Contents
Key takeawaysTips ranked by valueWhat the tier choice is worthScoring bundle fitWhat 40 to 55 renewals showedContract terms that add costA credible exit optionBroadcom lines and repliesRenewal timelineWhat to do nextFAQDecide which workload domains need VCF before Broadcom quotes. Moving a domain to VVF is worth $160 to $220 per core, more than the entire VCF discount in every size band we benchmarked.
- Settle the tier first. In all four size bands, moving a workload domain from VCF to VVF saves more per core than the full discount Broadcom offers on VCF.
- VVF list does not change with size. It stays flat at every size band, at roughly 40 percent of VCF list, and covers compute, storage and basic management.
- Most hosts do not need VCF. Hosts typically use less than half of the VCF feature set, and only some run NSX micro segmentation or SDDC Manager.
- A costed exit still shifts price. In the renewals we benchmarked, a documented Proxmox, Hyper-V or OpenShift pilot moved the discount further than anything else.
- Large buyers gain most. VVF net falls faster with volume than VCF net, so the tier choice compounds as core counts grow.
- Fix the mechanics at signature. The sixteen core minimum, the escalator, the SKU mix and the audit clause are far easier to change before you sign.
Which Broadcom negotiation tips save the most money?
The biggest saving comes from deciding which workload domains need VMware Cloud Foundation (VCF) and moving the rest to vSphere Foundation (VVF). The tier choice is worth more per core than the entire discount Broadcom will give on VCF. The costed exit option comes second, and the contract mechanics third.
Most negotiation advice starts with the discount percentage. We rank the tips by what they are worth and by when they have to be done, because several of them expire the moment Broadcom sends a quote.
| Tip | What it is worth | When it has to be done |
|---|---|---|
| Put every workload domain that does not use VCF features on VVF | More per core than the whole VCF discount, in every size band | Before Broadcom quotes |
| Price a credible exit on a representative workload domain | The largest single effect on the discount | Before the commercial talks open |
| Push VCF net to the low end of your size band | $10 to $130 per core, depending on size | After scope is fixed |
| Cap the annual escalator at 3 percent | The gap to Broadcom's default escalator, compounding each year | At signature |
| Get written exceptions to the sixteen core per CPU minimum | Large on small hosts, edge sites and development clusters | At signature |
| Lock the SKU mix and add a reduction window | Protects every later year from forced tier upgrades | At signature |
The wider picture of Broadcom's licensing changes sits in the Broadcom VMware pillar. This page covers how to run the negotiation itself.
How much is the VVF or VCF choice worth at each size?
Between $160 and $220 per core per year, measured against the net prices we see in each size band. The entire discount Broadcom offers on VCF is worth $10 to $130 per core. The tier wins in all four bands, including the largest, where Broadcom discounts hardest.
What does each bundle include?
VVF covers compute, storage and basic management: vSphere Enterprise Plus, vCenter Standard, VCF Operations and vSAN with 0.25 TiB of capacity per licensed core. VCF adds NSX networking and SDDC Manager, plus HCX, VCF Automation and 1 TiB of vSAN per core. Most workload domains never touch NSX.
VVF list holds flat at $135 per core whatever your size. That is roughly 40 percent of VCF list, which puts VCF list at about $350 per core. Our VCF and VVF feature comparison lists the components line by line.
| Total cores | VVF benchmark net | VCF benchmark net | Tier gap per core | VCF discount available |
|---|---|---|---|---|
| Under 500 | $120 to $130 | $340 to $360 | $220 | $10 |
| 500 to 2,000 | $100 to $115 | $300 to $330 | $200 | $50 |
| 2,000 to 10,000 | $80 to $100 | $260 to $300 | $180 | $90 |
| Over 10,000 | $60 to $80 | $220 to $260 | $160 | $130 |
Read the last column as the most Broadcom gave below the roughly $350 VCF list in that band. The tier gap compares the low end of each net range, VCF against VVF. In every row the gap is larger than the best available discount.
