Overtime and Shift Work on the Gantt: A Recovery Guide

Why does adding overtime rarely deliver the recovery contractors expect?

When a project falls two or three weeks behind, the first instinct is to extend working hours, push through weekends, and tell the crew to catch up. In practice, the recovery is almost always smaller than the time added. A construction team working ten-hour days for three consecutive weeks rarely produces 25 percent more output than a standard eight-hour day. Fatigue accumulates, error rates rise, rework increases, and supervisors who are also working extended hours make decisions with less care. Studies on construction labor productivity consistently find that sustained overtime reduces effective output per labor-hour by 10 to 25 percent after the first two weeks. On a Thai site in the dry season this is compounded by heat stress from extended afternoon work. The lesson is not that overtime never works — it does, in short bursts — but that a recovery plan built only on asking people to work longer will underdeliver unless you account for the productivity cost and choose the right method for each trade.

When does a night shift make more sense than extending the working day?

A night shift — a separate crew starting after the day shift ends — is fundamentally different from asking the same people to stay late. A dedicated second shift maintains each crew at a normal productivity level rather than grinding one team into fatigue. Night shifts work best when the work can be fully separated between teams without coordination overhead: structural waterproofing, large-area external render, floor screed across open bays, or concrete pours in the cooler evening hours. They work poorly when trades are tightly interlocked in a small space — MEP rough-in in a narrow ceiling void, tiling in a bathroom where two workers already fill the room — because handover errors create rework that costs more than the extra hours gain. Night-shift labor in Thailand typically costs 1.5 to 2 times the day rate. Include site lighting, safety supervision, and foreman coverage in the recovery cost estimate before committing a revised end date to the owner.

How to model accelerated working in SiteBoard's Gantt planning view

SiteBoard's Gantt planning view tracks tasks by calendar-day duration. The simplest way to model an accelerated schedule is to shorten the duration of affected tasks to reflect the higher daily output, and add a note in the task description explaining that the shorter duration assumes overtime or two-shift working. For a task that normally takes 10 days with a single day crew, a genuine two-shift plan on a large open work area might bring it to 6 or 7 days — not 5, because the productivity benefit is not a clean 2x due to shift handover time and material coordination. Link the accelerated tasks with finish-to-start dependencies as usual, and update on-site progress updates daily so you can see in week one whether the acceleration is hitting the target pace. Catching a shortfall early gives you time to adjust before you have told the owner a recovery date you cannot meet.

A worked example: recovering 15 lost days on a villa project in Chiang Mai

A contractor in Chiang Mai was building a three-bedroom villa on a 100-day contract. At day 55, late delivery of the roofing system and a mid-project subcontractor replacement for MEP rough-in had put the project 15 working days behind. The contractor identified three remaining tasks with genuinely independent, large work areas: basement waterproofing, external render across two large wall faces, and ground-floor screed. For these three tasks only, they added a four-person evening crew from 5 pm to 10 pm — a three-hour extension rather than a full separate night shift — delivering roughly 35 percent more productive hours per day on those tasks. On the Gantt, the durations were shortened from 10, 8, and 6 days to 7, 6, and 4 days. Total recovery was 7 working days at an additional labor cost of around 45,000 baht. The remaining 8-day gap was closed by resequencing two interior fit-out tasks to run concurrently. The project closed 2 days late — a significant improvement on the projected 15-day overrun.

How to account for productivity loss in your recovery estimate

When building a recovery schedule, never assume that overtime hours produce the same output as standard hours. A conservative planning assumption for Thai construction is that hours beyond the standard eight-hour day produce 70 to 80 percent of the output of a normal hour for the first two weeks, dropping to 60 to 70 percent in weeks three and four if overtime is continuous. A dedicated night crew on genuinely independent tasks can maintain closer to 85 to 90 percent productivity for the first three weeks before fatigue shows. Applying this logic: a task that takes 10 days at 8 hours per day, run instead at 11 hours per day with a 25 percent discount on the extra three hours, yields roughly 8.75 standard-equivalent hours of output per day — a saving of less than one day per ten. Knowing this number prevents you from promising a two-week recovery when the plan will only deliver five days.

How far ahead should a recovery Gantt extend?

A recovery schedule should run all the way to the handover milestone, not just to the point where you expect to have caught up. Planning only a few weeks of recovery hides downstream risks: saving two weeks in the structural phase can still result in a late project if the MEP resource plan was not updated at the same time. In the Gantt planning view, adjust all affected tasks from the current date to handover, update durations based on realistic productivity assumptions, and walk through each critical-path task to confirm the end date has genuinely moved. This full-horizon review is the difference between a plan that recovers time on paper and one that delivers on site. Once the recovery schedule is confirmed, share the updated Gantt through the owner sharing link so the owner sees a revised end date with a credible week-by-week path, not just a verbal assurance.

Keeping the owner informed during a recovery period

The most damaging version of a recovery is one the owner learns about only when the original handover date arrives and passes. Sharing the revised Gantt through the owner sharing link as soon as the recovery plan is confirmed changes the experience entirely. The owner sees each task's current percentage complete, the revised durations, and the new projected end date updating in real time as the foreman enters on-site progress updates each day. When the acceleration is working and tasks are finishing ahead of the revised plan, the owner's confidence in the team actually increases despite the original delay. Contractors who share the recovery plan openly and maintain daily on-site progress updates during the acceleration consistently report fewer payment disputes, because the owner has been a live witness to every adjustment.

FAQ: How long can a construction crew realistically sustain overtime?

Three to four weeks of extended hours — no more than two to three extra hours per day — is manageable for most trades in Thai residential and commercial construction without a significant quality drop, provided supervisors also rotate and get rest days. Beyond four continuous weeks, error rates and rework tend to climb faster than the extra hours can recover, turning the acceleration into a net loss. If the recovery plan requires more than four weeks of overtime, the honest assessment is that the project will be late and the goal should shift to minimising the overrun and communicating a credible revised end date to the owner. A two-phase approach — intensive acceleration for three weeks followed by standard hours and a frank owner update — is almost always more effective than a sustained push that collapses in week five.

FAQ: Does adding a second shift always cut the remaining duration in half?

No — and a contractor who plans on that assumption will miss the recovery target. The theoretical 2x comes from doubling productive hours, but in practice two-shift working on a construction site delivers roughly 1.4 to 1.6 times the daily output of a single shift, not 2x. The gap comes from shift handover time, setup and cleanup overhead, reduced pace in the transition window, and the difficulty of keeping material supply running for an evening crew. The practical rule is to plan for 1.5x daily progress on a two-shift task and treat anything beyond that as upside. If the first week shows stronger performance, tighten the Gantt at that point — do not tighten it before you have seen the data.

FAQ: Should you update the Gantt when starting a recovery schedule?

Yes — update the plan in SiteBoard's Gantt planning view to reflect the recovery schedule as soon as it is confirmed and communicated to the team. A plan that predates the recovery will show every task as late even when the recovery is working, which makes the data unreadable and discourages the team from trusting it. Set the recovery schedule as the active working plan from the start of acceleration, record the original contract end date in the project notes for the owner conversation, and track on-site progress updates against the recovery timeline. The key discipline is maintaining one honest, live plan that the whole team navigates by — not an outdated one that requires constant explanation.