Moving to the Cloud: A Migration Plan for a Small Business
A small business cloud migration plan works in phases: inventory what you run, decide per system whether to move, replace or retire it, build a secure landing area, pilot one low-risk workload, cut over with a rollback path and only then decommission the old environment. Skip the inventory and you'll migrate surprises.
The outline is provider-neutral and fits AWS, Google Cloud or Microsoft Azure. Adapt the steps to your size, and keep a written record of each decision.
Vendors Covered in this Article
Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.
How do you inventory what you have?
List every application, server, database, file share and integration, including the ones nobody wants to admit exist. For each, record:
- What it does and who uses it.
- Who owns it, and who is the technical contact.
- Its dependencies: databases, licenses, hardware, other systems and scheduled jobs.
- Its data: how much, how sensitive and how often it changes.
- Its uptime needs and what happens to the business if it's down for a day.
Talk to the people who use each system, not just the IT staff. A spreadsheet in a finance department that feeds an invoicing process is a system, and it will break if it's forgotten. This inventory becomes your project scope and your safety net.
What is the right strategy for each system?
Sort every item into one of these choices. Most small businesses end up using several:
- Retire: nobody uses it. The cheapest migration is deleting it.
- Replace: swap it for a hosted service, such as moving an in-house email or file server to a software subscription.
- Rehost: move the server largely as-is to a virtual machine in the cloud. It's the fastest, and it doesn't reduce operating work much.
- Replatform: move with modest changes, such as switching to a managed database.
- Keep: leave it on premises if it depends on local hardware or a migration isn't worth the risk.
Prefer replacing and retiring first, because they remove work rather than moving it. Rehosting everything gives you the same maintenance burden at a new address, with a monthly bill.
How do you build a safe landing zone before moving anything?
Set up the foundation first, because retrofitting security later is painful:
- Create separate accounts or projects for production and non-production.
- Connect the cloud to your identity provider, and enforce multi-factor authentication for every human and admin user.
- Give people roles with the least access they need, and avoid shared admin accounts.
- Design the network: private subnets for data, a controlled path in from the internet and no public database endpoints.
- Turn on audit logging and send it to a separate, protected location.
- Set budgets and alerts before workloads arrive.
- Define backups and test a restore.
Write your security expectations in your information security policy so the cloud follows the same rules. On patching, CISA's federal directive BOD 19-02 sets 15 days for critical vulnerabilities on internet-accessible systems1, a reasonable benchmark for your own target.
How do you pilot and cut over?
Choose a first workload that's useful but forgiving, such as an internal tool or a reporting database, and take it all the way through. The pilot proves your landing zone, your pipeline and your team's skills, and it produces real cost numbers.
For each production cutover, prepare a runbook with these parts:
- A rehearsal in a test environment, timed, with the results recorded.
- A data sync plan that keeps the source and target aligned until the switch.
- A cutover window chosen for low business impact, with clear go and no-go criteria.
- A rollback plan that keeps the old system intact and untouched for a defined period.
- A communication plan for staff and customers.
Say you migrate a customer database on a Saturday night. Rehearse the sync twice, keep the old database read-only afterward for two weeks and have someone verify key reports on Monday.
What comes after the move?
Migration isn't done at cutover. Plan these steps:
- Monitor and set service targets. Decide what good performance looks like using a reliability target, and alert when it slips.
- Prepare for incidents. Update your incident response plan and ransomware playbook for the new environment.
- Right-size and review cost monthly. The first bills are often higher than the estimate because both environments run in parallel.
- Train staff on the new tools and the on-call path.
- Decommission the old environment only after the agreed grace period, wiping disks and closing contracts and licenses.
Compare provider choices in the cloud provider comparison before you commit, and revisit the decision in a year with real data.
What Good Looks Like
Every system has a documented owner and migration choice, the landing zone enforces identity, network and logging rules, and each cutover has a rehearsed rollback.
Building The Capability (5-Stage Skill Ladder)
How to Get Started
Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.
Fits when you want the widest range of services and have someone comfortable operating them.
Fits when your workloads center on data, analytics or containers and your team prefers its tooling.
Fits when your business already runs on Microsoft software and identity, and you want that continuity.
Frequently Asked Questions
How long does a small business cloud migration take?
It depends on the number of systems and their dependencies. A handful of simple applications can move in weeks, while a complex environment takes months. The inventory phase gives you a realistic estimate.
Is the cloud cheaper than running our own servers?
Not automatically. Savings come from retiring unused systems, replacing servers with hosted services and right-sizing. A rehosted server can cost more than it did on premises. Model both environments, including staff time, before deciding.
What should we migrate first?
A useful but low-risk workload, such as an internal tool or reporting database. It validates your security setup, network and team skills before you touch systems that customers depend on.
Do we need a rollback plan?
Yes, for every cutover. Keep the original system intact and read-only for a defined period, rehearse the switch beforehand and set clear criteria for deciding to roll back.
Sources
Where we quote a benchmark, we show its source. Other figures in this guide are estimates or general guidance, so check them against your own numbers.
- Security patch remediation SLAs (CISA federal mandates, used as industry norm). CISA Binding Operational Directives 19-02 and 22-01 (CISA briefing hosted at NIST CSRC), 2022.
Related Guides
Information Security Policy for a Small Business: Outline and Examples
Write a short information security policy set for a small business: which policies you need, a section-by-section outline and example requirements.
SLOs for a Small Engineering Team: A Starter Worksheet
Define SLIs, pick a realistic availability target, set an error budget and decide what happens when it burns. A worksheet you can fill in.
Incident Response Plan for a Startup: A Fill-In Outline
An incident response plan outline for small engineering teams: roles, the first 15 minutes, communication steps, a security branch and a review process.
Ransomware Response Plan: Who Does What in the First 24 Hours
An outline for a ransomware response plan: roles, first-hour containment steps, the payment question, communications and how to recover safely.
AWS vs Google Cloud vs Microsoft Azure: Cloud Platforms for Startups Compared
Compare AWS, Google Cloud, and Microsoft Azure for startups: startup credits, Kubernetes engines, AI APIs, reliability budgets, and developer velocity.
Do You Need EDR? A Small Business Decision Guide
Endpoint detection and response goes beyond antivirus. See when a small business needs it, what to compare in demos, and how to roll it out.