MS Project Gantt Is Dead — What Now?

Build planning into your product

Start with the RevoGrid Gantt product page for live demos, Microsoft Project-style planning features, framework support, pricing, and trial access.

Microsoft states that as of August 2025, Project for the web and the Project and Roadmap apps in Teams were retired and users transitioned to Planner for the web and Planner in Teams. Existing Project for the web work is now accessible in Planner.

For a long time, when people said “Gantt chart”, many users heard “Microsoft Project”.

The classic workflow was familiar: open a project file, define tasks, connect dependencies, check the critical path, level resources, then export the plan for the next meeting. It was not always simple, but it became the mental model for project scheduling.

Now that model is changing.

Microsoft has moved Project for the web into Microsoft Planner, and Project Online has an official retirement date. The desktop Project product still exists, but the center of gravity for Microsoft cloud project management is clearly moving toward Planner.

WARNING

MS Project desktop is not disappearing. But the old cloud Project experience is being folded into Planner, and Project Online is being retired. For many teams, this is the moment to rethink where project scheduling belongs.

That creates a big question for software teams, SaaS builders, internal platform teams, ERP vendors, construction tools, manufacturing systems, and operations products:

If users are leaving the old Microsoft Project workflow, where will their schedules live next?

For many products, the answer will not be “another standalone project management app”. It will be inside the application users already work in. That means modern software needs a serious Gantt and scheduling layer.

What exactly changed with Microsoft Project?

Let’s separate the noise from the important facts.

Project for the web has moved into Planner

That means users who previously opened Project for the web are now expected to work through Planner endpoints and Planner experiences.

Project Online is retiring

Microsoft has also announced that Project Online is being retired.

Important dates:

Date	What happens
October 1, 2025	End of sale for Project Online-only SKUs for new customers
September 30, 2026	Official Project Online retirement date

Microsoft says existing customers can continue using Project Online with full support until September 30, 2026.

Project desktop is still alive

This is important: the change does not mean every Microsoft Project product is gone.

Microsoft says Project Online retirement does not affect:

• Project desktop
• Project Server Subscription Edition
• Planner

Project Standard 2024 and Project Professional 2024 are still listed by Microsoft as on-premises project management products. Project Standard includes traditional capabilities such as Gantt Chart view, task dependencies, milestones, baselines, critical path, resource leveling, calendars, and task sheets.

The opportunity: Gantt for the business software

A lot of companies keep the need for planning.

Each company still needs to answer:

• When does work start?
• Which task blocks which task?
• What happens if one task slips?
• Which tasks are critical?
• Which resources are overloaded?
• Which dates are hard constraints?
• Which deadlines are at risk?
• What changed since the baseline?

The difference is where those questions are answered.

In the old model, the schedule often lived in a dedicated project planning tool. In the new model, the schedule increasingly lives inside the system where the work already exists:

• a construction management platform
• a manufacturing planning system
• an ERP module
• a field service scheduler
• a product roadmap tool
• a SaaS workflow builder
• an internal operations dashboard
• a professional services platform
• a procurement or logistics system

That is why many teams will not only migrate from Microsoft Project to Planner. They will also move planning into tools they own, customize, and embed deeply into business processes.

And those tools need a strong Gantt.

A modern Gantt cannot be only a timeline renderer

A simple timeline answers one question:

Where should I draw this bar?

A real scheduler answers a harder question:

Given dependencies, calendars, constraints, deadlines, progress, resources, and manual edits, what should the correct dates be?

That difference matters.

A visual-only Gantt works for demos. It can draw bars, arrows, milestones, maybe even drag and resize tasks. But once users bring real project data, visual-only scheduling breaks down quickly.

Real project planning needs deterministic rules.

If Task B depends on Task A, the system must know what to do. If a task can only start after a permit date, the system must know what to do. If a resource is double-booked, the system must warn or level. If a deadline is missed, the system must show risk without silently changing user intent.

This is where the next generation of Gantt components will be judged.

Not by how beautiful the bars look.

By whether users can trust the schedule.

What a good Microsoft Project replacement needs

A useful Gantt for former MSP-style users should support the project management concepts they already understand, but without forcing every product to become a full Microsoft Project clone.

Here is the practical checklist.

Capability	Why it matters
Task hierarchy	Users expect phases, summary tasks, child tasks, WBS-style structure, and rollups.
Milestones	Important dates need zero-duration markers, not fake one-day tasks.
Dependencies	Real plans need finish-to-start, start-to-start, finish-to-finish, and start-to-finish relationships.
Lag and lead	Schedules need waiting time and overlap, not only strict sequential chains.
Forward and backward scheduling	Some projects schedule from a start date; others plan backward from a required finish date.
Manual and automatic tasks	Users need both authored dates and dependency-driven dates.
Constraints	“Start no earlier than”, “must finish on”, and similar rules are essential for real-world planning.
Deadlines	Deadlines should show risk without always moving the task.
Calendars	Weekends, holidays, working hours, resource calendars, and time zones all affect duration.
Actuals and progress	In-progress and completed work should influence remaining schedule logic.
Resource planning	Assignments, workload, over-allocation warnings, and leveling matter once people or machines are shared.
Critical path and slack	Users need to know which tasks can slip and which tasks control the finish date.
Diagnostics	When rules conflict, the system should explain the problem instead of silently producing surprising dates.

