Cloud migration services that plan each workload's route before anything moves

Our cloud migration services start with a map of what depends on what. No case study of a cloud move is published, so judge us on a discovery first.

Illustration of servers and applications moving from an on-premises data center into three cloud environmentsIllustration of servers and applications moving from an on-premises data center into three cloud environments

CLOUD MIGRATION

What does a move to the cloud involve?

A cloud migration moves servers, applications and their infrastructure off on-premises hardware, or out of another cloud, onto AWS, Azure or Google Cloud, or into a hybrid setup that keeps some systems where they are. It ends when your team has taken over the new environment.

Some workloads need more than a move. When an application has to be refactored or re-architected before it can run in the cloud, that is a legacy modernization project of its own. A large database with schemas to map and records to reconcile is a data migration, a separate service of ours.

Diagram of an application and its server moving from an on-premises rack to public, private and hybrid cloudsDiagram of an application and its server moving from an on-premises rack to public, private and hybrid clouds

MIGRATION ROUTES

Which route does each workload take?

The 7 Rs, and the two we run

We run rehost and replatform. The 7 Rs are the usual names for every route a workload can take, and the assessment gives each workload its route. Most estates end up using more than one.

Rehost (lift and shift)

The workload moves as it is, onto cloud virtual machines, with no change to the code. The provider's own replication tooling does the copying, such as Azure Migrate on Azure or AWS Transform MGN (formerly AWS Application Migration Service) on AWS. A rehost is the right first move when the hardware or the data center contract sets the deadline. It is a poor one when the license bill is the problem, because the same licenses and oversized servers move with it.

Replatform

Replatforming makes a few targeted changes on the way so the application can use managed services, such as Amazon RDS or Azure SQL Database in place of a self-run database, or containers instead of hand-built servers. The core code and architecture stay as they are. The application's own database moves with it.

Relocate

Relocation lifts a whole VMware environment onto the provider's version of the same platform, such as Azure VMware Solution or Amazon Elastic VMware Service. The virtual machines themselves don't change.

Refactor and re-architect

Refactoring changes the code and re-architecting changes the structure, for example splitting a monolith into services. The assessment flags the workloads that need either before they can move.

Repurchase, retire or retain

Repurchase swaps an application for a SaaS product. Retire switches off what nobody uses. Most estates have at least one server nobody can explain, and that is the server that stalls a wave when its owner turns up mid-cutover. Retain keeps a system where it is for now, which is often the right call for one with a few years of support left.

Hybrid setups that keep some systems on-premises

Some systems stay on-premises for a while, and the move then has to keep both sides working as one, with a single identity and a network link between them. The link is usually a site-to-site VPN, or a private line such as Azure ExpressRoute, AWS Direct Connect or Google's Cloud Interconnect. Hybrid identity is part of the Azure work we offer, and it connects your on-premises Active Directory to Microsoft Entra ID through Entra Connect.

Between clouds

A move from one cloud to another takes the same routes. The work is in the managed services that have no exact twin on the other side, and each of those is a small replatform of its own.

Diagram of one workload branching into seven migration strategies, with rehost and replatform markedDiagram of one workload branching into seven migration strategies, with rehost and replatform marked

HOW A MOVE RUNS

How we plan and run a cloud move

A move runs as project delivery after a discovery phase, and we can set up CI/CD for the new environment on the way. If you would rather keep the move in-house, add our cloud and DevOps engineers to your team instead.

  1. STEP 1

    Assessment and dependency map

    We review your setup, release process and team skills, trace which systems call which, and give each workload its route. You get a recommended platform, target architecture and delivery plan.

  2. STEP 2

    Landing zone and target environment

    When your requirements call for a landing zone, it comes first, with account or subscription structure, identity and policy guardrails. The rest is code in reviewed pull requests.

  3. STEP 3

    Waves grouped by dependency

    We recommend waves of systems that depend on each other, and your requirements decide their final shape. Each wave is replicated into the target and tested there before it cuts over.

  4. STEP 4

    Cutover with a way back, then handover

    Every cutover has a rollback or fallback path. When the last wave is live, your team takes over and our engineers' access is removed.

PLATFORMS

Which cloud should the move land on?

Our cloud consulting practice covers AWS, Azure and Google Cloud, and the right one is usually the cloud your team and your licenses already point to. A Microsoft-heavy estate often fits Azure. A team that already runs part of its stack on AWS rarely gains from adding a second provider.

Pick the cloud for the years after the move. A lower compute price on one provider rarely outweighs retraining the team that has to run it, and identity, networking and monitoring are the parts that take longest to relearn.

