How to Schedule Wet Trades to Prevent Rework and Callbacks

Why do wet trades cause the most callbacks on Thai construction projects?

Wet trades — render and plaster, waterproofing, floor screed, and tile or stone finishing — are the final visible layer of any building, and they are also the most interdependent. Each trade applies a wet material that must bond to the layer below and reach a minimum cure time before the next trade can follow without risk of failure. When tiles are laid on a screed that is still green, or waterproofing is applied over plaster that still holds moisture, the result is hollow tiles, cracked screeds, and water ingress that show up not during the project but three to twelve months after handover — when the owner calls, the work is already covered up, and the dispute is expensive and time-consuming. Thai SME contractors lose significant money each year to wet trade callbacks that are entirely preventable with a well-planned sequence on the Gantt.

What are the wet trades and what order must they follow?

The wet trades on a typical Thai residential or commercial project follow a fixed hierarchy. First comes plaster and render, applied to masonry walls as a base for paint or tile, requiring a minimum seven-day cure in Bangkok temperatures before tiling or painting begins. Next is bathroom and balcony waterproofing, applied in wet areas, needing twenty-four to forty-eight hours between coats and a mandatory twenty-four-hour flood test before any floor tiles start. Third is floor screed, which levels the structural slab and requires fourteen to twenty-eight days cure before tiling, depending on thickness and mix. Last is tile and stone installation. Any step taken out of order, or any cure time compressed, creates a defect that may not be visible for months but will eventually show up as hollow tiles, delamination, or water staining that the contractor must fix at their own cost.

How do cure times become Gantt task durations in SiteBoard?

In SiteBoard's Gantt planning view, each cure period becomes a distinct task with a minimum duration, not a lag hidden inside another task. For bathroom waterproofing, model three tasks: apply first coat, apply second coat the following day, then a one-day flood test. Set the flood test finish as a predecessor for all bathroom floor tile tasks. For floor screed, create a screed pour task and a fourteen-day screed cure task, then set the cure finish as the predecessor for all tile-laying tasks on that floor. For wall plaster, set a seven-day plaster cure task before any wall tile or paint task begins. These cure tasks look like idle time on the schedule, but they are the minimum physics of the materials. Making the cure periods visible on the Gantt gives the project manager the earliest honest date for each finishing trade and protects the contractor from pressure to start tiling before the screed is structurally ready.

What is the most common sequencing mistake on Thai residential sites?

The single most damaging wet trade mistake on Thai residential sites is tiling too soon after screed — sometimes as few as three to five days after a pour when the site is under time pressure. Cement-sand screed at forty millimetres thick typically takes fourteen to twenty-one days to reach ninety percent of its design strength in Bangkok conditions. Tiles bonded to green screed will appear sound during installation, but as the screed continues to shrink and lose moisture, the bond breaks down. Tiles hollow out and within six to twelve months they begin to delaminate, crack, and sometimes tent, lifting off the floor. Re-tiling a 150-square-metre apartment after handover can cost 80,000 to 150,000 baht in labour and materials alone, plus dispute management time and reputational damage. The same project, properly sequenced with a fourteen-day screed cure dependency on the Gantt, would have been defect-free.

A real example: a Pathum Thani townhouse where wet trade rework cost 65,000 baht

A contractor in Pathum Thani built three linked townhouses at 3.2 million baht each on a twenty-six-week programme. After the floor screed was poured in week nineteen, the site supervisor allowed tiling to start in week twenty-one — only eight days after the pour — to keep the tile subcontractor on programme. Six weeks after handover the owner reported hollow tiles throughout the main bedroom and two bathrooms. Investigation found the screed had not cured fully before adhesive was applied. Re-tiling the affected areas cost 65,000 baht, disputed between the main contractor and the tile subcontractor. Had the Gantt planning view in SiteBoard included a fourteen-day screed cure task as a dependency before tile-laying, the site supervisor could not have started tiling early without visibly pushing the programme, and the project manager would have caught the conflict before work began.

How do you plan wet trades across multiple rooms and floors in SiteBoard?

