Microsoft Project Online Has Retired: Practical Alternatives for Small Teams

Microsoft Project Online reached its official end of life on 30 September 2026, closing an era of web-based enterprise project management built around SharePoint and Project Web App. Organisations that relied on Project Online must now transition their scheduling, resource allocations, and portfolios to modern tools. While large corporations often move straight to enterprise cloud architectures, small teams and agencies frequently discover that those heavy enterprise stacks create unnecessary overhead.

Understanding what has changed, where Microsoft is pointing its users, and what lighter alternatives offer can help you decide how your team should schedule work going forward.

What happened to Microsoft Project Online?

Microsoft scheduled the complete retirement of Project Online for 30 September 2026, as announced on the Microsoft Tech Community blog (checked on 2026-09-29). The deprecation process began earlier in the year: since 1 April 2026, tenants were no longer permitted to provision new Project Web App (PWA) sites, according to Microsoft Tech Community updates (checked on 2026-09-29).

Project Online was originally designed for Project Management Offices (PMOs) needing structured timesheets, complex resource pools, and centralised portfolio governance. Over the last decade, Microsoft shifted its development focus toward the unified Microsoft Planner experience and Microsoft Power Platform. While Project Online environments are winding down, the traditional desktop client remains: standalone perpetual licences like Project Standard 2024 and Project Professional 2024 continue to be maintained for desktop-bound project engineers.

What is Microsoft's recommended replacement path?

Microsoft directs teams departing Project Online towards the modern Microsoft Planner platform inside Microsoft 365. Rather than maintaining a distinct online application, Microsoft merged tasks from To Do, basic Planner, and Project for the web into a single rebranded Planner application.

However, the capabilities you receive depend entirely on subscription tiers:

1. Basic Planner (included in Microsoft 365 suites): Basic plans provide Kanban boards, light checklists, and basic task assignment. On the basic tier, task dependencies can be viewed if created elsewhere, but you cannot create or edit dependencies yourself, as documented on the Microsoft Planner plans and pricing page (checked on 2026-09-29).
2. Planner Plan 1: Priced at $10.00 per user per month excluding sales tax in the US, or €8.70 per user per month excluding VAT in Spain (checked on 2026-09-29). This tier unlocks timeline (Gantt) views, sprints, backlogs, milestones, and the ability to create task dependencies.
3. Planner and Project Plan 3: Priced at $30.00 per user per month excluding sales tax in the US, or €26.00 per user per month excluding VAT in Spain (checked on 2026-09-29). It adds baselines, portfolio management, critical path tracking, advanced dependencies with lead and lag times, and rights to run the Microsoft Project desktop client on up to five PCs per user.
4. Microsoft 365 Copilot: If you want generative assistance to build plans from written goals, Copilot is available as an add-on costing €26.00 per user per month excluding VAT, and its plan generation agent only operates on premium plans, as outlined in the Planner Copilot FAQ (checked on 2026-09-29).

For large enterprises that already run their business data on Dataverse and Power Apps, migrating to Planner Plan 3 or custom Power Platform solutions preserves enterprise governance. But for small teams, agencies, and lean studios, adopting Plan 3 across several team members introduces substantial monthly expenses.

Where does Microsoft Planner fall short for smaller teams?

While Microsoft Planner provides extensive enterprise features, its tiered structure introduces friction that catches smaller organisations by surprise.

Disjointed feature availability

Certain straightforward tools exist only in the basic tier, while traditional planning features reside exclusively in premium tiers. According to the official Planner feature comparison (checked on 2026-09-29), recurring tasks that repeat on an automated schedule are only supported in basic plans, not in premium plans. Conversely, creating task dependencies requires at least Plan 1, and configuring lead or lag timing on those dependencies requires Plan 3, according to the Microsoft Project blog (checked on 2026-09-29).

Similarly, task chat conversations have been rolled out to basic plans, while premium plans still utilise legacy conversation threads, as reported on the Microsoft Tech Community (checked on 2026-09-29). Even developer tooling exhibits this split: Microsoft released an agentic server using the Model Context Protocol, but the Microsoft Planner MCP server documentation (checked on 2026-09-29) indicates it only supports basic plans.

Baseline and portfolio constraints

In classic Project Online, baselines and multi-project reporting were core capabilities. In the modern Planner setup, baselines are restricted to Plan 3 and allow only one active baseline per plan, as detailed on the Planner baselines update (checked on 2026-09-29). Portfolios also require Plan 3 to create (Plan 1 users can merely view them), and tracking health requires manual status selection rather than automated progress derivation, as described in the portfolios guide (checked on 2026-09-29).

Manual date adjustments

When a deadline moves or a task takes two days longer than estimated, traditional project managers expect the schedule to recalculate immediately. In many collaborative tools, dates do not cascade automatically unless rigid constraints are configured. Project managers spend hours manually dragging bars across a timeline whenever a single colleague falls ill or a deliverable is delayed.

What should small teams evaluate when replacing Project Online?

If your team does not need enterprise resource levelling across hundreds of employees, you do not need to replace Project Online with another massive PMO suite. Small teams benefit from evaluating tools across four practical criteria:

1. True dependency mechanics: Does the software prevent people from starting work before upstream steps are ready? If a dependency is merely a visual arrow on a Gantt chart, it does not stop team members from working on blocked tasks out of order.
2. Automatic schedule calculation: When an estimate expands, does the system adjust future start dates and forecast project completion without manual intervention? High-maintenance timelines become obsolete the moment reality diverges from the initial plan.
3. Workload realism: Can one person realistically execute three sequential deliverables simultaneously? Tools that schedule tasks on parallel tracks for the same person without accounting for daily availability present a false picture of project delivery.
4. Licensing clarity: Enterprise tools frequently demand multi-tiered licensing where managers need expensive subscriptions while assignees require intermediate plans just to update status dates.