That is the minimum for serious planning.

But there is one more requirement that many Gantt products miss.

The hidden requirement: no context loss

Most business applications are not born as Gantt chart.

They start as tables.

A task table may already have:

• custom columns
• validation rules
• inline editors
• dropdowns
• status badges
• permissions
• formulas
• row grouping
• keyboard workflows
• filters
• selection behavior
• cell templates
• custom context menus

Then the product team adds a Gantt view.

This is where many implementations fail.

They treat Gantt as a separate screen with a separate data model and a separate editor. The result feels like leaving the app and entering a different tool. Users lose the table context that made the product useful in the first place.

A better approach is:

Keep the table. Add scheduling intelligence and timeline projection on top.

This is the core idea behind RevoGrid Gantt.

RevoGrid Gantt is included in Pro Advanced and built as a plugin on top of the base RevoGrid data grid. RevoGrid owns rendering, virtualization, editing, keyboard interaction, selection, and scrolling. Gantt adds timeline projection, task bars, dependency links, scheduling rules, resources, assignments, baselines, critical path, and timeline tools.

In other words: the Gantt view does not replace the grid.

It extends it.

That means users can move between a classical data table and a Gantt view without losing the context of the data they were already working with.

Why this matters after MS Project

Former Microsoft Project users already understand the table-plus-timeline mental model.

MS Project was never only a chart. It was always a combination of:

• task sheet
• Gantt view
• dependencies
• calendars
• resources
• constraints
• actuals
• critical path

So if those users move into a new web system, they will expect the same conceptual power.

But modern SaaS users also expect something MS Project was not designed around:

• embedded workflows
• custom business data
• web-native collaboration
• design-system integration
• application-specific editors
• responsive performance
• framework-friendly implementation
• predictable JSON data ownership

This is the gap RevoGrid Gantt is trying to fill.

Not a heavyweight MSP clone.

Not a decorative timeline.

A lightweight, understandable, developer-friendly Gantt and scheduler that fits inside real products.

RevoGrid Gantt: table-first planning with real scheduling

The important part of RevoGrid Gantt is not only that it renders task bars.

The important part is that the scheduling logic behaves like a deterministic CPM-style engine.

At a high level, the scheduler:

1. validates task, dependency, calendar, resource, and assignment data;
2. normalizes rows into scheduled tasks;
3. builds the dependency graph;
4. computes task dates in dependency order;
5. applies constraints and reports conflicts;
6. optionally levels resources;
7. rolls up summary tasks;
8. emits diagnostics;
9. computes critical path and critical dependencies;
10. projects the result back into rows and Gantt layout.

That gives users something much more useful than a painted timeline: a schedule they can reason about.

Calendars and working time

A reliable Gantt must understand that five calendar days and five working days are not the same thing.

RevoGrid calendars support working weekdays, holidays, working-hour windows, hours per day, and timezone metadata. Tasks and resources can reference calendars, and scheduling can skip non-working time when calculating duration.

This is essential for teams that plan across:

• public holidays
• regional calendars
• part-time resources
• non-working weekends
• working-hour windows
• timezone-specific schedules

Resource leveling

Resource planning is where many Gantt implementations become fragile.

It is easy to draw two tasks assigned to the same person at the same time. It is harder to explain that the person is over-allocated and then resolve the conflict deterministically.

RevoGrid supports resource over-allocation diagnostics and deterministic delay-based leveling for eligible automatic tasks. The goal is predictable behavior: users should be able to understand why a task moved.

Critical path and slack

Critical path is one of the most important concepts former MS Project users expect.

It answers:

Which tasks have zero slack before they delay the project finish?

RevoGrid computes early and late dates, total slack, and critical state after scheduling rules are applied. Critical tasks and dependencies can then be highlighted in the Gantt view.

For users, this is the difference between “a schedule changed” and “we understand what changed the finish date”.

Bryntum, Webix, and RevoGrid: different philosophies

There are already serious JavaScript Gantt options in the market.

Bryntum Gantt

A mature, powerful JavaScript Gantt suite. It positions itself around performance, customization, a strong scheduling engine, dependencies, constraints, scheduling modes, exports, and integration with frameworks such as React, Angular, and Vue.

Webix Gantt

Another complete JavaScript Gantt component. It focuses on project management applications with tasks, projects, milestones, dependencies, working calendars, critical path, resources, workload visualization, and integration into modern web apps.

