Azure · SQL Server · Snowflake · Databricks · Fabric
Azure landing zones, datacenter and workload migration, SQL Server and data platform modernization, and disaster recovery you have actually tested. Fixed scope, fixed price, and cutovers scheduled outside business hours.
What we do
BitMerge is not a managed service provider. It's a specialist engineering practice you bring in for the work that has a deadline, a rollback plan, and consequences if it goes wrong.
Migrate
Datacenter and workload migration into Azure. VMware and Hyper-V estates, production SaaS platforms, and post-acquisition consolidation. Dependency mapping, wave planning, a pilot before the fleet, and a rollback plan you would actually be willing to use.
Modernize
SQL Server version upgrades and moves to Azure SQL. Warehouse and analytics work across Snowflake, Databricks, and Microsoft Fabric. Power Platform and reporting layered on top, with performance baselines taken before and after.
Recover
Disaster recovery designed against the SLA, RTO, and RPO you have actually committed to — not the aspirational ones. Backup architecture, replication, written runbooks, and a failover test that demonstrates the numbers hold.
Where to start
A Cloud & Data Platform Assessment. Read-only, two to three weeks, and it ends in a written migration plan with a cost model attached.
Fixed feeA scoped project from the roadmap — a landing zone, a migration wave, a SQL modernization, a DR build. One number, one statement of work.
Fixed feeAn optional Guardian plan: cost and posture review, backup verification, and a DR test on a schedule, so the environment doesn't quietly drift.
Monthly · 30 days' noticeStart here
A read-only review of your estate that ends in a document you can budget against: a migration wave plan, a cost model, a prioritized risk register, and an honest gap analysis of what your disaster recovery would actually do under load.
Two to three weeks depending on estate size. Nothing in your environment is changed. The fee is credited against any project you book within sixty days.
What gets reviewed
Projects
Pricing depends on estate size and condition, so every engagement starts with a short conversation and a written quote. A fifty-VM estate that's documented and a fifty-VM estate that isn't are not the same job.
Management groups · Hub-spoke · Policy · IaC
The foundation everything else sits on. Management group and subscription design, hub-and-spoke networking, hybrid connectivity, identity integration, governance and policy, cost controls and tagging, all defined as infrastructure-as-code rather than portal clicks nobody can reproduce.
VMware · Hyper-V · Azure · Colocation
Moving real estates: VMware and Hyper-V workloads, production SaaS platforms, and consolidation after an acquisition. Dependency mapping, wave planning, rehost-versus-replatform decisions, a pilot wave, cutover in a scheduled window, and post-migration right-sizing once you have real telemetry.
This is the core of the practice, and the work with the deepest track record behind it.
SQL Server · Azure SQL · HA/DR
Version upgrades off end-of-support builds, and moves to Azure SQL Managed Instance or SQL on Azure VMs. Compatibility assessment, performance baselines captured before and after, high-availability design, and a tested rollback path.
Snowflake · Databricks · Fabric · Power BI
Warehouse migration and platform build-out across Snowflake, Databricks, and Microsoft Fabric. Ingestion pipelines, workload migration off legacy warehouses, cost and performance tuning, and Power Platform and Power BI on top for the people who actually consume the data.
Replication · Backup · Runbooks · Failover testing
DR designed against the SLA, RTO, and RPO in your customer contracts. Replication and backup architecture, recovery runbooks and methods of procedure written so someone other than the author can execute them, and a documented failover test that proves the numbers.
Most DR plans have never been run. That's the problem worth paying to fix.
Entra ID · Active Directory · Exchange · M365
Hybrid Active Directory and Entra ID design, Exchange and Exchange Online migration, tenant-to-tenant consolidation after a merger, conditional access and privileged access review, and cleanup of the sync errors nobody has looked at in two years.
PowerShell · Bash · CI/CD · Monitoring
Replacing manual runbook steps with scripted ones. Build and deployment automation, configuration drift control, monitoring and alerting that reflects what actually matters, and the operational documentation that makes a small team able to run a large estate.
Plesk · Exchange · Mail deliverability
For agencies, hosts, and anyone running multi-tenant web and mail infrastructure. Platform migration and consolidation, mail deliverability and authentication, spam and filtering architecture, and virtualization of estates still running on physical hardware.