How do the main alternatives compare?

To replace Project Online, teams usually choose between sticking to desktop software, adopting generic collaboration boards, investing in modern enterprise suites, or moving to automated project schedulers.

Note on tokidu: tokidu is the project manager built by Sinergia Barcelona, the team behind this article.

When does Microsoft Project Desktop make sense?

For specialist planners who require complete control over work breakdown structures, Earned Value Analysis, and contractually mandated critical path network diagrams, standalone desktop software remains unmatched.

Project Standard 2024 costs €929.00 and Project Professional 2024 costs €1,659.00 as one-off purchases (the Microsoft store page does not specify whether VAT is included; checked on 2026-09-29). What desktop Project does better than any lightweight web application is deep mathematical modelling: it handles finish-to-start, start-to-start, finish-to-finish, and start-to-finish links, complete with negative and positive lag offsets. It also tracks multiple versioned baselines across multi-year infrastructure schemes.

However, desktop software operates in isolation. Sharing plans requires emailing .mpp files back and forth, losing live team updates, and forcing one person to act as an administrative scribe who enters task updates manually.

How does tokidu approach project scheduling?

At Sinergia Barcelona, we designed tokidu, the project manager by Sinergia Barcelona, around a fundamentally simpler premise: small teams do not need cumbersome enterprise PMO configurations; they need to know what to work on today and when the project will genuinely finish.

Instead of making project managers manually adjust calendar bars, tokidu uses an automated sequencing engine:

• Starts when … is done: Dependencies in tokidu follow a strict finish-to-start rule. A task waits for predecessor tasks in the same project to finish. While waiting, the task stays locked. The moment the preceding task is marked complete, the downstream task unlocks automatically and notifies the assignee. If the upstream task is reopened, downstream items lock again automatically. Circular dependencies are strictly rejected.
• Dates that work themselves out: When you assign an estimated duration to a task, tokidu distributes each team member's work across their defined working days. It assumes that nobody can work on two tasks at the exact same time. The system calculates when each task and the overall project will wrap up, highlighting immediately if an item will exceed its deadline. Only tasks with a specified duration are counted; tokidu never guesses a duration.
• Why that date?: Rather than wondering why a task is scheduled for a certain Friday, tapping on any projected date presents a single, plain sentence explaining what pushed it there—identifying which preceding task it is waiting for or what prior work is occupying that person.
• Up next: Team members do not need to sift through elaborate project charts to find their duties. The «Up next» view displays only the tasks that can be performed right now, completely hiding locked or waiting items.
• Gantt: An interactive timeline visualises tasks from the moment they can start until completion, showing clear dependency arrows. Durations can be adjusted directly by dragging on a desktop computer.

To keep scheduling clear and reliable, tokidu deliberately avoids unnecessary enterprise complications: there are no complex lead or lag offsets, no start-to-start links, no automated recurring tasks, no internal checklist subtasks, and no milestone markers. Features such as critical path calculations are not present under that name; instead, tokidu highlights the tasks that govern your finish date and explains each individual date.

tokidu offers a Free plan that is free forever without a payment card, supporting up to 2 active projects, 60 open tasks, and 1 person to assign tasks to. Paid tiers (such as Pro at a planned price of €3 a month or €29 a year, Team at a planned price of €9 a month or €89 a year, and Max at a planned price of €19 a month or €189 a year, all VAT included, as outlined on our pricing page) are not on sale yet. The Max plan includes an Assistant that turns pasted task lists into scheduled work, though it does not import raw files.

How to migrate away from Project Online step by step

Transitioning away from a retired platform can feel daunting, but breaking the migration into manageable phases keeps your active deliveries secure.

1. Audit active projects against legacy archives

Most organisations using Project Online retain dozens of dead or completed projects in Project Web App. Do not attempt to migrate completed projects into a new live system. Export completed projects as archive files or historical spreadsheets to retain records, and only migrate projects with remaining milestones and active deadlines.

2. Standardise dependency structures

In Project Online, planners frequently applied complex constraint dates («Must Finish On», «Start No Earlier Than») combined with lead and lag times. Often, those constraints were workarounds for simple human realities: waiting for client approval, awaiting materials, or waiting for a colleague to finish an initial draft.

Before recreating tasks in a new tool, simplify relationships into pure finish-to-start chains: task B cannot start until task A is delivered. Stripping away arbitrary calendar pins allows scheduling engines to forecast deadlines naturally.

3. Establish individual capacities

One fundamental flaw in unmanaged project boards is assuming team members have infinite parallel capacity. When moving to a modern platform, configure realistic working days. If a developer or copywriter works four days a week, your schedule should reflect that availability so target completion dates correspond with reality.

4. Run a pilot on one active deliverable

Never migrate an entire company overnight. Select one upcoming project, input its tasks, establish durations, assign owners, and link predecessors. Observe how the schedule responds when real-world delays occur. When your team sees that an updated task automatically recalculates downstream deliverables without administrative headaches, adoption follows smoothly.