Updated September 2026
Cloud repatriation is moving workloads from public cloud to infrastructure you control: owned hardware in a colocation facility, rented bare metal, or a hybrid of dedicated and cloud. It is not a rejection of cloud. It is a decision about which workloads belong where, made on total cost over three to five years rather than on this month's bill.
At seven figures of annual cloud spend, the destination is not a few cabinets. It is a wholesale colocation lease at several hundred kilowatts to several megawatts, a bare metal fleet, or both. The two things cloud prices worst are constant utilization and high egress. If your workloads have both, the gap is wide. If they are bursty and light on data transfer, it is narrow or negative, and staying on cloud is the right answer. Which side you are on is arithmetic, not philosophy, and the repatriation case has to beat optimized cloud, not list-price cloud.
Send your annual cloud spend, the main workloads and what is driving the question. It comes to the principal directly, and you'll hear back within 24 hours, including whether the analysis is worth running for you.
Spending less than $50,000 a month on cloud, or already decided to move? You may not need the analysis. Send the requirement and we place it: bare metal or colocation quoted across the providers that fit, at once, paid by the provider you choose, at no cost to you. Send the requirement and you'll hear back within 24 hours.
What Is Cloud Repatriation, and Why Is It Back?
Cloud repatriation is the migration of applications, databases, compute and storage from AWS, Microsoft Azure, Google Cloud or a multi-cloud setup to dedicated infrastructure. In practice that means one of three things: owned hardware in a colocation facility or on-premises, rented bare metal servers, or a hybrid where steady workloads move and elastic ones stay. The on-premises version is rarer than it was, because most companies would rather rent space in a professional facility than run a server room, but the economics are the same.
It came back for a simple reason. Cloud is priced for elasticity. You pay a premium for the ability to scale down that a 24/7 workload never uses, and you pay again for every gigabyte that leaves the provider's network. Companies that were small enough for that premium to be an IT line item are now large enough for it to be a board question.
Cloud Repatriation Examples
Dropbox
Saved about $75 million over two years by moving most of its storage off AWS, as disclosed in its IPO filing.
37signals
Left the cloud in 2023 and published the numbers, projecting savings of more than $1 million a year.
AWS RTB Fabric
Launched in October 2025 to discount egress by up to 80 percent for real-time bidding, because adtech companies were leaving over it.
And the largest repatriation motion in 2026 is AI. Companies renting GPU capacity from cloud providers and neoclouds are moving to owned rack-scale hardware in colocation, because a GB300 NVL72 rack running training or steady inference at about 140 kW is the most expensive thing to rent and the most predictable thing to own. That is a different scale of decision from moving a web tier: it is multi-megawatt, liquid-cooled, and it needs a facility that can serve the density. Our AI and GPU colocation page covers the facility side, and the GPU rental guide shows what renting costs against owning. This page covers whether to make the move at all.
Cloud Repatriation Trends in 2026
AI leads the move
Steady GPU workloads are the most expensive thing to rent and the most predictable thing to own, so AI training and inference have become a central part of the repatriation decision. See neocloud providers for the rental side.
Leaving got cheaper
Since 2024, AWS, Google Cloud and Azure all waive egress fees on the data you take with you when you leave, which removed one of the strongest reasons to stay.
Providers discount to keep you
Products like AWS RTB Fabric show cloud providers cutting prices where customers were leaving. A documented alternative is now real negotiating leverage.
Hybrid is the end state
Few companies leave cloud entirely. The steady core moves to dedicated infrastructure and the variable edge stays, connected privately. See hybrid cloud colocation.
Cloud vs Colocation vs Bare Metal: The Four Paths Compared
Every repatriation decision resolves into one of four scenarios. This is the comparison most companies build in their heads without writing down, and it is worth writing down, because the differences in speed and capital are larger than the differences in cost.
| Scenario | What it is | Time to production | Capital required | P&L treatment | Best for |
|---|---|---|---|---|---|
| Stay on cloud, optimized | Reserved instances, savings plans, rightsizing, egress renegotiation | Weeks | None | Opex, fully variable | Bursty workloads, small footprints, teams with no infrastructure capacity |
| Bare metal | Rented dedicated servers, no hypervisor, no capital | Weeks to three months | None | Opex at a lower rate | Steady compute that needs to move fast, or a first step before colocation |
| Colocation, owned hardware | You buy servers and lease space and power. At scale, a wholesale lease of 500 kW and up with committed power and a ramp schedule | Nine to fifteen months | Hardware amortized over three to five years | Capex plus depreciation; opex for the facility | Large steady footprints and AI clusters where three-year economics matter more than speed |
| Hybrid | Steady workloads on dedicated, variable ones on cloud, private interconnect between them | Follows the dedicated path chosen | Depends on the dedicated path | Mixed | Most companies above $50,000 a month. This is where the analysis usually lands. |
How Does Each Scenario Show Up in the P&L?
