Weekly vs monthly cycles

Last reviewed 9 September 20261,694 words8 min read

What gets updated when, and by whom

๐ŸŸฉ In one line: The weekly cycle runs the site โ€” a look-ahead, a percent-complete sheet and a short list of constraints โ€” and stays out of P6; the monthly cycle runs the contract โ€” the full status, reschedule, narrative and submission โ€” and is the only one that produces a record anyone will rely on later.

๐Ÿ‘ค Who this is for: J/M. Read statusing-a-schedule and monthly-update-checklist first; this page is about how the two rhythms fit together.

e / c
First, let's be honest about why this page exists

Planners either update P6 every week โ€” and spend Sunday through Tuesday statusing while the site moves on without them โ€” or update it monthly and arrive at the site meeting with dates three weeks stale. The first burns the planner out; the second makes the schedule irrelevant to the people building the job.

The answer is two products from one dataset. The weekly product is light, fast and built with the site team. The monthly product is the contractual record. The mistake is trying to make one document do both.

๐ŸŸจ The standard โ€” what "good" looks like
SourceWhat it says (paraphrased)
FIDIC 1999 Cl 4.21 / 2017 Cl 4.20Monthly progress reports to the Engineer, in a prescribed format, including progress against programme.
FIDIC 1999 Cl 8.3 / 2017 Cl 8.3Revised programme whenever the previous one is inconsistent with actual progress.
NEC3 / NEC4 Cl 32.2Revised programme at intervals stated in Contract Data (commonly 4 weeks) and whenever instructed.
SCL Protocol, 2nd ed. โ€” Core Principle 1 and Part BProgramme updated at regular intervals, progress records maintained contemporaneously; monthly updates as the minimum record.
PMI Practice Standard for Scheduling, 3rd ed.Status frequency should match the level of control; short-cycle status for execution, formal periodic status for reporting.
Last Planner System (Lean Construction Institute)Weekly work plan with percent plan complete; six-week look-ahead for constraint removal.

๐ŸŸข Rule: weekly is for the site and lives in the look-ahead; monthly is for the contract and lives in P6 โ€” data date on the same day every month, never moved to suit the numbers.

How it actually works

1. The two cycles side by side.

WeeklyMonthly
PurposeDirect the next three to six weeks of workRecord position, forecast, report to Employer
ToolLook-ahead (P6 layout exported to Excel, or Excel directly); percent sheetP6 full status and reschedule
Data dateRolling; last working day of the weekFixed: same calendar day each month (commonly 25th or month-end)
Who provides inputSection engineers, foremen, subcontractor supervisorsSame people, formally, via signed progress sheets
Who owns itConstruction manager, with the planner facilitatingPlanner, with PM approval
OutputTwo pages: this week's plan vs actual; next three weeks; constraint listUpdated XER, narrative, change table, S-curve, near-critical table, delay register
P6 changesNone โ€” or actuals only, no logic or duration changesActuals, remaining durations, approved logic and duration changes, calendar changes
RetainedWeekly sheets filed; not submittedArchived XER + narrative; submitted; becomes TIA base

2. The weekly routine โ€” four hours, not two days.

DayTask
Thursday PMCollect percent complete and remaining durations for activities in the current look-ahead window from each section; five-minute conversation each, not a form.
Sunday AMEnter actual starts and finishes only. Do not reschedule the full network; run the look-ahead filter and export. Note any activity that won't start on time and why.
Sunday PMWeekly site meeting: last week's planned vs achieved; this week's commitments; next three weeks' activities; constraints (drawings, materials, access, approvals) with owner and date.
OngoingConstraint list updated as items clear; items not cleared by the week they're needed are escalated.

Percent plan complete โ€” planned activities that started on time divided by planned activities โ€” is worth tracking. Below 60% consistently means the look-ahead is fiction; above 85% means it's working.

3. The monthly routine. See monthly-update-checklist for the full sequence. In outline: freeze the data date; collect signed progress from every section; status actuals and remaining durations; reschedule; compare to last update (Schedule Comparison); review near-critical paths; apply approved changes only; produce the narrative; PM review; submit. Three to five working days from data date to submission on a typical building job.

4. Where the two meet. Weekly actual dates feed the monthly update โ€” by month-end the planner already has four weeks of actual starts and finishes and only needs remaining durations and percent complete for in-progress work. Monthly rescheduling resets the look-ahead: the first weekly meeting after the update uses the new dates.

