one direction.
SAP shows the steepest collapse, from 60 to 75 percent on large perpetual ECC deals to near zero on perpetual S/4HANA, discounting all but withdrawn to make RISE subscriptions the only economically rational path.
Oracle eroded by policy, roughly two points a year from 57 percent in 2019 to 44 today, with sharp list increases added from June 2026; Microsoft held stable through 2024 then dropped roughly thirteen points after the programmatic Enterprise Agreement price level discounts were removed.
And Salesforce pairs gentle percentage erosion with repeated list increases, so net cost rises even where the percentage survives.
The compounding is the finding buyers miss.
Discount declines compounded with list price increases.
So a customer holding its 2019 percentage still took a double digit net price rise, because a stable discount applied to a raised list is a raise wearing a discount's clothes: the Salesforce curve understates its own truth for exactly this reason.
And the honest measure everywhere is achieved net price against applicable list at signature, normalized to comparable annual contract value bands, which is what the panel measures.
The decline is structural, and the mechanism explains why it continues.
In the perpetual era a discount was a one time concession on a one time fee.
In the subscription era a discount given once is given every year of the term and usually every term after, making discount discipline a board level revenue quality metric, since recurring revenue at protected prices is what the market values these vendors on.
Sales compensation was redesigned to defend list, the concession authority moved from the account team to the deal desk, and the Broadcom VMware demonstration proved the market rewards repricing a locked in base.
The negotiation value moved, and the buyers who know it still win.
The value shifted from the discount percentage to contractual protection, alternatives, and timing: buyers with a credible, documented alternative developed twelve months out landed 8 to 15 points deeper than otherwise identical deals without one, well leveraged deals still beat the average curves.
And the account team across the table has lost the authority to concede more than the floor, which makes the preparation aimed at the deal desk's constraints rather than the rep's goodwill.
The 2019 discount is not coming back, and the playbook that assumed it is the one to retire.
The four curves, 2019 against 2026
| Vendor | 2019 average | 2026 average | The shape of the decline |
|---|---|---|---|
| SAP | 60 to 75 percent on large perpetual ECC | Near zero on perpetual S/4HANA | Collapse, through the S/4HANA to RISE transition |
| Oracle | 57 percent | 44 percent | Steady policy erosion, two points a year, plus June 2026 list rises |
| Microsoft | Stable through 2024 | Down roughly 13 points since 2025 | A step change when programmatic EA discounts were removed |
| Salesforce | Gradual erosion begins | Modest percentage decline, higher net cost | A stable looking percentage on a repeatedly raised list |
Each shape has its lesson.
SAP falls fastest because the product carrying the discount changed underneath the customer, the perpetual withdrawal steering everyone to subscription; Oracle erodes without any single announcement to point to, which is exactly why nobody fought it.
Microsoft's step change shows what removing a program does overnight; and Salesforce proves the percentage is the wrong metric entirely, since a surviving discount on a raised list still raises the net.
The benchmark measures achieved net price against applicable list at signature, the only comparison that catches all four.
Why the discount is disappearing, mechanically
- Subscription economics: a discount given once is given every year of the term and every term after, a permanent reduction in the recurring asset the share price is built on.
- Discount discipline as a board metric: revenue quality now means recurring revenue at protected prices, and finance teams govern the floor accordingly.
- The compensation redesign: sales plans defend list, and the concession authority moved from the account team to the deal desk, where your relationship does not reach.
- The Broadcom demonstration: repricing a locked in base was rewarded by the market, and every vendor board watched it happen.
- The compounding with list: declines stacked on price rises, so holding the 2019 percentage still meant a double digit net increase.
The renewal negotiation timing playbook
The protection, alternative, and timing levers that replaced the discount conversation, worked end to end.
Get the white paper →The buyer response, aimed at the deal desk
The negotiation value moved from the percentage to three places the deal desk still concedes: contractual protection, the capped uplift, the co terminus date, and the reduction rights that decide what any rate is worth in year three.