This is the question finance teams ask and infrastructure teams rarely answer, and it matters more than the headline saving.
Cloud today
All operating expense, fully variable, and you own nothing at the end.
Bare metal
Still operating expense, at a lower rate. EBITDA improves without capital outlay, and nothing lands on the balance sheet.
Colocation with owned hardware
A large share of the cost moves to capital expenditure and depreciation. EBITDA improves more, because the recurring cost drops further, but there is cash up front and a depreciating asset to carry.
Hybrid
Steady workloads on whichever dedicated path fits, variable ones on cloud, so the P&L shows a mix.
Two things follow. For an adtech, SaaS or data company, infrastructure usually sits in cost of revenue rather than below the line, so a saving improves gross margin directly, which is a better story for investors than an operating expense reduction. And the choice between owning hardware and renting it is a financing decision, separate from the infrastructure decision. Hardware leasing is standard, and it lets you take the colocation path without the capex if that is what the balance sheet prefers.
How Metro Colo Advisory Works on This
Most companies at this scale have three options: run the analysis internally, hand it to a large consultancy, or go straight to quotes. Each has a problem. Internal teams have never priced dedicated infrastructure and tend to leave out the operations and platform costs, which overstates the saving by a third. Large consultancies charge for the name and staff it with associates who have never negotiated a colocation lease. Quotes price a configuration you have not decided on yet, and each vendor prices its own product.
What we do instead
A fixed-fee engagement, four to six weeks, that produces the scenario-by-scenario P&L this page describes. Every cost counted, priced from operator contracts the practice has visibility into rather than published rates, with an editable model your finance team keeps. The fee is the same whether the answer is move, stay, or somewhere in between. That is what keeps the recommendation clean.
Why the practice is positioned for this
The placement side has been in front of more than fifteen operators in the last ninety days with live requirements. That is where the dedicated-side pricing in the model comes from. It reflects what a facility will actually sign, not what its website says, and that difference is usually larger than the margin the whole decision turns on.
What you get
The four scenarios as a P&L over three to five years with cash timing. The movable list and the stay list, workload by workload. What egress actually costs you and what it would cost on dedicated infrastructure. A recommendation with the migration sequence and the risks per workload. And the model, so when your spend or cloud pricing changes, you rerun it rather than starting over.
What it costs
Engagements run from $50,000 to $195,000 depending on scope, half on signature and half on delivery. Against seven figures of annual cloud spend, the fee is what it costs to know the number before the decision rather than after. When the answer is move, the placement side takes the requirement to operators and brings back competitive proposals for wholesale colocation, bare metal or both. That part is paid by the operator, not by you.
Cloud Egress Fees: What Do They Actually Cost?
Egress is what a cloud provider charges to move data out of its network. Ingress is free. Egress is priced per gigabyte, and for a business that sends a lot of data outward it is often the largest single line on the bill, and the one that disappears most completely on dedicated infrastructure. AWS egress fees, for example, start at 9 cents per gigabyte on its published US pricing and step down with volume, with Google Cloud and Azure in the same range.
The egress line grows fastest as a business scales, because it tracks traffic rather than infrastructure. The workloads where it dominates are the ones that answer requests: real-time bidding, ad serving, content delivery, API platforms, anything that returns data to the outside world at volume. AWS's RTB Fabric product, launched in October 2025, is the clearest evidence of how much this matters. It cuts egress by up to 80 percent for real-time bidding traffic, but only when both sides of the exchange are on AWS, and not in every region. Google and Microsoft have not announced an equivalent. If your egress bill is large and you are not on AWS, or you are on AWS but your counterparties are not, that line is a strong candidate to move.