Google Cloud as the target

Compute Engine takes rehosted virtual machines, copied with Google's Migrate to Virtual Machines. Cloud SQL and GKE are the usual targets for a replatform.

COST

What sets the price of a migration?

Three things set the estimate, and the fourth sets the bill after it. You can narrow the estimate before anyone quotes by listing the workloads you want moved, who owns each one and what each one talks to.

The move itself is a one-off piece of work. The bill that surprises people is the monthly one after the cutover, and the choices that set it are made during the move, while the target is still easy to change.

Number of workloads moved

Every server, application and database needs its own inventory line, target design, move and test.

Rehost or replatform depth

Adopting a managed service on the way, such as a database or a container platform, means one more thing to build, test and cut over.

Dependencies and cutover

Systems that call each other move together, or keep talking across two environments where latency can hurt. The dependency map sets how many waves that takes.

Running costs after cutover

Instances sized for the old hardware, SQL Server and Windows licenses and egress charges on hybrid traffic all show up on the monthly bill. Sizing is reviewed after each wave.

HOW WE WORK

Engagement models and terms

Project delivery

We deliver an agreed scope end to end: plan, team, milestones, QA and release. You make the product decisions and accept each milestone.

Fixed price per milestone after discovery, or time and materials against an estimate.

  • Scope changes go through change requests.
  • On fixed-price projects, a warranty period agreed in the contract, usually 30–60 days.

Discovery phase

A short phase before a build: requirements and backlog, architecture, an estimate and a delivery plan.

Fixed price, usually 2–4 weeks.

  • The outputs stay with you whether or not the build continues with us.

Staff augmentation

Our engineers join your team and work in your process and tools. You manage them; we source, vet and contract them.

Hourly rate per engineer by role and seniority, invoiced monthly.

  • If an engineer isn't the right fit, we replace them within 14 days at no extra cost.
  • Knowledge transfer to the replacement engineer isn't billed.
  • There is no free trial period.
  • 30 days' notice to take an engineer off or end the engagement, with no minimum term beyond it.
  • Out-of-hours on-call only if agreed with you; we set its hours and billing together.
  • After two years, moving an engineer onto your own staff is something we can discuss.

Whichever model you choose

  • We reply to an inquiry from the contact form within 24 hours.
  • We sign an NDA before work starts, and IP passes to you on payment.

Plan the move before you commit to it

Send us the list of systems you run and where you want them to land. We will suggest a first step that fits the size of the move.

FAQ

Questions to settle before a cloud move

Should we modernize the application before we move it?

Move first when the application already runs on a cloud virtual machine as it is and a date is pressing. Modernize first when it can't, because its operating system or runtime is past end of support or it depends on specific hardware. Lifting a broken application gives you a broken application in the cloud, with a monthly bill attached. The assessment shows which case each workload is in, and the refactoring itself runs as legacy modernization.

How long will the move take, and how much downtime should we plan for?

On project delivery a discovery phase comes first, usually 2–4 weeks, and it ends in a delivery plan. The timeline for the move itself comes from that plan, because the number of workloads and the links between them shape the waves. Downtime depends on the workload. A stateless web tier can usually switch over with little interruption, while a database taking constant writes is the usual reason a cutover needs a planned window. With replication running, the final sync is part of that window, and a way back stays real only until the target starts taking writes the source doesn't have.

Can we judge your work before we commit to the whole move?

Yes, in two ways. A discovery phase can be bought on its own, and it ends in an estimate and a delivery plan for your move. And our AWS and Azure engineers have passed a practical task on a real cloud setup, judged on the security and cost of the design. Much of our cloud work sits under client NDAs, some of it on Azure. The discovery terms are in the Engagement models and terms section.

What access will your engineers need to our cloud accounts?

Named individual accounts that you issue in your own IAM or SSO, each limited to what that engineer's role needs. The work happens inside your accounts, so your data stays where you already keep it. Where we handle your data, a DPA is part of the engagement.

Who runs the new environment after the cutover?

Your team, unless you would rather keep us on. Managed cloud services and FinOps are both service lines of ours, and managed services run in CET/CEST business hours by default, with the schedule shifted to match a US team.

Are you an AWS, Azure or Google Cloud partner?

No. We are in no cloud provider's partner program and work as an independent team. Our AWS and Azure engineers are certified.

Can we stop partway through the move?

Yes, on project delivery. A fixed-price project can stop at the end of any milestone, and a time-and-materials one at the end of any bi-weekly demo. You pay for the work done up to that point, IP passes to you on that payment, and we hand over the code, the documentation and the discovery outputs.