On a multi-room project, the most efficient approach is to run wet trades in a zone-by-zone rotation rather than finishing one room at a time. In SiteBoard's Gantt planning view, group tasks by zone — ground floor wet areas, ground floor dry areas, first floor, and so on — and plan each zone as a mini sequence: plaster cure, waterproofing and flood test, screed cure, then tiles. Set dependencies so waterproofing in zone A starts on the day plaster reaches its cure date, and tile-laying in zone A cannot start until both the flood test and the screed cure are complete. With three or four zones in rotation, the tile subcontractor always has a cured zone ready to work in, eliminating idle time while still respecting cure physics. The on-site mobile updates in SiteBoard let the tile crew confirm a zone is ready before they mobilise, and the owner sharing link shows which rooms are tiled without a daily phone call.

How does the wet trades schedule connect to the owner's move-in date?

The wet trades schedule is usually the controlling path for the final six to eight weeks of a residential project. When the owner asks for a move-in commitment, the realistic earliest date must account for the minimum cure chain: screed poured, fourteen days cure, tiles laid, three days grout cure, then cleaning and final inspection. Planning backward from the requested move-in date in SiteBoard's Gantt planning view, adding the minimum cure periods back through the sequence, quickly shows whether the date is achievable. If the resulting screed pour date falls before the structural frame will be ready, the move-in date carries real defect risk. Presenting the owner with a cure-based schedule through the owner sharing link replaces an uncomfortable negotiation with a data point from the Gantt that both parties can see and reason about.

What to include in a wet trades progress update in SiteBoard

When tracking wet trade milestones in SiteBoard as on-site progress updates, record more than the percentage. For waterproofing, note the product name, coat applied, and date, and attach a photo of the flood test in progress. For screed, record the mix design, pour date, thickness at three sample points, and the date the cure begins. For tiles, photograph the screed moisture check if one was taken. These details take under two minutes to record from the site on a mobile phone, but they create a timestamped, photo-backed record that serves as evidence if a defect is later disputed. Thai courts and arbitration bodies consider contemporaneous site records as strong evidence; a photo of a completed flood test dated before tiling started is far more useful than a contractor's verbal assurance that the sequence was followed.

FAQ: Can you speed up screed curing when the programme is under pressure?

Screed curing can be legitimately accelerated by using a rapid-setting or fibre-reinforced screed mix designed for shorter cure periods, typically reaching adequate strength in seven to ten days rather than fourteen to twenty-eight. Rapid-set screeds cost roughly fifteen to thirty percent more per square metre than standard mixes. Keeping the screed surface moist for the first three days also improves early-strength gain in hot dry weather. What cannot be recommended is simply ignoring the cure period: no product accelerates bonding that has not yet occurred. If you switch to a rapid-set screed mid-project to recover time, update the predecessor duration in SiteBoard's Gantt planning view immediately so the tile start date reflects the actual safe date, not the original standard-screed assumption.

FAQ: Who is responsible for wet trade sequencing on a multi-subcontractor site?

Responsibility for wet trade sequencing sits with the main contractor or project manager, not with the individual subcontractors. The plastering crew will plaster when the site is ready; the tile subcontractor will tile when called; neither monitors cure periods unless the main contractor explicitly tracks them. In SiteBoard's Gantt planning view, the dependency chain between cure tasks and tile tasks makes the sequencing constraint visible to everyone with access to the project. When a subcontractor calls asking to mobilise early, the project manager can open the Gantt, see that the screed is on day nine of fourteen, and confidently say no rather than deferring to memory. Setting these dependencies is not complex — three to five tasks per zone, taking thirty minutes to set up once at the start of the project.

FAQ: How do you use the wet trades record from SiteBoard after handover?

After handover, the most valuable record from the wet trades schedule is the timestamped sequence of on-site progress updates in SiteBoard, particularly the screed pour date, the waterproofing flood test photo, and the tile start date. If an owner reports hollow tiles twelve months after handover and the contractor can show, through the SiteBoard progress record, that tile-laying started fourteen days after screed pour, the contractor has a defensible record that the sequence was followed correctly and the defect may have another cause. Thai construction disputes often turn on whether the contractor followed the specification. A dated, photo-backed record from SiteBoard is evidence that an Excel file or a LINE chat history cannot reliably provide, and it costs nothing extra to create during the build.