On 2,000 cores the tier decision is worth about $400,000 a year and the whole VCF discount about $100,000. Broadcom prefers to talk about the discount, because it is bounded and it comes up late, when scope is already fixed in the proposal. Score bundle fit first and the discount becomes the second argument.
What does the tier choice look like in a worked example?
Say you run 1,800 cores and Broadcom proposes VCF across all of them. Your host scoring shows that 400 cores sit in domains running NSX, and 1,400 cores run plain vSphere on external storage. The figures below use band midpoints and are illustrative.
| Option | Cores and rate | Annual cost |
|---|---|---|
| Broadcom default: all VCF | 1,800 at $315 | $567,000 |
| Split: VCF where NSX runs | 400 at $350 | $140,000 |
| Split: VVF for the rest | 1,400 at $107.50 | $150,500 |
| Split total | 1,800 cores | $290,500 |
| Saving from the tier choice | $276,500 | |
| Full VCF discount off list in this band | 1,800 at $50 | $90,000 |
We priced the 400 VCF cores at the small band rate on purpose. A smaller VCF order can lose volume pricing, so assume the worst case when you model the split. Even then the tier choice is worth about three times the full discount Broadcom gives on VCF in this size band.
Broadcom VMware Negotiation Brief
Bundle fit scoring, per core benchmarks by size band and the contract terms to ask for at renewal.
Get the white paper →How do you score bundle fit host by host?
Measure which VCF features each host uses and license only those hosts on VCF. The default Broadcom proposal puts every workload under the more expensive tier, and a host by host score is what reverses it.
- Inventory every host. Record total cores, sockets, memory and feature use, since the per core bill is built from that list whether or not you have seen it.
- Score each host against the VCF feature set. Most hosts use less than half of it, and only some need NSX micro segmentation.
- Reserve VCF for the domains that use NSX and SDDC Manager. Put the majority on VVF, which covers compute and storage at the lower price.
- Drop undeployed suites. The Aria, Tanzu and HCX SKUs that arrive inside a bundle and were never installed should not shape your tier.
Which reports show what each host actually uses?
- Broadcom's core counting module. Knowledge base article 313548 links the FoundationCoreAndTiBUsage PowerCLI module. It reads vCenter and reports licensable cores per host with the sixteen core minimum applied, plus vSAN capacity per cluster.
- RVTools. The vHost tab gives sockets and cores per host for every vCenter, which makes a quick cross check on the Broadcom output.
- NSX Manager. The host transport node list shows which clusters are prepared for NSX. Clusters that are not prepared are VVF candidates.
- Distributed firewall rules. A cluster prepared for NSX that only carries the default rule is not using micro segmentation.
- vSAN capacity per cluster. Compare claimed TiB against the 0.25 TiB per core that VVF includes, and price any shortfall both ways.
What have 40 to 55 Broadcom renewals shown about pricing?
Scale earns a better rate on VVF faster than on VCF. In roughly 40 to 55 VMware and Broadcom renewals we benchmarked in 2024 and 2025, VVF net fell 44 percent across the bands, from about $125 per core to $70. VCF net fell 31 percent, from about $350 to $240.
So the buyers with the most negotiating weight gain the most from the tier choice. The gap per core narrows as you grow, from $220 down to $160, but the core count grows far faster.
- On 400 cores. The tier is worth about $88,000 a year.
- On 10,000 cores. It is worth about $1.6 million a year.
How deep do Broadcom discounts go?
Discounts ran 10 to 30 percent off list for VVF in mid sized environments and 20 to 40 percent off list for VCF in large ones. Broadcom protects price harder than legacy VMware did, and the band tightened sharply after the acquisition closed in November 2023. Our VCF discount band benchmarks break this down further.
None of this retires the exit argument. A documented exit option still saved 20 to 40 percent, and a costed Proxmox, Hyper-V or OpenShift pilot was what got the discount to shift at all.
Why we would not open with the discount percentage
The usual advice is to benchmark the discount and push for the deepest percentage you can get. We think that puts the effort in the wrong place, because the percentage is applied to a scope Broadcom chose.
