New Outlook vs Classic Outlook: The Migration Microsoft Just Postponed Again

In late February 2026, Microsoft posted notice MC949965 in the Microsoft 365 Admin Center stating that the opt-out phase for the new Outlook for Windows in Enterprise environments would now begin in March 2027, rather than April 2026 as previously scheduled. The official reasoning was that Microsoft is “seeing strong and accelerating adoption of new Outlook” and wanted to give organizations more lead time to prepare.

That language is doing the same kind of work as the Secure Boot “degraded security state” framing. You don’t postpone a forced rollout by a full year because adoption is accelerating; you postpone it because the receiving end is not ready. Microsoft’s own published feature comparison table, which Office Watch has called “admirably transparent or a quietly damning admission,” lists the gaps: limited support for .pst files, no offline mode equivalent to the classic client’s, no support for COM and VSTO add-ins, no support for Exchange on-premises beyond basic IMAP-equivalent features, missing folder navigation features, missing email templates, and a long list of smaller items that individually look like minor inconveniences and collectively explain why enterprises have refused to flip the switch.

This page is about what the new Outlook actually is, what it isn’t, what Microsoft has now committed to in writing, and what to do about it depending on whether you are an individual user, a small business, or running a managed enterprise fleet.

What was supposed to happen

Microsoft’s stated rollout plan has three stages:

StageStatusDescription
Opt-inCurrent stateNew Outlook is off by default. Users see a “Try the new Outlook” toggle and can switch voluntarily. Both versions can run side-by-side.
Opt-outNow starting March 2027 (was April 2026)New Outlook becomes the default. Users can still revert to classic. IT admins keep policy controls.
CutoverNo earlier than March 2028Users cannot switch back to classic. New Microsoft 365 deployments ship with new Outlook only.

The classic Outlook desktop client is committed-supported through “at least 2029” — Microsoft’s documentation phrase. The qualifier is doing real work. In earlier rollout language, Microsoft said classic Outlook would be retired in 2029. The current language says classic will be supported “at least” through 2029, with no firm cutover date for organizations using perpetual licenses (Office LTSC) or active Microsoft 365 subscriptions.

Existing classic Outlook installations, both perpetual and subscription, will continue to be supported and download-available. Existing enterprise customers with a Microsoft 365 subscription that includes desktop apps will be able to download classic Outlook separately and use it for no additional fee, even after the cutover stage. This is a meaningful commitment that did not exist in 2023 when the original rollout language was published.

What “new Outlook” actually is

This is the part that gets misreported most often. The new Outlook for Windows is not a rewritten desktop application. It is a desktop wrapper around Outlook on the Web (OWA) — the same web-based mail client that runs at outlook.office.com. The desktop wrapper provides notifications, file system access, and some platform integration; the mail logic, the rendering engine, the calendar code, and the contact management all run in a web context.

This architectural choice explains every limitation of the new client. PST files do not import easily because PST is a Win32 file format and the new Outlook does not have native Win32 file APIs. COM and VSTO add-ins do not work because COM is a Windows-specific binary integration model that does not exist in a web context. Offline mode is limited because IndexedDB and service-worker-based offline storage are not equivalent to the classic Outlook’s OST file architecture. Exchange on-premises support is limited because OWA was designed for Exchange Online; on-prem connectivity is bolted on via IMAP/SMTP and does not include calendar federation, free/busy lookup, or shared mailbox features.

If you understand that the new Outlook is OWA-with-a-shell, every “missing feature” complaint becomes predictable. The features that are missing are the features that depend on Win32 architecture. They are not coming back. Some of them, like PST import, are being rebuilt as web-equivalents (PST import is scheduled to roll out starting April 2026 per Microsoft’s published roadmap). Others, like COM add-ins, are being retired permanently in favor of a web add-in model that is significantly less capable.

What is and is not missing right now

As of April 2026, the Microsoft-published feature comparison shows these gaps in new Outlook compared to classic:

Working but limited. PST file support (export-only as of late 2025; import scheduled for April 2026 rollout), offline mode (basic IndexedDB-based; no full OST equivalent), email signatures (rebuilt but with some formatting limitations), shared mailboxes (working; some delegate features missing), conditional formatting (rebuilt November 2025).

Not working. COM add-ins (will not be supported; web add-ins are the replacement). VSTO add-ins (same; not supported). Outlook Data Files for backup (export to PST works; import into new Outlook arriving April 2026). Multiple Outlook profiles (single profile with multiple accounts is the model). On-premises Exchange (supported via IMAP/SMTP only; no calendar federation, no free/busy, no public folders, no shared mailbox via on-prem).