Both are valid options, and both show the same market signal: companies still need serious Gantt functionality after the Microsoft Project transition.

RevoGrid Gantt

RevoGrid Gantt starts from the grid. That makes it especially interesting when your users already work in a data-heavy application and you want to add Gantt without replacing the user’s workflow.

Product direction	Best fit
Microsoft Planner	Teams already committed to Microsoft 365 work management
Bryntum Gantt	Full-featured advanced JavaScript Gantt requirements
Webix Gantt	Ready-made enterprise Gantt widget with broad UI toolkit integration
RevoGrid Gantt	Table-first products that need Gantt and scheduling without context loss

This is not about saying one tool wins every use case.

It is about choosing the right abstraction.

If you need a standalone planning suite, choose a standalone planning suite.

If you need a Gantt inside your product, attached to your own rows, columns, editors, templates, validation rules, and business data, a table-first Gantt becomes much more attractive.

What RevoGrid is not trying to be

It is important to be honest here.

RevoGrid Gantt is close to Microsoft Project in core scheduling concepts, but it is not a bit-for-bit Microsoft Project clone.

That is intentional.

Microsoft Project has decades of scheduling behavior, legacy compatibility, prompts, side effects, import/export expectations, enterprise server models, and advanced edge cases. Recreating all of that would make the product heavy and harder to understand.

RevoGrid focuses on a deterministic web scheduling subset that developers can reason about:

• task hierarchy
• summaries
• milestones
• manual and automatic task modes
• FS / SS / FF / SF dependencies
• lag and lead
• constraints
• deadlines
• calendars
• actuals and progress-aware scheduling
• effort modes
• resource warnings and deterministic leveling
• total slack
• critical path

That is the right trade-off for many modern products.

The goal is not to bring every old MSP behavior into the browser.

The goal is to give users the scheduling power they need inside the app where work actually happens.

Migration checklist for product teams

If your users are coming from Microsoft Project, Project for the web, or Project Online, do not start by drawing bars.

Start with the data and the workflow.

1. Inventory the schedule model

Before choosing a Gantt component, list what your users actually have:

• tasks
• hierarchy
• dependencies
• milestones
• calendars
• resources
• assignments
• baselines
• deadlines
• constraints
• percent complete
• actual start and finish
• custom fields
• attachments
• permissions
• reports

A Gantt that cannot express your core model will create migration pain later.

2. Decide where the source of truth lives

Will the schedule be owned by:

• the frontend?
• your backend?
• Microsoft Dataverse?
• an ERP?
• a custom project database?
• imported JSON?
• an .mpp migration workflow?

For embedded planning, the source of truth is often your application, not the Gantt component. The Gantt should project and update your model without forcing a closed data shape.

3. Preserve the table experience

If your product already has a powerful task table, do not throw it away.

Users should keep:

• familiar columns
• custom editors
• existing templates
• keyboard navigation
• validation
• context menus
• filters
• permissions
• selection behavior

Adding a Gantt should feel like adding a timeline dimension to the existing table, not opening a separate project management universe.

4. Treat conflicts as product UX, not engine errors

Scheduling conflicts are normal.

A task may violate a constraint. A dependency may push a task beyond a deadline. A resource may be over-allocated. A manual task may ignore a predecessor.

The engine should report these conditions clearly so the product can guide the user.

Good diagnostics build trust.

Silent date changes destroy trust.

5. Start simple, then expose more power

Most users do not need every MSP scheduling knob on day one.

A good migration path is:

1. show tasks and milestones;
2. add hierarchy and dependencies;
3. add calendars and working days;
4. add deadlines and warnings;
5. add critical path;
6. add resources and workload;
7. add leveling and advanced constraints.

This is how you keep the product understandable.

Final thought: the Gantt is not dead. The silo is.

Microsoft Project shaped how a generation of users thinks about scheduling.

But the future of Gantt is not only in standalone project planning tools.

The future is embedded.

Users want to plan work where the work already exists. Developers want a scheduling engine they can understand. Product teams want a Gantt that does not destroy their table, templates, editors, or data model.

That is why the Microsoft Project transition matters.

It creates a moment where teams will re-evaluate their planning stack.

Some will choose Planner.

Some will choose advanced standalone Gantt suites.

Some will build planning directly into their own products.

For that last group, RevoGrid Gantt offers a strong direction:

Keep the grid. Add the timeline. Preserve the context. Schedule deterministically.

That is the next Gantt users will actually enjoy.

Sources

• Microsoft Support: Frequently asked questions about Microsoft Planner
• Microsoft Planner product page
• Microsoft Project plans and pricing
• RevoGrid Gantt Overview
• RevoGrid Gantt MSP Compatibility
• RevoGrid Gantt Engine Logic
• RevoGrid Gantt Critical Path and Slack
• RevoGrid Gantt Resource Leveling
• Bryntum Gantt
• Webix Gantt