At 40 percent off the roughly $350 list, VCF still costs $210 per core, well above the $135 VVF list on hosts that never run NSX. Settle the tier in your own inventory before Broadcom quotes, then argue the discount with an alternative already priced.
Broadcom is the least pleasant negotiation in enterprise software right now, and one of the most winnable, because the gap between a prepared customer and an unprepared one is so large.
Our session Negotiating with Broadcom covers the wider negotiation, from the opening quote to signature.
Which contract terms raise the bill after the tier is set?
Three mechanics add cost regardless of which bundle you choose: the sixteen core minimum, the annual escalator and the SKU mix. Each is negotiable at signature and much harder to change at renewal.
How does the sixteen core per CPU minimum affect small hosts?
Broadcom applies a sixteen core minimum per CPU to every subscription. A two socket host running two eight core CPUs still pays for thirty two cores. Small hosts, edge sites and development environments take the worst of it.
Say you run 30 edge hosts, each with two eight core CPUs. That is 480 physical cores, licensed as 960, and at the $135 VVF list rate the extra 480 cores cost $64,800 a year. Ask for the exception in writing, or consolidate small hosts before the renewal count, using our core licensing guide for the counting rule.
How much does the annual escalator cost over a term?
Broadcom defaults to 6 to 8 percent a year. Negotiate it to 3 percent and decline the multi year version, because the year one price is the only one the agreement fixes.
| Year | At 7 percent a year | At 3 percent a year |
|---|---|---|
| Year 1 | $290,500 | $290,500 |
| Year 2 | $310,835 | $299,215 |
| Year 3 | $332,593 | $308,191 |
| Total | $933,928 | $897,906 |
The 7 percent rate is the middle of Broadcom's default range. The cap saves about $36,000 over three years in this example, and more with every year the agreement runs. See our guide to price cap negotiation for cap wording.
Why lock the SKU mix and add a reduction window?
Lock the SKU mix so a future bundle change cannot force a tier upgrade. Then negotiate a scheduled reduction window at the renewal anniversary. Where the subscription renews annually, that cuts both ways. Each renewal gives Broadcom a chance to reprice, and gives you a point to reduce if the right is written in.
What else belongs in the order form?
- Named core minimum exceptions. List the small hosts and edge sites to be licensed at physical cores, by host name.
- A reduction notice date. State how many days before each anniversary you must give notice, so the window cannot lapse unnoticed.
- An audit clause based on annual self attestation. Surprise inspection allows compliance pressure to run alongside the commercial talks.
- Perpetual rights on record. Where legacy perpetual licenses still exist, confirm them in writing and do not trade them away, since they cannot be recreated.
What makes a VMware exit option credible to Broadcom?
Broadcom takes an exit seriously when you name the alternative and show a costed pilot. A stated intention to leave does not move the price, and account teams can tell a priced plan from a threat.
- Pick a real alternative. Proxmox, Hyper-V, OpenShift and KVM are the credible options. Pick the one your team could actually run in production.
- Pilot on a representative workload domain. A lab does not price a migration, and Broadcom knows the difference.
- Cost the whole move. Include migration labor, retraining, backup and monitoring changes, and the months of dual running.
Our VMware exit plan guide and the Proxmox and VMware comparison go through the costing in detail.
What will the Broadcom account team say, and how should you answer?
Expect the conversation to be steered toward the discount and toward VCF everywhere. These are the lines we hear most often, with replies that keep the talks on scope.
- "VCF is the strategic platform, and VVF is for smaller customers." Reply: quote VVF for the domains our scoring shows do not use NSX or SDDC Manager, and we will compare.
- "The better discount depends on standardizing on VCF." Reply: our per domain model shows the tier is worth more than that discount, so show us both prices by domain.
- "Pricing is set centrally and the partner has no room." Reply: then we will size the order to what we use and finish costing the alternative.
- "Sign for multiple years and we will hold the price." Reply: hold year one pricing with a capped escalator, a locked SKU mix and a reduction window, or we renew for one year.
