Converting MS Project Schedules to Primavera P6: 7 Differences That Change Your Dates

Same project, different logic, different dates

A contractor on a water treatment upgrade once submitted their first P6 update after converting from Microsoft Project. The client's reviewer sent it back within two days. The number of critical activities had more than doubled, three key milestones were showing negative float, and a curing sequence that had carried four days of float now had one. On site, nothing had changed. The crews were exactly where they'd been the week before.

Untangling it took the best part of two days. There was no corruption and no software bug. There were three schedule settings and one calendar that behaved differently in P6 from the way they had in MS Project. Everything the reviewer flagged traced back to those four things.

If you're moving a programme out of Microsoft Project this year — because Project Online is being switched off on 30 September 2026, or because a new contract insists on P6 — this is the article I wish that contractor had read first. It doesn't walk through menus. It explains the seven places where the two tools quietly disagree, what each disagreement does to your dates and float, and how to fix it. It sits inside the Learning Tracks programme on PMMilestone Academy and pairs with the Knowledge Pillars library.

Same CPM engine, different assumptions

Think of two survey crews setting out the same building from the same drawings, but working to different site datums. Both are doing correct work. Their levels still won't agree until someone reconciles the datums. Microsoft Project and Primavera P6 are like that. The forward and backward pass maths is the same; the reference points aren't.

The first step is knowing where each MS Project object actually lands when it arrives in P6. Some map cleanly. Some map but need a second look. A few have nowhere to go at all. Terminology used below — total float, data date, level of effort, retained logic — is defined in the Project Controls Glossary.

Difference 1 — Calendars and the hours-per-day trap

MS Project keeps durations in minutes and converts them to days using a single project option, Hours per day, which defaults to 8 and has no connection to the calendars you've actually built. P6 stores durations in hours and, depending on how the administrator has set it up, can convert hours to days using each activity's own calendar.

Consider a rail station upgrade running possession-based night works: four 10-hour shifts, Monday to Thursday. The planner built the calendar correctly but never touched the Hours per day option.

Nothing on the programme moved. The number that changed was the label the planner typed, not the time the site works to. But try telling that to a reviewer scrolling through 400 activities that all seem to have lost a fifth of their duration.

Two more calendar issues deserve a mention. MS Project lets you define work weeks for date ranges — for example longer summer days and shorter winter days — while P6 calendars use a standard work week plus date-specific exceptions. And holiday periods, such as a three-week Christmas and New Year shutdown, need to be checked explicitly in P6, because a single missing exception shifts every date downstream of it.

Difference 2 — Constraints, typed dates and deadlines

MS Project allows one constraint per task, plus a separate deadline field. P6 allows a primary and a secondary constraint and has no deadline field at all. Most types have a clear equivalent. Two need real care.

The subtle part is an MS Project option called Tasks will always honor their constraint dates, found under File › Options › Schedule and switched on by default. With it on, a Finish No Later Than constraint can hold a task to its date even when the predecessors can't possibly finish in time. MS Project shows the impossible date and flags negative slack. In P6, Finish On or Before only affects late dates, so the early dates follow the logic and the activity moves later. The date that appeared “achievable” in MS Project never was.

Then there are typed dates. Enter a date by hand in the Start column of an automatically scheduled task, in a project scheduled from its start date, and MS Project responds by attaching a Start No Earlier Than constraint. There's no warning dialog. On a school campus programme I looked at, roughly a third of the construction tasks had one. None of them was a hard constraint, so the usual checks didn't complain. But they locked activities in place so that early finishes upstream couldn't pull anything forward, which quietly skews both float and any later delay analysis — the sort of hidden constraint that turns up repeatedly in the Project Failure Database.

Difference 3 — Total float and the critical path

MS Project reports total slack as the smaller of start slack and finish slack. P6 lets you choose how total float is computed: start float, finish float, or the smallest of the two. P6 also lets you define critical activities by a float threshold or by the Longest Path. If the settings differ between the two tools, the critical path can look quite different even though the logic is untouched.

Here's what that looked like on the water treatment upgrade. The chart compares float before and after conversion for eight activities. The logic was identical in both files.

1. Wall pours (cure lag): 4 → 1 day. The curing lag had been entered as elapsed time in MS Project. Once it was being measured on a five-day working calendar in P6, seven days of curing took longer in real terms and float shrank. Always check how elapsed lags arrived.

2. M&E installation: 10 → 8 days. Pure display. The same 80 hours of float, shown in 8-hour days in MS Project and 10-hour days in P6.

3. Tie-in (Must Start On): −2 → 0 days. The constraint arrived as Mandatory Start. P6 holds the constrained activity at zero float and pushes the negative float onto its predecessors, where it actually belongs.

4. Commissioning: 0 → 2 days. Start float was 0 but finish float was 2. MS Project reports the smaller value; the P6 project was set to finish float.

