The deadline is behind you, which changes the maths
Microsoft publishes an Extended End Date of 7/15/2026 for both SharePoint Server 2016 (opens in new tab) and SharePoint Server 2019 (opens in new tab). Mainstream support for 2016 ended back on 7/14/2021 and for 2019 on 1/10/2024. Nothing switches off at an end-of-support date — that is precisely the problem. The farm carries on serving pages, users notice nothing, and the only visible change is that security updates stop arriving for a server that holds your contracts, HR files, and project history.
Most guidance written on this topic is a countdown, aimed at a reader who still had months. If you are reading this now, you are past that. The practical consequence is narrow but real: the assessment and the first migration wave belong in this quarter rather than next year, and any project that was going to be phased over twelve months is better phased over three. The work itself has not changed at all.
What the free Microsoft tool moves, and what it cannot
The SharePoint Migration Tool (opens in new tab) is free, supports SharePoint Server 2010, 2013, 2016, and 2019 as sources, and is genuinely good at what it does. Microsoft’s own summary of what SPMT supports (opens in new tab) covers files, folders and list items, permissions, version history, content types and term stores, navigation for out-of-box sites, and incremental re-runs that pick up only what changed. Microsoft also ships a free assessment tool, SMAT, which scans a SharePoint Server 2010, 2013, or 2016 farm and reports the issues it finds before you start.
The sentence in that documentation that decides most projects is this one: “Out-of-the-box” SharePoint sites that don’t use any coding or third-party tools can be migrated. Almost no farm that has been in production since 2016 is out-of-the-box. There is a branded master page, a web part somebody’s predecessor wrote, a Business Connectivity Services connection to the ERP, a PerformancePoint dashboard, an InfoPath expense form. Each of those is a separate conversation, and the honest version of this project is that the content copy is the cheap part.
Where each part of the farm lands
This is the working checklist we take into the assessment session. It is deliberately blunt about which rows are a copy and which are a rebuild, because the rebuild rows are where the budget actually goes.
| In your farm |
Where it lands in SharePoint Online |
Decided before the first wave |
| Site collections and subsites |
Modern team and communication sites, usually flattened — one site per team rather than a subsite tree |
Which subsites deserve their own site, which merge, and which are an archive nobody has opened in four years |
| Content databases |
Nothing. There are no content databases in SharePoint Online; the content inside them is copied, the database itself is left behind |
How long SQL stays online as a read-only reference, and who signs off that it can finally be decommissioned |
| Files, list items, and version history |
Document libraries and lists, with the amount of file history you choose to preserve |
How many versions are worth carrying, since version count drives both migration time and destination storage |
| Permissions and AD groups |
Microsoft 365 groups and SharePoint groups, applied after identities are mapped |
Which broken-inheritance exceptions get rebuilt and which get cleaned up while you have the excuse |
| Content types and term store |
Site columns, content types, and the managed metadata term store, all supported by the migration tool |
Which content types are still applied to real documents, and which are duplicates of each other |
| Out-of-box web parts |
Many migrate; a long list does not. Business Data and BCS web parts, Excel Web Access, Visio Web Access, PerformancePoint, Content Search, and the search-driven content web parts are all marked unsupported in Microsoft’s supported web parts table (opens in new tab) |
Which pages depend on an unsupported web part, and what replaces it — often a modern web part or an embedded Power BI report |
| Custom web parts and full-trust solutions |
Rebuilt as client-side SharePoint Framework components (opens in new tab), replaced by a built-in feature, or retired |
Whether anyone still uses it, whether source code survives, and whether the original problem still exists |
| SharePoint Designer workflows |
Rebuilt as Power Automate flows. SharePoint 2013 workflow was removed from existing Microsoft 365 tenants on 2 April 2026, so there is no engine waiting for them. Farms that ran Nintex Workflow instead of Designer get the same rebuild through our Nintex to Power Automate migration services |
Which workflows are still firing, who owns each one, and which approvals should be redesigned rather than re-created |
| InfoPath forms |
Rebuilt as Power Apps, the alternative Microsoft names in its own deprecation notice |
Which forms are still submitted, and which were a workaround for a list column all along |
| Branding, master pages, and page layouts |
Not carried across. Modern sites use site designs, themes, and header layouts instead |
How much of the branding was a requirement and how much was a 2016 intranet fashion |
| Search configuration |
Rebuilt: managed properties, result sources, and query rules are farm-level configuration, not content |
Which custom search experiences people actually use, versus the ones built for a launch and never revisited |
One operational detail that catches teams out: several web parts only migrate if the destination tenant has custom script allowed, and Microsoft notes that the setting should be changed at least 24 hours before you start and left in place until the migration completes. It is a small thing that quietly costs a weekend when it is discovered on the Saturday.