The alternative, credible and documented twelve months out, worth 8 to 15 points on otherwise identical deals because the desk prices retention risk even when the rep cannot; and the timing, the calendar that reaches the desk before its quarter closes rather than after its floor is set.
Well leveraged deals still beat every curve in the panel, which is the point of the response: the averages describe the unprepared.
The current band placement method runs in the real discount benchmark, the lever ranking in the negotiation leverage report, the list movements the discounts push against in the price increase index, and the demonstration that started the era in the Broadcom VMware pricing report.
- Percentile standing for your exact deal size and industry, from real closed transactions
- Scenario simulation before the call: test alternative terms and see the financial impact of each
- A negotiation playbook, talking points, and a two page executive brief on day one
What the 800 contracts show
Across the 800 contracts our team benchmarked between 2019 and 2026, 200 per vendor in comparable deal size bands, measuring achieved net price against applicable list at signature:
Deeper on deals with a credible documented alternative developed twelve months out.
The average discount fell at all four vendors, every year since 2019, four shapes, one direction.
The methodology's honesty is its value: averages smooth wide individual variation, well leveraged deals still beat the lines, deals whose scope changed too much between cycles are excluded, and the panel promises what any single deal can achieve exactly nothing, because bands, leverage.
And timing move outcomes by double digits.
The budget conclusion is the report's bluntest sentence, the 2019 discount is not coming back: the next budget plans against the current curve and the compounding list, the negotiation invests in the protection, alternative, and timing levers the deal desk still concedes.
And the discount percentage retires as the success metric it never deserved to be.
Where the common advice on the vanishing discount is wrong
The common advice is to negotiate harder on the percentage and wait for the vendor to restore the discount levels of 2019. We disagree. Across the 800 contracts in this panel the average achieved discount fell at Oracle, SAP, Microsoft and Salesforce every year since 2019, and the deal desks now concede protection, alternatives and timing far more readily than points off a list they control. The buyers who did best measured net price per unit against the prior term at signature, capped the uplift, and brought one costed alternative. The percentage retired as the success metric; the net price did not.
Discount bands by vendor in 2026
The four curves above are the trend. The table below is the position today: the band a prepared buyer closes in at each vendor, stated at the spend tier it applies to, from closed agreements in the Redress engagement file and the VendorBenchmark deal base.
Each row links to the full benchmark page with the metric, tier and term behind it.
| Vendor | Band from closed agreements | What moves it | Detail |
|---|---|---|---|
| Microsoft EA | Median E5 discount 28 percent; 10 to 18 percent at Tier A rising to 28 to 38 percent at Tier D | User tier, Copilot attach, Azure commitment | Full benchmark |
| Oracle technology | Net cost per unit per year by product family; support uplift target 0 to 3 percent | Entitlement baseline, a funded alternative, the May close | Full benchmark |
| Oracle Java | List 15 dollars per employee per month under 1,000 employees, stepping to 5.25 at 40,000 to 49,999 | Employee count definition and band position | Full benchmark |
| SAP | 20 to 28 percent below 1M ACV, 28 to 38 at 1M to 5M, 45 to 60 above 10M on perpetual; RISE 10 to 50 by tier | Spend tier above all; uplift cap and FUE count | Full benchmark |
| Salesforce | Sales Cloud 20 to 40 percent, Service Cloud 15 to 35, Agentforce 5 to 15, SELA 35 to 55 | Cloud mix and estate size | Full benchmark |
| ServiceNow | 8 to 12 percent median under 500K ACV, 27 at 2M to 5M, 30 to 40 above 5M | Spend tier, a written competitive proposal, Q4 close | Full benchmark |
| Broadcom VMware VCF | 185 to 275 dollars per core per year against a 350 dollar list; 45 to 55 percent off above 15,000 cores | Core tier, term length, a costed exit | Full benchmark |
| AWS EDP | 5 to 20 percent off most spend on top of Savings Plans | Commitment size, ramp shape, Marketplace scope | Full benchmark |
| Google Cloud | Committed use discounts plus a negotiated agreement percentage that stacks on top | Resource versus spend commitment, portability | Full benchmark |
| Cisco EA | Discount ladder 20 to 35 percent by commit tier; median first quote gap 24 percent | Quarter timing, suite composition, true forward | Full benchmark |
| IBM ELA | 19 to 34 percent below the opening best and final, median 27 | ILMT reconciliation, VPC sizing, a Red Hat carve out | Full benchmark |
| OpenAI ChatGPT Enterprise | 10 to 18 percent at 500 to 2,000 seats rising to 33 to 40 above 15,000 seats on 36 months | Seat band, term, credible competition | Full benchmark |
A quote inside the band is a starting point, not a result. The spread between a first quote and a benchmarked signature runs 10 to 20 points at most vendors, and the spend tier moves the band further than anything said at the table.