- "This offer expires at quarter end." Reply: our renewal date has not moved, and we will sign when the scope is right. Our note on Broadcom quarter end timing explains when that pressure is real.
When should each step happen before a Broadcom renewal?
Start about 12 months out. The tier work has to be finished before the first quote, and the exit pilot needs months to produce a real cost.
| Months before renewal | What to do |
|---|---|
| 12 | Run the host inventory, score bundle fit, list perpetual rights and undeployed SKUs |
| 6 | Finish the per domain tier split, start the exit pilot on a representative domain, ask for quotes by domain |
| 3 | Put the priced alternative on the table, negotiate the discount, core minimum exceptions and escalator cap |
| 1 | Check the order form wording: SKU lock, reduction window, audit clause, then sign |
What to do next
- Inventory and score every host. Do it before Broadcom quotes anything, since the tier is settled in your data and not at the table.
- Move every domain that does not use NSX to VVF. Model it per domain against the VCF net price in your size band.
- Cost an exit pilot on a representative workload domain. A documented alternative shifts the discount further than anything else.
- Negotiate the sixteen core minimum for small hosts and edge sites. Get the exception written into the order at signature.
- Cap the escalator at 3 percent. Decline the multi year version and lock the SKU mix alongside it.
- Get help where it pays. Our negotiation practice builds the bundle fit score and prices the alternative for you.
Frequently asked questions
How does VVF compare to VCF on cost?
At list, VVF costs $135 per core a year and VCF about $350. The difference buys NSX, SDDC Manager, HCX, VCF Automation and a larger vSAN allowance per core. If a workload domain uses none of those, the extra spend buys nothing it needs.
Is the tier decision really worth more than the discount?
Yes. Above 10,000 cores the tier is still worth $160 per core against $130 for the full VCF discount, and that is the band where the two are closest. Below 500 cores the tier gap is $220 per core while VCF net sits within $10 of list.
What does that mean in annual cash?
Multiply the per core gap by the cores you can move to VVF. On 2,000 cores that comes to about $400,000 a year, against roughly $100,000 for the whole VCF discount. On 10,000 cores the tier is worth about $1.6 million a year, which is why large buyers should start the host scoring early.
Which workloads actually need VCF?
Workload domains that use NSX micro segmentation or overlay networking, and those run through SDDC Manager. Storage heavy clusters are a pricing call: VCF includes 1 TiB of vSAN per core against 0.25 TiB with VVF, so compare VCF with VVF plus extra vSAN capacity before you decide.
What discounts are actually available on Broadcom VMware?
In the renewals we benchmarked, 10 to 30 percent off list for VVF in mid sized environments, and 20 to 40 percent off list for VCF in large ones. Ask Broadcom for the net price per core for each bundle, which is easier to compare with benchmarks than a percentage off a list you may not have seen.
Does the exit option still matter?
Yes, because a priced alternative is what gets the discount to shift at all. It does nothing for the tier, which you settle in your own inventory. Finish the host scoring before the quote, and have the pilot costed before the commercial talks start.
What is the sixteen core minimum?
Broadcom licenses at least 16 cores for every physical CPU, whatever the chip has. Broadcom's own PowerCLI counting module applies that minimum host by host, so run it before renewal to see which small hosts are inflating the count and whether consolidating them is cheaper than licensing them.
What escalator does Broadcom default to?
6 to 8 percent a year. Ask for a 3 percent cap written into the order form and tied to the same SKUs and quantities, so a later change in Broadcom's packaging cannot reset the base price the cap applies to.
Why does discount depth differ between the bundles?
VVF net falls 44 percent from the smallest band to the largest, against 31 percent for VCF. Our reading is that Broadcom holds VCF pricing harder because it wants customers on the full stack, and prices VVF to keep volume from leaving.
When should the tier be decided?
Before Broadcom sends a quote. Start the host inventory about 12 months before the renewal date, so the per domain split is finished and you can ask for quotes by domain. A quote that already covers every core as VCF is much harder to reopen than one you shaped from the start.