Coming. Mail delegate permissions granted from new Outlook (rollout July 2026). Teammates’ calendars in left navigation pane (March 2026 preview). PST file import (April 2026). Inserting signatures into calendar events (postponed multiple times; April 2026 latest target).

Improved over classic. AI features tied to Microsoft 365 Copilot (Summary, Draft, Coach), pinning and snoozing emails, theming, faster search indexing in Exchange Online, native dark mode.

The Office Watch summary captures the dynamic: Microsoft maintains “a carefully worded feature comparison table showing what the company will admit is missing or incomplete from the new version. The list appears to be admirably transparent or a quietly damning admission but really needs to be read with the proverbial ‘pinch of salt.’”

The Exchange on-premises problem

This is the migration blocker that does not get enough attention. New Outlook treats Exchange on-premises and third-party hosted Exchange “like any other Internet mail service, with no more calendar integration or other features than you would get with Yahoo or Google” (per the Directions on Microsoft analysis). For an organization running mail on Exchange Online, this is irrelevant. For an organization running Exchange Server 2019 on-premises — and there are many such organizations, particularly in regulated industries, government, and healthcare — the new Outlook is functionally a different product that does not do the things their users currently rely on.

Microsoft’s stated path forward is to move those organizations to Exchange Online. The economics of that migration are non-trivial. Enterprise Exchange Online licensing costs more per seat than on-premises Exchange Server CAL licensing for organizations that already own the on-prem licenses. The data residency, compliance, and integration constraints that drove the on-premises decision in the first place are not changed by the new Outlook’s existence.

The honest reading of the Outlook rollout is that Microsoft is using the deprecation of classic Outlook as a forcing function for Exchange Online migration. This is also a meaningful part of why the opt-out deadline keeps slipping: enterprises with on-premises Exchange estates have not, and will not, complete the migration in 12 months.

What the rollout looks like by audience

Personal Microsoft accounts (Outlook.com, Hotmail, Live). Already migrated. Personal accounts began moving in 2025; the classic Mail and Calendar apps were retired in late 2024 and replaced with the new Outlook. If you are reading this on a Microsoft 365 Family or Personal subscription, the new Outlook is your default experience. The “Try the new Outlook” toggle has been replaced with “Switch back to classic Outlook” in most cases.

Small business (Microsoft 365 Business Standard / Premium). Opt-out phase rolled out in 2025; new Outlook is the default for new installations and many existing installations have been auto-toggled. Users can still revert to classic, but the friction is rising — the toggle is less prominent, and IT-managed migration policies are being applied progressively.

Enterprise (Microsoft 365 E3, E5, A3, A5, G3, G5). Opt-out phase delayed to March 2027 as of the February 2026 announcement. Until then, new Outlook is opt-in only. Administrators retain the Admin-Controlled Migration policy to either accelerate or block the rollout for managed users.

GCC, GCC High, DoD. No timeline announced. Microsoft has stated these will be addressed separately; expect lag of 6–18 months behind the Enterprise commercial rollout.

Office LTSC and perpetual licenses. Classic Outlook through Office LTSC 2024 is supported through October 2029 minimum. Office 2021 ends support October 13, 2026 — perpetual license users on Office 2021 should plan their migration around the Office 2021 EOL rather than the Outlook rollout.

What to do depending on who you are

You are an individual or small business user on Microsoft 365. Try the new Outlook with one user or your own account. Do not commit the whole organization until you have validated the features your workflow depends on. Particular things to test: signatures with images and HTML formatting, attachment behavior with .pst archives, calendar integrations with any third-party scheduling tools, COM add-ins (if you use any — they will not work and there is no path), shared mailbox access for any users who manage delegated calendars or inboxes.

You are the IT person at a small business and your users are getting auto-migrated. You have a few months left of the opt-out window before Microsoft tightens further. Use the Admin-Controlled Migration policy to either accelerate the rollout (if you have validated and want consistency) or pin users to classic (if you are not ready). The relevant Cloud Policy is “Choose whether to use new Outlook.” Setting it to “Disabled” prevents the opt-in toggle from appearing.

You are an enterprise administrator. The March 2027 opt-out date is firm enough to plan around but not firm enough to stake your career on. Treat it as a working assumption with a 6-month buffer. Validate new Outlook against your specific add-in inventory, your Exchange topology, and your security tooling. The big questions: do any of your security or compliance add-ins use COM/VSTO architecture (most do; the replacement web-based versions may not be feature-complete)? Are you on Exchange Online or on-premises (if on-prem, the migration becomes a multi-year project, not a 12-month one)? Do you have any PST-based archive workflows (you will need to migrate those to Online Archive or similar before the cutover)?