Repatriation From AWS, Azure and Google Cloud: The Exit Waivers
The cost of getting data out used to be one of the strongest reasons to stay. That changed in 2024. All three large cloud providers now waive the egress fees on data you move out when you leave, following the direction of the European Data Act, and all three apply it worldwide. The conditions differ, and they matter for a phased repatriation.
| Provider | Exit waiver | The conditions |
|---|---|---|
| AWSMarch 2024 | Data transfer out to the internet credited when you move off AWS | Request it through AWS support first. Built for moving all your data out; a partial move needs support's approval for credits. Transfer through CloudFront, Direct Connect, Snow Family and Global Accelerator is not covered |
| Google CloudJanuary 2024 | Data transfer fees waived when migrating off Google Cloud | Case-by-case approval, and you must move all your data out and end the relationship within 60 days of approval |
| Microsoft AzureMarch 2024 | Free data transfer out to the internet when leaving Azure | For customers moving their data out to another provider or an on-premises data center, by request |
Sources: DCD on the AWS waiver; Network World; UK CMA cloud market investigation, egress appendix. Terms can change; confirm with your provider before you plan around them.
Cloud vs On-Premise Cost: How Do You Calculate TCO?
Total cost of ownership is the comparison over three to five years, not year one. Whether the question is cloud vs colocation, cloud vs on-premises, or colo vs cloud for a single workload, the method is the same. Dedicated infrastructure loses year one to migration cost and, if you buy hardware, capital outlay. It wins years two through five if the workload is steady. The crossover is usually somewhere in year two, and the cumulative gap by year five is what the decision rests on.
The cloud side of the model is the bill by workload, with an optimized-cloud line to make the comparison fair. The dedicated side has to carry everything a cloud bill hides:
Hardware
Amortized over three to five years, with a refresh cycle built in.
Facility
Colocation rent per kilowatt or bare metal rental, power, cross-connects, remote hands. Primary-market colocation averaged about $196 per kW per month in CBRE's 2025 survey, and rates rose again in the first half of 2026.
Interconnect
Private links back to whatever stays on cloud, plus transit and peering for the traffic that used to ride the cloud provider's network.
Operations
Whoever runs the hardware, usually a managed service provider rather than new hires, and usually the line companies forget.
Platform layer
Whatever replaces the managed services you use today: databases, load balancers, Kubernetes, observability. Each is a product on cloud and a project on dedicated infrastructure.
Migration
The one-time cost of moving, including the months both environments run in parallel.
Risk buffer
Capacity headroom, redundancy, and the cost of a hardware failure when the replacement takes eight weeks.
Bare Metal or Colocation: Which Dedicated Path?
This is the decision inside the decision, and it is mostly about speed and capital.
Bare metal: faster, no capital
Bare metal is a dedicated physical server you rent from a provider: no hypervisor, no other tenants, no capital outlay. Against cloud it is usually a 25 to 40 percent saving on steady compute, and it needs no capital to test, which is why most companies should run it first. Provisioning takes days to weeks because the hardware already exists. It costs more per month than owning over three years, but it lands the saving in a quarter rather than a year. For a company that needs a result before a fiscal year end, it is usually the only path that gets there.
Colocation: lower cost, slower to reach
Owned hardware in a colocation facility is the lower-cost path over three years and the slower one: procurement, delivery, racking and burn-in run nine to fifteen months from decision to production. The saving is larger because you own the asset, and the facility contract is negotiable in ways a bare metal rental is not. For a large steady footprint where the three-year number matters more than the first-quarter number, it is the right answer.
Many companies do both: bare metal for the first wave, colocation for the second once the workload profile is proven on dedicated infrastructure. That sequence gets the saving early and the lower cost later, and it turns the migration from one bet into two smaller ones.
Which Workloads Should Move, and Which Should Stay?
The distinction is utilization and data movement, and it is worth applying workload by workload rather than to the bill as a whole.
Strong candidates to move
AI training and steady inference on GPU clusters. Rented GPU capacity is the highest-cost line in cloud, and a cluster at consistent utilization is the clearest repatriation case there is. At rack scale this means owned GB300 or Rubin hardware in a liquid-cooled wholesale facility. Workloads running 24/7 at consistent usage: databases, data warehouses, core application tiers, batch on a fixed schedule. Anything with high outbound transfer: bidding, serving, content delivery, high-volume APIs. Storage-heavy applications where the bill is mostly capacity you never release. Data residency or compliance requirements where physical control is an advantage.
Better left on cloud
Extreme variability: anything spiking ten times or more against baseline. Anything under roughly $20,000 a month in isolation, where the overhead of dedicated infrastructure exceeds the saving. Applications built on proprietary managed services that would need re-platforming. True global distribution across many regions at once. Anything expected to change shape in the next twelve months.
The result is almost always hybrid. The steady core moves, the variable edge stays, and private interconnect keeps them connected without touching the public internet. Our hybrid cloud colocation guide covers the architecture and the private links; this page covers the decision.