Your first five moves
- Measure net price against list at signature, the only metric that catches the Salesforce shape.
- Plan the budget against the current curve, because the 2019 discount is not coming back.
- Develop the documented alternative twelve months out, the 8 to 15 points that still works.
- Aim the preparation at the deal desk's constraints, since the account team lost the authority to concede.
- Move the negotiation to protection, alternatives, and timing, where the value went. The cost optimization practice runs the benchmark with you.
Frequently asked questions
Are enterprise software discounts really disappearing?
Across 800 benchmarked contracts, 200 per vendor in comparable deal size bands, average achieved discounts fell at Oracle, SAP, Microsoft, and Salesforce every year since 2019: SAP collapsed from 60 to 75 percent on perpetual ECC to near zero on perpetual S/4HANA, Oracle eroded from 57 to 44 percent.
Microsoft dropped roughly thirteen points after 2025, and Salesforce's stable looking percentage sits on a repeatedly raised list.
Why did SAP discounts collapse?
Because the product carrying the discount changed underneath the customer: perpetual S/4HANA discounting has been all but withdrawn to make RISE subscriptions the only economically rational path, the steepest decline in the panel.
The discount did not shrink on the same product; the discounted product was retired, which is the pattern's most aggressive form.
What removed Microsoft's EA discounts?
The programmatic Enterprise Agreement price level discounts were removed from 2025, producing a roughly thirteen point fall in two years after stability through 2024, a step change rather than an erosion.
The volume tiers that once rewarded scale automatically now reward nothing, which moves the entire Microsoft conversation to the negotiated layer and the structural levers.
Why do vendors defend list price so hard now?
Subscription economics changed what a discount costs: once, it was a one time concession on a one time fee; now it recurs every year of the term and usually every term after, a permanent reduction in the recurring revenue asset the share price is built on.
Discount discipline became a board level revenue quality metric, compensation was redesigned to defend list, and the Broadcom demonstration proved the market rewards repricing a locked in base.
Can buyers still get good software deals?
Yes, and the panel proves it: well leveraged deals beat every average curve, and buyers with a credible documented alternative developed twelve months out landed 8 to 15 points deeper than identical deals without one.
The value moved from the discount percentage to contractual protection, alternatives, and timing, aimed at the deal desk that now holds the authority, and the averages describe the unprepared.
How should budgets adjust to the vanishing discount?
Plan against the current curve and the compounding list, not the 2019 memory: a customer holding its old percentage still took a double digit net rise because the list beneath it climbed, so the honest budget metric is achieved net price at signature.
The negotiation investment moves to the capped uplift, the co terminus date, the reduction rights, the twelve month alternative, and the calendar, the levers that still concede.
Which vendor still gives the deepest discount in 2026?
Measured off list, SAP perpetual above 10 million of annual contract value at 45 to 60 percent, Salesforce SELA at 35 to 55 and Broadcom VCF above 15,000 cores at 45 to 55 lead the table, but each sits on a list the vendor controls.
Net price per unit against the prior term is the only comparison that holds across vendors.
How do I benchmark my own quote against this report?
Run the quote through the free VendorBenchmark decoder, which normalizes it to the metric and spend tier the band is measured in, then read the vendor benchmark page linked in the table for what moves a deal from the bottom of the band to the top.