You are running Exchange on-premises. This is the hardest scenario. New Outlook does not work the way your users expect it to with on-prem Exchange. The honest planning horizon is 18–36 months, and the planning starts with “do we move to Exchange Online or do we accept that classic Outlook is our endpoint until 2029.” Both answers are legitimate; pretending the migration to new Outlook is straightforward is not.

When to stop

A specific category of advice in support forums for new Outlook issues is wrong and worth flagging. If new Outlook is not working with your Exchange account, the answer is not to “remove the registry key that blocks new Outlook” or “force-toggle from classic via the AppX manifest.” The forum advice that suggests these workarounds is operating on the assumption that new Outlook should work for your scenario and is being blocked. For Exchange on-premises and several other scenarios, new Outlook genuinely does not support what classic Outlook supports. The toggle is not the problem; the architecture is.

If you are stuck in a sign-in loop in either client (outlook keeps asking for password is the classic symptom), see our dedicated guide — that is a cached credentials issue and is not related to the new vs classic question.

If your search has stopped working in either Outlook variant, the cause is almost always the search index rather than the client architecture; see the search troubleshooting guide.

The pattern

What this rollout illustrates is the gap between Microsoft’s strategic preference and customer reality. Microsoft would prefer all of its customers on Exchange Online, all running web-based clients, all on a subscription cadence, all compatible with Copilot AI features. The product strategy, the rollout cadence, and the licensing changes all point in that direction. Customer infrastructure does not yet support that strategy at the scale Microsoft is targeting.

The result is a deprecation cycle that keeps slipping. Each slip is rationalized as adoption pacing; each slip reveals that adoption is not actually accelerating in the segments that matter. The 2029 floor for classic Outlook support is the conservative read on how long this transition actually takes; the realistic read is that classic Outlook in some form will remain available longer than that, because the alternative is breaking enterprise workflows that Microsoft cannot afford to break.

The practical conclusion: if you are planning around an April 2026 cutover, stop. If you are planning around a March 2027 cutover, build in a 6-month buffer. If you are planning around a 2028 or 2029 cutover, you are roughly right, and the question is what you do with the runway.

FAQ

Will classic Outlook stop working in 2026? No. The opt-out phase that was scheduled for April 2026 — where new Outlook would have become the default but classic remained available — has been postponed to March 2027. The cutover phase, where classic is no longer accessible, is no earlier than March 2028. Classic Outlook installations will be supported through “at least 2029.”

Why is everyone complaining about new Outlook? The new Outlook for Windows is a desktop shell around Outlook on the Web. Several features that depend on Win32 architecture — PST files, COM/VSTO add-ins, full offline mode, on-premises Exchange integration — either do not work or work in a more limited way. For users who depend on these features, the new client is functionally a different product.

Can I block the new Outlook from being installed? For managed environments, yes — the Admin-Controlled Migration policy in the Microsoft 365 Admin Center allows IT to disable the new Outlook toggle for users in their tenant. For unmanaged consumer devices, there is no permanent block; the toggle is always available, and Microsoft is progressively making it more prominent.

What happens to my COM/VSTO add-ins? They do not work in new Outlook. Microsoft’s intended replacement is the web add-in model (Office.js-based add-ins that work in both Outlook on the Web and the new desktop client). Web add-ins have a more restricted security model and may not be capable of replicating all functionality. If your organization depends on a specific COM add-in, contact the vendor to ask about their web add-in roadmap.

Will my PST files work in new Outlook? PST export from new Outlook works as of late 2025. PST import into new Outlook is scheduled to roll out starting April 2026 per Microsoft’s published roadmap. As of late April 2026, the rollout has begun but has not reached all tenants. Until import is generally available, users with PST archives should keep classic Outlook installed for archive access.

What about Outlook for Mac? The legacy Outlook for Mac (desktop) is being retired in coordination with the Exchange Web Services (EWS) deprecation on October 1, 2026. The “new Outlook for Mac” is the supported successor; over 95% of Microsoft 365 users have already migrated. If you are in the remaining 5%, the migration deadline is much closer than the Windows side.

Should I switch to new Outlook now? For personal email and individual users on Microsoft 365 Personal/Family, yes — you are already there. For small business, test first. For enterprise on Exchange Online, validate against your add-in inventory before committing. For enterprise on Exchange on-premises, no — the new client does not support your scenario meaningfully, and the classic client will be available through 2029.

Official references

Last updated: April 28, 2026.

↑ Back to top