Built on hands-on operation of a hosting estate at roughly 300 domains, 150 websites, and 6,000 mailboxes.
Ongoing
Scheduled engineering and review on a monthly agreement, with thirty days' notice either way.
Monthly cloud cost and posture review, backup and replication verification, patch and update oversight, a change log, and a quarterly written report.
Quote on requestEverything in Essentials, plus infrastructure-as-code and policy management, monitoring and alerting maintenance, capacity and cost forecasting, and one documented DR failover test per quarter.
Quote on requestA fractional cloud and data architect. Roadmap ownership, licensing and reservation strategy, technical oversight of your existing MSP or internal team, project design, and vendor evaluation.
Quote on requestPrepaid blocks of ten, twenty, or forty hours for ad-hoc engineering, drawn down in scheduled sessions. Twelve-month expiry.
Quote on requestGuardian plans cover engineering, review, and scheduled change work. They do not include end-user helpdesk, hardware support, or emergency response.
Method
A written statement of work with a defined outcome and a number attached. No hourly meters, no scope-creep invoices, no surprises at the end of the month.
Migrations and cutovers run evenings and weekends. That's the standard, not a premium add-on. Your users sign in Monday to a system that works.
Nothing moves at scale until a representative wave has moved, been validated, and been rolled back at least once in testing. Migrations fail on the assumptions nobody checked.
Every engagement ends with as-built documentation and, where relevant, an executable runbook. It's yours outright — no proprietary tooling, and nothing that makes leaving expensive.
Scope
Most of this industry will tell you it does everything. Here is where BitMerge stops, so you can find the right partner for the rest.
If you already have an MSP or an internal IT team, that's usually the ideal setup. BitMerge handles the migration and data platform engineering they don't do often enough to be fast at, and everyone stays in their lane.
In writing
The Cutover Guarantee
Migrations run evenings and weekends as standard. If a cutover has to move into your working day, that portion of the engagement is free.
The Runbook Guarantee
Where disaster recovery or failover is in scope, the runbook is executed at least once before the engagement closes, and the results are written down. An untested recovery plan is not a plan.
About
Migrations rarely fail on the technology. They fail on the dependency nobody mapped, the runbook nobody tested, and the cutover window that turned out to be four hours shorter than it needed to be.
BitMerge is a senior cloud and data platform engineering practice — small on purpose — built around the parts of a migration that carry actual risk. Estate discovery. Wave planning. Recovery that has been proven rather than assumed.
Nineteen-plus years of systems administration, infrastructure engineering, and cloud architecture sits behind the work, including SaaS datacenter migrations, SQL Server platform transitions, and disaster recovery runbooks written against real contractual SLA, RTO, and RPO targets.
We are not trying to become your IT department. If you have an MSP or an internal team, keep them — that's usually the right structure.
Depth
Questions
Usually not, and usually you shouldn't want us to. They handle day-to-day operations and the things that break at 9am. BitMerge handles migration and data platform engineering — work most teams do a handful of times a career. The two fit together well, and we're glad to work directly with yours.
Because the same package can be a two-week job or a three-month one depending on what's actually in the estate, and a published number would be wrong in one direction or the other. What we will do is give you one fixed figure in writing before any work starts, so the estimation risk sits with us rather than you.
Within one business day, in writing. Scheduled work runs evenings and weekends. If you need someone reachable by phone within the hour, we're not the right fit — and we'd rather say so now than in month three.
Small and mid-sized organizations, and agencies or hosts running multi-tenant infrastructure. Below a certain size a fixed project fee rarely makes sense for you. Above it you need a delivery team rather than a specialist, and we'll say so.
Yes, and it's common. Subcontracted and white-label engineering for MSPs and IT firms is a normal part of the practice — you keep the client relationship, we do the platform work underneath it.
Nothing. You hold your own subscriptions, your own administrative accounts, and your own documentation throughout. Any access granted to us is yours to revoke at any time, and we'll walk you through doing it.
Contact
Fifteen minutes, no charge, no deck. Tell us what you're moving or what you're worried about, and you'll get an honest read on whether it's a real project, roughly what it involves, and whether we're the right people for it.
Prefer email? [[ hello@bitmerge.co ]]