5. Who does what.

RoleWeeklyMonthly
PlannerRuns the look-ahead export; facilitates the meeting; maintains the constraint listOwns the update start to finish
Construction managerChairs the meeting; commits to the week's workReviews remaining durations; signs off the forecast
Section engineersReport actual progress; identify constraintsSign the monthly progress sheet for their area
SubcontractorsAttend; commit; reportSubmit their monthly update in agreed format
Project managerInformedApproves the narrative before submission
QS / commercialโ€”Receives the update for valuation and any notices
๐ŸŸฅ Where people go wrong
  1. Rescheduling the full network weekly. Every reschedule produces new dates that the site then reacts to. Actuals weekly, reschedule monthly.
  1. Moving the monthly data date. "We'll update on the 3rd because the pour finished on the 2nd" turns a record into a negotiation. Same day, every month.
  1. Letting the look-ahead drift from P6. After two months of Excel edits the look-ahead shows a sequence P6 has never seen. Export it fresh each week.
  1. Weekly percent complete collected by email. Nobody answers. Walk the site or phone the section engineer.
  1. Using the weekly meeting for the delay argument. Weekly is for what happens next week. Delay is recorded contemporaneously in the daily diary and dealt with in the monthly narrative and notices.
  1. Skipping the monthly update because the recovery plan is "nearly ready". The update records what happened. The recovery plan proposes what will. You need both, and the update comes first.
โš–๏ธ When you're challenged

"Why can't the P6 schedule just be updated every week?" It can hold actual dates every week and it does. A full reschedule every week produces a new forecast every seven days, which nobody has time to absorb and which makes the monthly trend unreadable. The look-ahead gives the site the detail; the monthly update gives the contract the record.

"The look-ahead says plaster starts Tuesday but P6 says next month." P6 was rescheduled at the data date three weeks ago; the look-ahead reflects the section engineer's commitment made on Sunday. If the commitment holds, the next update will show it. If P6 and the look-ahead disagree by more than a week on a critical activity, that's the discussion for this week's meeting.

"Can we shift the update to the 5th this month to include the milestone?" No โ€” the data date is the 25th on every update since Baseline 0, and the milestone will appear as an actual in next month's. Moving the date once means every comparison against it is unreliable.

๐Ÿ“„ Related pages
โœ๏ธ Worked example

A residential tower in Sharjah, monthly data date 25th, weekly meeting Sunday 09:00.

Week ofWeekly outputMonthly link
1 Jun 25Look-ahead L12โ€“L15 structure, L3โ€“L6 blockwork; PPC 78%; constraint: L14 rebar shop drawings not approved, owner: consultant, need by 8 JunActual finishes for L11 slab entered
8 JunPPC 71%; L14 rebar delayed 3 days โ€” drawing approved 10 Jun; constraint escalated to PM on 8 JunDiary entry and email trail filed against activity A1240
15 JunPPC 82%; sequence recovered by pouring L14 in two halvesActuals entered
22 JunSummer hours start 15 Jun already in calendar; look-ahead re-exportedActuals entered; remaining durations collected for the update
25 Junโ€”Data date. Update 08 statused, rescheduled, narrative issued 30 Jun. L14 drawing delay: 3 days on a path with 5 days float; recorded in delay register; no notice required beyond the weekly correspondence, but the correspondence is now attached to the update.

The planner spent roughly four hours per week on the weekly cycle and four days on Update 08. The site had current dates every Sunday; the Engineer had a formal record every month; the L14 delay was documented on the day it happened rather than reconstructed six months later.

๐Ÿ“– References
  • FIDIC Conditions of Contract for Construction, 1999 โ€” Cl 4.21, 8.3; 2017 โ€” Cl 4.20, 8.3 (check the edition in your contract).
  • NEC3 ECC / NEC4 ECC โ€” Cl 32.2 (check the edition in your contract).
  • Society of Construction Law, Delay and Disruption Protocol, 2nd ed. (2017) โ€” Core Principle 1; Part B, records.
  • PMI, Practice Standard for Scheduling, 3rd ed. โ€” schedule maintenance.
  • Lean Construction Institute โ€” Last Planner System, weekly work planning.
  • Oracle Primavera P6 Professional User Guide โ€” Filters; Layouts; Apply Actuals.

From the field

Experience from working planners. Unreviewed โ€” read it as experience, not guidance.

Add what you know about weekly vs monthly cycles. What worked, what the consultant pushed back on, what you would do differently next time. A paragraph is plenty.

Contributors get their name and one link on the site โ€” your own templates, course or consultancy. We take nothing and hold nothing.

Add your experience