Subscription Edition is a real option, and it is not a reprieve
The other supported destination is SharePoint Server Subscription Edition, and Microsoft documents a database-attach upgrade path (opens in new tab) from 2016 and 2019. If you have a genuine data-residency requirement or a line-of-business application welded to the farm, it deserves a serious look. What it does not do is save your legacy artifacts. Microsoft’s list of what is deprecated or removed in Subscription Edition (opens in new tab) is explicit that InfoPath Forms Services, SharePoint 2010 workflows, SharePoint Designer 2013, and Basic authentication were supported only until 14 July 2026, and that PerformancePoint Services and both versions of Access Services were removed outright.
| Consideration |
SharePoint Server Subscription Edition |
SharePoint Online |
| Who patches it |
You do, on your servers, forever |
Microsoft does, as part of the subscription you already pay for |
| InfoPath forms |
Supported only until 14 July 2026 — a rebuild either way |
Rebuilt in Power Apps |
| SharePoint Designer workflows |
SharePoint Designer 2013 supported only until 14 July 2026 |
Rebuilt in Power Automate |
| Full-trust custom code |
Still deployable, and still yours to maintain against every update |
Not deployable; rebuilt in the SharePoint Framework or replaced |
| PerformancePoint dashboards |
Removed from the product; Microsoft points to Power BI |
Rebuilt as Power BI dashboards |
| Copilot, Teams, and modern search |
Not available in the on-premises product |
Included, once permissions are clean enough to trust them |
| Ongoing cost shape |
Server licences, SQL, hardware or IaaS, and the admin time to run all of it |
Included in Microsoft 365 licensing you are almost certainly already buying |
For the small and mid-sized businesses we work with, the second column usually wins on the ongoing-cost row alone, because the farm is already being run part-time by someone whose actual job is something else.
The three phases of the project
The table below shows scope and typical timeline rather than prices. An honest number needs the farm inventory first — the same principle behind how our fixed-fee process works. For reference on the alternative, published US rate surveys in 2026 put most independent IT consultants at roughly $100–$250 per hour, with senior specialists commonly $150–$300; treat those as market data rather than a quote, and note that an hourly meter on a migration rewards the slowest possible migration. We quote one written figure for the whole project instead, which is the argument we make at more length in fixed-fee versus hourly consulting. A farm is the heaviest version of this work; if your content sits on a file share instead, the same scoping logic applies on a smaller scale and our guide to SharePoint migration cost for a small business covers the drivers that set it.
| Phase |
Scope |
Typical timeline |
Fee |
| Phase 1 — Farm assessment and design |
Scan of every web application, site collection, and content database; the workflow, form, and custom-code catalogue; keep, archive, and retire decisions; the target site and content-type design; the identity map; a wave plan and cutover date |
About 1 week |
Fixed fee — scoped after your free consultation |
| Phase 2 — Build and pilot |
Destination sites, libraries, content types, term sets, and groups built; rebuilt Power Automate flows and Power Apps forms tested; one volunteer site collection migrated end to end; the plan corrected before anyone else moves |
About 1 week |
Fixed fee — scoped after your free consultation |
| Phase 3 — Waves, cutover, hypercare |
Remaining site collections migrated in waves with validation at each step, an incremental pass at cutover, the old farm set read-only, per-team training, admin handover, then 30 days of included hypercare |
1–2 weeks, plus 30 days of hypercare |
Fixed fee — scoped after your free consultation |
Everything lands in the tenant you own. The sites, content types, flows, apps, documentation, and admin access are yours throughout, so your team can carry on in-house the moment hypercare ends.
Who this is for
The usual fit is a company of roughly 20 to 500 people running a single SharePoint Server 2016 or 2019 farm that has quietly become load-bearing — an intranet, a handful of departmental document stores, two or three approval workflows, and one form that finance refuses to live without. Often the trigger is an insurance questionnaire or a client security review asking, in writing, whether any unsupported server software is in scope. That question rarely stops at SharePoint: the same cupboard usually holds a mail server that lost support on 14 October 2025, which is a separate project with its own artifacts — mailboxes, public folders, and an MX cutover — covered by our Exchange Server to Exchange Online migration. The work is delivered by senior US-based engineers from our Orange County, CA base and remotely across all US states.
If the content you are worried about sits on a Windows file share rather than in a farm, the shape of the project is different and our file server to SharePoint migration playbook covers that path instead — plenty of companies do both, in that order. Once the farm content lands, our SharePoint intranet and document management consulting turns the migrated sites into a workspace with navigation and governance behind it, and the workflow rebuilds usually grow into wider approval workflow automation. Because Microsoft 365 Copilot inherits whatever permissions it finds, teams planning a Copilot rollout are better off finishing this clean-up first rather than indexing twelve years of over-shared farm content.