For more than a decade, VMware was the safe choice for running servers on-premises. Licences were paid, the platform was familiar, and the hardware refresh was the only decision that came up every few years.
That calm has gone. Since Broadcom acquired VMware, licensing has moved to subscription bundles and many customers have seen their renewal costs rise sharply. Renewal time has become a decision point: renew for another term, or use the moment to rethink where your workloads should live.
86%
of enterprises surveyed are actively reducing their VMware footprint
Source: CloudBolt research, January 2026
59%
have seen VMware prices rise by more than 25%
Source: CloudBolt research, January 2026
72%
of the workloads they're migrating are heading to public cloud infrastructure
Source: CloudBolt research, January 2026
The survey covered 302 IT decision-makers at North American organisations with 1,000 or more employees. Australian organisations of every size are having the same conversation at renewal time.
The renewal is the wrong question
It's tempting to treat this as a licensing problem: negotiate the renewal, or find the cheapest place to run the same virtual machines. But the virtual machines are not the point. The applications and data on them are, along with the people who use them and the risk they carry.
Moving to the public cloud is an opportunity to ask, for every workload: does this still need to exist, and if so, what's the best way to run it now?
A lift and shift moves your problems to someone else's data centre. A rationalised migration leaves them behind.
Why lift and shift alone falls short
Copying every virtual machine to the cloud exactly as it is can be fast, but it carries over everything that wasn't working: oversized servers, unpatched operating systems, flat networks and applications nobody uses any more. In the cloud, oversized means overpaying every month. Microsoft's own guidance notes that rehosting does not resolve existing performance, reliability or architectural issues.
A better approach: rationalise, then migrate
From VMware to a modern cloud platform
- Step 1
Assess
Inventory every VM, dependency, licence and cost. Measure real utilisation, not provisioned size.
- Step 2
Rationalise
Decide per workload: retire, replace with SaaS, rehost, refactor or rebuild.
- Step 3
Build the landing zone
Identity, networking, security and governance in place before the first workload arrives.
- Step 4
Migrate in waves
Low-risk workloads first, with tested rollback and cut-overs out of hours.
- Step 5
Modernise
Move to managed databases, app services and right-sized compute where it pays.
- Step 6
Manage & optimise
Monitoring, patching, backup and cost reviews through managed services.
Assess the estate
We inventory every virtual machine, its dependencies, licences, utilisation and cost. Real utilisation, not provisioned size, is what drives the cloud design. Most estates have servers sized for peaks that never came.
Rationalise workload by workload
Each workload gets one decision. Some are retired. Some are replaced with a SaaS product that does the job better, such as file servers moving to SharePoint. Some are rehosted quickly. The ones that matter most are refactored or rebuilt to take advantage of managed databases, app services and automatic scaling.
Cloud rationalisation: one decision per workload
Retire
Little or no business value left
Turn it off and stop paying for it
Replace
A SaaS product does the job
e.g. file servers to SharePoint, on-prem apps to their cloud editions
Rehost
Needs to move fast, works as is
Lift to Azure VMs, right-sized from real usage
Refactor
Small changes unlock big savings
Managed databases, app services, containers
Rearchitect / rebuild
Core to the business, holding you back
Redesigned for scale, resilience and cost
Retain
Not ready to move yet
Stays put, connected to the cloud, revisited later
Build a landing zone before the first workload moves
A landing zone is the foundation: identity, networking, security policies, governance and cost management, set up once and applied to everything that follows. Done properly, it means your cloud is secure by default rather than secured after the fact.
Connect users and branch offices with zero trust
Traditional networks trusted anything inside the perimeter. In the cloud there is no perimeter to hide behind. A zero-trust model verifies every request by identity and device health, segments workloads so a problem in one can't spread, and gives staff and branch offices secure access to what they need, and nothing else.
A landing zone built on zero trust
Staff anywhere
Branch offices
Managed devices
Zero-trust access layer
Production workloads
Data & backups
Shared services
Keep it running well
Migration is the beginning. Ongoing managed services monitor performance and security, patch and back up workloads, and review costs every month, so the savings you planned are the savings you keep. Right-sizing and reserved capacity can then cut steady-state costs further.
What changes for the business
- Cost that follows usage: pay for what you use, with reservations and savings plans for the steady base, and no hardware refresh cycle.
- Resilience by design: backups, disaster recovery and multi-zone options that would be expensive to build on-premises.
- Security that's consistent: one identity platform, one policy set and one monitoring view across offices, people and workloads.
- Speed: new environments in hours, not procurement cycles.
- A platform ready for AI: data and applications in a modern cloud are far easier to connect to AI services safely.
A 30-person firm with three ageing VMware hosts
Instead of renewing licences and replacing hosts, the file server moves to SharePoint, line-of-business apps move to their cloud editions, and the one remaining server runs in Azure. The server room is switched off.
A 300-person organisation with 60 VMs across two sites
An assessment finds a fifth of the VMs can be retired. The rest move in waves into a landing zone with zero-trust access for both offices, and disaster recovery is included rather than a separate project.
A 3,000-seat enterprise facing a multi-year VMware renewal
Workloads are rationalised by business unit, with critical systems refactored onto managed services and others rehosted to hit the renewal deadline. Ongoing management keeps the environment patched, monitored and cost-reviewed every month.
When staying put makes sense
The public cloud is not the answer for every workload. Some systems have latency, data sovereignty or licensing constraints that keep them on-premises for now. A good rationalisation identifies those honestly, keeps them connected to the cloud, and revisits them as the constraints change.
The bottom line
The VMware renewal has turned an infrastructure decision into a business one. Organisations that use it to rationalise, modernise and secure their estate end up with a platform that costs less to run, is safer by design and is ready for what comes next. Those that only move the virtual machines end up paying cloud prices for yesterday's problems.
Sources
- New CloudBolt research: 86% of companies actively reducing their VMware footprint (CloudBolt, January 2026)
- Some enterprises are dropping VMware, just not all at once (Network World)
- Cloud rationalization: the 5 Rs (Microsoft Cloud Adoption Framework)
- What is an Azure landing zone? (Microsoft Cloud Adoption Framework)