Every SMB still running SharePoint Server gets the same question from the executive team this year: “Should we just move to the cloud version?” If you’re on SharePoint Server 2016 or 2019, the question has a deadline attached. Both editions reach end of support on July 14, 2026.
The honest answer for most teams is yes. But not for the reasons vendors give you. Here’s what actually changes, and the one thing that decides how hard the move is.
The short version, as a decision
Two questions decide it. A 'no' to both points almost every Ontario SMB to SharePoint Online.
What you lose
Vendors skip this part. Here’s the real list, in order of how much it hurts.
01 Deep customization
.wsp packages), sandboxed solutions, and farm-level features do not exist in SharePoint Online. The modern extension model is SPFx (SharePoint Framework) plus the Power Platform. Some custom code rewrites cleanly; some has no equivalent because the problem is now solved a different way.02 SharePoint Designer workflows
03 InfoPath forms
04 Direct SQL and server access
The InfoPath deadline is the one to circle on the calendar:
28days until InfoPath Forms Services shuts off in SharePoint Online.
That date matters even if you stay on-prem, more on that below.
What you gain
No more server
Underrated. You stop patching Windows Server, SQL, the SharePoint binaries, and the supporting stack. In a 10-50 person shop, that’s 2-4 hours a month of sysadmin time that simply disappears.
Modern collaboration that actually works
Real-time co-authoring in Word, Excel, and PowerPoint. Teams integration. External sharing without standing up an extranet farm. The gap between on-prem SharePoint and how people expect to work in 2026 is wide.
Native Copilot, nothing to install
SharePoint Server has no built-in Copilot. On-prem content can reach Microsoft 365 Copilot, but only through the SharePoint Server Microsoft 365 Copilot connector: a Graph connector, currently in preview, that needs its own connector agent and Active Directory sync. In SharePoint Online, Copilot indexes your content natively with nothing to set up. If AI is on the executive wish list, SharePoint Online is the shorter path. On-prem means building and maintaining extra infrastructure to get there.
A better compliance posture
Microsoft Purview, retention policies, sensitivity labels, DLP, all native and far easier to operate than the on-prem equivalents. For any SMB eyeing ISO 27001 or SOC 2, that removes a lot of manual effort.
Predictable monthly cost
Per-user-per-month instead of Software Assurance plus CALs plus server licences plus SQL plus Windows Server plus hardware plus patching labour. For SMBs under 200 users, the SharePoint Online number is almost always lower.
The file move is the easy part. The real cost is the customizations, and you can’t price what you haven’t counted.
The part that decides everything: your customizations
Two companies with the same headcount can have wildly different migrations. A 60-person firm with plain team sites and document libraries is a clean 3-6 week project. A 60-person firm with a publishing portal, twenty SharePoint Designer workflows, and a stack of InfoPath forms is a project measured in months.
This table is the one to bookmark. It’s the difference between a guess and an estimate:
| Asset type | Moves cleanly? | Notes |
|---|---|---|
| Document libraries | Yes | SPMT or Migration Manager (both free); ShareGate for large or complex libraries |
| Lists (no custom forms) | Yes | Metadata and views transfer |
| Team sites (no customization) | Yes | Permissions need a review |
| Publishing sites | Partial | No direct modern equivalent, usually a rebuild |
| SharePoint Designer 2013 workflows | No | Rebuild in Power Automate |
| InfoPath forms | No | Rebuild in Power Apps or Microsoft Forms |
| Full-trust solutions | No | Rewrite as SPFx or retire |
| Sandboxed solutions | No | Deprecated in SharePoint Online, remove |
| Custom master pages | No | Modern sites don’t use master pages |
If your farm is mostly the top rows, migration is fast. If it’s heavy on the bottom rows, budget the rebuild work. That’s where the real time and money go.
The tools, and what they actually cost
Robocopy won’t cut it. SharePoint Online throttles aggressive uploads, so you need a proper migration tool. Three options matter in 2026:
- SharePoint Migration Tool (SPMT), a free Microsoft client you download and run. Fine for most volumes.
- Migration Manager, also free, built into the SharePoint admin center. A central console for file shares and SharePoint Server sources.
- ShareGate, the common paid third-party option. Worth it for large or complex migrations that need detailed reporting and incremental sync.
Hidden costs on both sides
On-premises: server hardware refresh every 5-7 years, SQL licensing (many SharePoint 2016 deployments quietly run Enterprise SQL), patching labour, backup infrastructure, SSL certificate rotation, Windows Server CALs.
SharePoint Online: Power Automate premium connectors if your workflows need them, third-party migration tooling for anything non-trivial, governance setup time (SharePoint Online’s sharing defaults are more permissive than most on-prem farms), and external-sharing cleanup, which everyone defers and which bites about a year later.
When staying on-prem makes sense
SharePoint Server Subscription Edition is the legitimate “stay on-prem” option. It’s on the Modern Lifecycle, fully supported, with no end of support before the end of 2035. That’s unlike 2016 and 2019, which both retire July 14, 2026.
Staying makes sense in a short list of cases:
- An explicit regulatory ban on cloud storage for the workloads in scope.
- A remote site with unreliable internet that still needs local file authoring.
- A genuinely large customization estate the business won’t fund rebuilding yet, though Subscription Edition only buys time, it doesn’t avoid the work.
One thing it does not buy you: InfoPath. InfoPath Forms Services retires July 14, 2026 on Subscription Edition too. Staying on-prem gives you a supported platform, but your InfoPath forms are going away regardless. They still need rebuilding in Power Apps or Microsoft Forms.
The honest recommendation
For almost every Ontario SMB in the 10-200 user range running SharePoint 2016 or 2019: migrate to SharePoint Online. Operational savings, a stronger compliance posture, and native Copilot all point the same way. The pain is concentrated in the customization rebuild, and if you don’t have heavy customizations, there’s barely any pain at all.
Which is exactly why you inventory first. Count site collections, total storage, customizations, workflows, and forms. The estimate has almost nothing to do with user count and almost everything to do with that list.