A Worked Example at Enterprise Scale
The numbers below are illustrative, built for a data platform company at $1.2 million a month on cloud. They are the shape of the analysis rather than any client's figures.
| Monthly line | Cloud today | Hybrid, year one | Hybrid, year three |
|---|---|---|---|
| Cloud compute and storage | $720,000 | $310,000 | $290,000 |
| Cloud egress | $280,000 | $60,000 | $55,000 |
| Cloud managed services | $200,000 | $120,000 | $110,000 |
| Bare metal rental | 0 | $210,000 | $40,000 |
| Colocation facility | 0 | $25,000 | $95,000 |
| Hardware amortization | 0 | $20,000 | $130,000 |
| Interconnect and transit | 0 | $35,000 | $40,000 |
| Operations and platform | 0 | $70,000 | $75,000 |
| Migration, amortized | 0 | $60,000 | 0 |
| Monthly total | $1,200,000 | $910,000 | $835,000 |
| Annual saving | $3.5 million | $4.4 million |
Illustrative model, Metro Colo Advisory, September 2026. Free to cite with a link to this page.
Egress is the largest reduction, and it arrives immediately
From $280,000 a month to $60,000 in year one, before anything else has finished moving.
Bare metal first, colocation second
Year one uses bare metal for speed while the colocation build runs in parallel, which is why bare metal is high in year one and low in year three.
The line in-house models leave out
Operations and platform, at $70,000 to $75,000 a month. Without it this analysis would show $5.3 million a year instead of $4.4 million, and the company would find the difference the hard way.
The colocation footprint in year three is roughly 2 to 3 MW, which is a wholesale lease rather than a cabinet agreement, with committed power, a ramp schedule and the contract terms that come with it.
Annual spend, the main workloads, which clouds, and what is driving the question. If the numbers do not justify the engagement, we will say so.
What Does a Cloud Exit Strategy Actually Look Like?
The sequence matters as much as the destination. A cloud exit strategy is the order in which workloads move, the parallel-running period for each, and the point at which the cloud commitment is reduced. Done in the wrong order, it costs more than staying.
Workload assessment, weeks one to three
Billing export by workload, utilization data, and a conversation with the people who run the systems. This produces the movable list, the stay list, and the kilowatt requirement for the dedicated side.
Path selection and facility or provider search, weeks three to six
Bare metal, colocation or both. If colocation, two or three facilities that fit the connectivity, density and compliance profile, with competitive proposals. Our site selection guide covers the facility criteria.
Contracting, weeks six to ten
Facility or provider contract through legal review, with escalation, minimum commitment and renewal terms negotiated before signature rather than discovered after. The data center lease guide covers what to look for.
Hardware, weeks eight to sixteen for bare metal, months four to twelve for owned
Bare metal provisions in days once contracted. Owned hardware is a procurement cycle.
Migration in waves, three to six months for live workloads
Non-critical first, production once stability is confirmed, with both environments running in parallel for each wave. The data center migration guide covers the mechanics.
Cloud commitment reduction
Reserved instances and committed-use contracts are reduced as workloads leave, not before. Cutting the commitment early is the most common expensive mistake.
What This Page Cannot Tell You
It cannot tell you which of your workloads are steady, because that depends on utilization data you have and we do not. It cannot tell you what your egress line is, because that is on your bill. It cannot tell you what dedicated infrastructure would cost for your footprint in your markets, because that depends on operator pricing that is contracted, not published. And it cannot tell you where your company lands inside the 30 to 50 percent range, because that depends on all three. Those four things are the analysis. The framework above is what the analysis uses.
Frequently Asked Questions
Should we move AI workloads off cloud?
If they run at steady utilization, almost certainly. Rented GPU capacity is the most expensive line in cloud, and a training or inference cluster running around the clock is the clearest repatriation case there is. The move is to owned rack-scale hardware in a liquid-cooled colocation facility, which at about 140 kW per rack is a wholesale lease, not a cabinet agreement. The economics are usually decisive. The hard part is finding a facility that can serve the density on your timeline, which is a short list and getting shorter.
When does cloud repatriation to colocation make financial sense?
When three things are true: the workload runs at steady utilization, roughly 60 percent or more around the clock; the spend is large enough that the overhead of dedicated infrastructure is small against the saving, usually $50,000 a month or more; and the comparison still wins against optimized cloud, not list price. High egress makes the case stronger. If any of the three is missing, optimizing cloud is usually the better answer.
Will AWS waive egress fees if we leave?