Difference 4 — Relationships and lags

MS Project allows only one relationship between any two tasks. P6 allows several. That sounds minor until you look at how linear construction work gets modelled. For a trunk sewer, excavation, pipe laying and backfill should really be linked with both a start-to-start lag (the following crew can start once there's enough open trench) and a finish-to-finish lag (it can't finish until the leading crew is done). In MS Project you have to pick one, so most planners use the SS lag and hope the finish takes care of itself.

Conversion is the right moment to rebuild those as SS + FF pairs in P6. It usually lengthens the tail of the linear sequence slightly and gives a far more honest picture of float at the end of the run.

Lag needs attention too. P6 has an explicit schedule option, Calendar for scheduling relationship lag, with choices of predecessor activity calendar, successor activity calendar, 24-hour calendar or project default calendar. For something like concrete curing, my strong preference is not to hide it in a lag at all. Model curing as a visible activity on a seven-day calendar. It shows on the Gantt, it has its own float, and nobody needs to remember a setting to understand it. The same reasoning drives the method choices in the delay analysis knowledge pillar.

Difference 5 — Status date, data date and progress

In MS Project, the status date is informative. Work that should have started can still appear in the past. In P6 the data date is enforced: remaining work is always scheduled on or after it. Convert a file where unstarted tasks sit behind the status date and the first F9 will move them, together with everything downstream.

P6 also asks you to choose how out-of-sequence progress is handled: Retained Logic, Progress Override or Actual Dates. Retained Logic keeps the remaining duration of an out-of-sequence activity waiting for its predecessors; Progress Override lets it continue as if the logic had been satisfied. The choice can move a completion date by weeks on a heavily progressed schedule. Many specifications name the required method, so check the contract before accepting whatever default the database was set up with — and see how the same decision plays out on real programmes in the mega-project case studies.

Difference 6 — Activity types: milestones, LOE and summaries

MS Project offers tasks, summary tasks and milestones. P6 offers Task Dependent, Resource Dependent, Level of Effort, WBS Summary, Start Milestone and Finish Milestone activity types. Conversion is a good opportunity to put each activity into the type that matches how it really behaves.

Difference 7 — Baselines, custom fields and IDs

An MPP file can hold up to eleven baselines inside it. In P6, a baseline is a separate copy of the project, and baselines inside an MPP won't arrive as usable P6 baselines. Re-establish them deliberately and record how. Custom fields need mapping to UDFs of matching data types, or the import errors out. And hold on to each task's original Unique ID in a dedicated UDF until handover; correspondence that quotes old task numbers doesn't disappear just because the software did.

Convert or rebuild? A decision table

Conversion is not always the right answer. Sometimes the cleanest route to a credible P6 submission is to rebuild the programme properly and use the old file only as a reference.

Common mistakes when converting MS Project to P6

Every one of these has cost a project team credibility in front of a reviewer. Patterns like them recur across the Project Failure Database and the delay claims library.

Expert tips

A few habits make the difference between a conversion that survives review and one that generates a hundred comments. Where the arithmetic matters, run it through the SPI calculator, the CPI calculator and the EAC calculator so the numbers in your conversion report reconcile with the numbers in your progress report.

Lessons learned

1. A settings problem looks exactly like a performance problem. On the water treatment upgrade, the reviewer read the jump in critical activities as a project in trouble. It was a float calculation option and a lag calendar. Explain the settings before you present the dates.

2. Display differences need a paragraph, not a debate. On the night-works rail job, one worked example — 40 hours shown as 5 days in MS Project and 4 days in P6 — closed dozens of review comments in a single response.

3. Typed dates are the most common hidden issue. On the school campus programme, stripping out the hand-typed SNET constraints nobody could justify exposed a services sequence that had far less float than anyone believed. Better to learn that during conversion than during a claim.

4. Linear work deserves proper logic. Rebuilding SS + FF pairs on a trunk sewer showed the final section finishing later than the MS Project version suggested, which changed the resourcing plan for the last month.

Post-conversion QA scorecard (template)

Rate each check green, amber or red, and attach the completed scorecard to your first P6 submission.

Any red rating should stop the submission. Amber is fine as long as the conversion report explains it clearly enough that a reviewer who has never seen the MS Project file could follow the reasoning.

Closing thoughts

Converting a schedule from MS Project to Primavera P6 is less about moving data and more about translating assumptions. Calendars, constraints, float, lags, progress, activity types and baselines each carry a small difference in meaning between the two tools. Individually they're easy to understand. Together, and unexamined, they can make a healthy project look like it's in serious trouble. Work through the seven differences, one setting at a time, and your first P6 submission will say what your site team already knows.

Written by Dr. Hassan Eliwa, PhD, founder of PMMilestone.org, whose peer-reviewed publications and books cover project controls, planning and forensic schedule analysis. Continue with the Primavera P6 learning track, the negative float in P6 guide, or the wider Knowledge Pillars library.