Yes, on the data you move out when you leave. Since March 2024 AWS credits data transfer out to the internet for customers moving off AWS, by request through AWS support. It is built for moving all your data; a partial move needs support's approval. Google Cloud and Azure offer similar exit waivers. The ongoing egress on workloads that stay on cloud is still charged.
What are the cost differences between AI cloud and GPU colocation?
Renting GPUs from a cloud provider or neocloud is priced per GPU-hour and suits bursty or short-term work. Owning the hardware in colocation means paying for the servers up front or by lease, plus colocation at roughly $150 to $250 per kW per month depending on market and density. For a cluster running at steady utilization for two years or more, owning usually costs well under renting. See GPU rental for rental prices and AI and GPU colocation for the facility side.
How long does cloud repatriation take?
Bare metal: weeks to provision, three to six months to migrate live workloads. Colocation with owned hardware: nine to fifteen months from decision to production, because hardware procurement and facility build-out run in series. Most enterprise moves take six to eighteen months end to end, with the saving arriving in waves as each workload lands. The analysis that precedes the decision takes four to six weeks.
What is the difference between bare metal and colocation?
Bare metal is a dedicated server you rent; the provider owns the hardware and the facility. Colocation is space and power you rent for hardware you own. Bare metal is faster and needs no capital; colocation is cheaper over three years and gives you control of the asset. Many companies start on bare metal and move to colocation once the workload is proven on dedicated infrastructure.
Do egress fees really justify leaving cloud?
For some workloads, egress alone does. At 5 to 9 cents per gigabyte, a business sending 500 terabytes a month outward pays $25,000 to $45,000 in egress before anything else. On dedicated infrastructure that line falls to a transit commit and a port fee. Real-time bidding, ad serving, content delivery and high-volume APIs are the workloads where egress decides it, and AWS's RTB Fabric product exists because it was losing exactly those customers.
Should we just renegotiate with our cloud provider instead?
Yes, and the analysis is what gives you the leverage. A documented cost for the alternative is what makes a cloud provider move on price. Some companies run the analysis, take it to their account team, and get 15 to 25 percent off without moving anything. That is a fine outcome. The comparison has to be against optimized cloud, not list price, or the repatriation case is overstated.
What does cloud repatriation do to EBITDA?
Bare metal improves EBITDA directly, because the cost is still operating expense at a lower rate. Colocation with owned hardware improves it more, because the recurring cost drops further, but part of the cost moves to depreciation below EBITDA and there is a capital outlay. For companies where infrastructure sits in cost of revenue, the saving improves gross margin as well. Hardware leasing lets you choose the treatment the balance sheet prefers.
Who runs the hardware after we move?
Usually a managed service provider rather than new hires, with the colocation operator handling remote hands and hardware replacement. It is a cost line in the model, typically 5 to 10 percent of dedicated infrastructure spend, and the line in-house comparisons most often leave out. Bare metal providers include hardware maintenance in the rental, which is part of why it is the simpler first step.
Colocation vs cloud: which costs less?
For bursty, variable workloads, yes. For steady workloads running 24/7, almost never over three years. The comparison usually favors dedicated infrastructure by 30 to 50 percent on the movable share once utilization is above roughly 60 percent, because cloud is priced for the ability to scale down that a steady workload never uses. Colocation is the practical form of on-premises for most companies, since it provides the facility without the burden of running one.
Is cloud repatriation a one-way decision?
No. The hybrid architecture most companies land on keeps cloud connectivity through private interconnect, and workloads can move back if their profile changes. Bare metal contracts are typically one to three years; colocation contracts three to five. Neither is permanent, and the analysis includes what it would cost to reverse if the assumptions turn out wrong.
Find Out Where Your Company Lands
If the question is a P&L across scenarios at seven figures of annual spend or more, send us your spend, the main workloads and what is driving the question. It comes to the principal directly, and you'll hear back within 24 hours. If your footprint is smaller, start with the calculator for a first read on your numbers.
Same fee whether the answer is move, stay, or somewhere in between.
Deciding and modeling: the cloud vs colocation calculator, the colocation pricing guide, the hybrid cloud colocation guide and FinOps consulting.
The dedicated side: bare metal, wholesale colocation, GPU rental and AI and GPU colocation.
Executing the move: the migration guide, the site selection guide, the data center lease guide.
Markets and providers: compare all providers, or the guides to Northern Virginia, Dallas, Chicago, Los Angeles, Phoenix, Denver and the New York metro. For how placement works, see how it works and why companies use an independent advisor.
