Snagging is what you do when quality control failed earlier. An inspection and test plan moves the check to before the work is buried. Here's how to build one that gets used.
Think about what a snagging list actually is. It's a record of work that was done wrong, discovered after the fact, that now has to be taken apart and redone. Every item on it is a small failure of quality control that happened earlier — often much earlier, sometimes behind a wall or under a screed that now has to come up.
Snagging is necessary. But a long snag list isn't a sign of thorough close-out — it's a sign that quality was being inspected out at the end rather than built in along the way. The tool that shifts you from one to the other is the inspection and test plan, the ITP. And most small contractors either don't use one or treat it as a form to file rather than a system to run.
An inspection and test plan sets out, for a specific element of work, the quality checks that will happen, when they'll happen, who does them, against what standard, and what evidence gets recorded. It answers a simple question for every important stage of the work: before we go any further, how do we confirm this is right?
Take a drainage run. The ITP might say: check the trench formation and bedding before pipes are laid; check the pipe falls and joints before backfill; carry out an air or water test before the trench is closed; record the results. Each of those is a checkpoint placed before the work becomes impossible or expensive to inspect. Once that trench is backfilled, checking the bedding means digging it up. The ITP makes sure you looked while looking was easy.
That's the whole philosophy: put the inspection where it's cheap, not where it's expensive.
Two terms worth knowing, because they're what makes an ITP more than a checklist:
The skill in writing an ITP is putting hold points where they matter and not where they don't. Make everything a hold point and the job grinds to a halt waiting for sign-offs. Make nothing a hold point and the critical checks get skipped under programme pressure. Reserve hold points for the work that's buried, safety-critical, or irreversible.
Here's the economic argument, because it's the one that lands. The cost of putting a defect right multiplies the later you catch it.
An ITP is a machine for catching defects at the top of that list instead of the bottom. It doesn't eliminate snagging entirely — nothing does — but it dramatically shrinks the list, and it moves the checks to the point where fixing things is cheap and quick. On the projects where I've seen ITPs used properly, the close-out was calm, because most of what would have been on the snag list had already been caught and fixed weeks earlier, in place, before anything covered it.
The failure mode with ITPs is producing them to satisfy a quality requirement, then never actually running them. A plan nobody follows is worse than no plan — it's false comfort. To make one that works:
1. Cover the elements that matter. You don't ITP everything. You ITP the work that's buried, safety-critical, hard to inspect later, or a common source of defects — drainage, waterproofing, structural connections, M&E first fix, anything going behind a finish.
2. Tie each check to a real standard. "Check it's OK" is useless. The standard is the specification, the drawing, the manufacturer's requirements, the relevant code. The inspection confirms compliance with something specific.
3. Set the hold and witness points deliberately. Where must work genuinely stop for a check? Where does someone else need the chance to attend? Mark them, and set the notice periods for witness points so people get fair warning.
4. Make recording easy, and do it at the point of inspection. Photos, results, sign-off, dated — captured when the check happens, not written up later. The evidence is worth as much as the check itself: it proves the work was right when it was covered, which is gold if a defect is ever alleged later.
5. Assign ownership. Who carries out each check, and who signs the hold points? An ITP with no named responsibility doesn't get run.
There's a second payoff that people underrate. An ITP run properly produces a contemporaneous record that the work was inspected and correct at each stage — dated, photographed, signed. If a defect is alleged after handover, or a dispute arises about workmanship, that record is your defence. "Here's the inspection, here's the test result, here's the photo, signed on this date" is a very different position from trying to remember whether anyone checked the bedding eighteen months ago. Like a good site diary, the ITP record wins arguments — and increasingly, on higher-risk buildings, that structured quality evidence is exactly what the golden thread expects you to have.
At Construction AI, inspections are structured checks against the specification, with pass/fail line items, photos and sign-off captured on site at the point of inspection — hold and witness points recorded, evidence attached, all tied to the element of work. Instead of an ITP that lives in a folder and never gets run, the checks happen in the flow of the work and the record assembles itself. Which means when close-out comes, the quality was built in, and you can prove it.
Snagging will always exist, but it should be the exception, not the plan. Put your inspections where the work is still open, hold the points that genuinely matter, record the evidence as you go, and you'll close out calmer, cheaper, and with a record that protects you long after the client's moved in. Build quality in. Don't inspect it out.
What is an inspection and test plan (ITP) in construction?
An ITP sets out, for a specific element of work, the quality checks that will happen, when, who does them, against what standard, and what evidence is recorded — placing inspections before the work is covered or buried so defects are caught early.
What's the difference between a hold point and a witness point?
A hold point is a mandatory stop: work cannot proceed until the inspection is done and signed off. A witness point is a check a second party (client's rep, CA, building control) is invited to attend; if they don't attend within the notice period, work can usually proceed.
Why is an ITP better than relying on snagging?
Snagging catches defects after the fact, when they're expensive to fix and may need finished work ripped out. An ITP catches them while the work is still open and cheap to correct. The cost of a defect multiplies the later it's found.
Which elements of work should have an ITP?
The work that's buried, safety-critical, hard to inspect later, or a common source of defects — drainage, waterproofing, structural connections, M&E first fix, and anything going behind a finish. You don't ITP everything.
Does an ITP help with disputes or the Building Safety Act golden thread?
Yes. A properly run ITP produces a dated, photographed, signed record that work was inspected and correct at each stage — a strong defence against later defect allegations, and exactly the kind of structured quality evidence the golden thread expects on higher-risk buildings.
Stephen Mckenna MCIOB
30+ years in UK commercial construction, from site management to director level. Now building the project management tools he wished he'd had.
Tier-one contractor processes shouldn't only be available to tier-one contractors. Drawing registers, document control, RFIs, programmes, financial tracking — built for how you actually run projects, priced so any construction business can access them. Try everything free for 30 days.
By subscribing, you agree to our Privacy Policy.
The last 5% of every project takes 50% of the effort. Here's how to manage defects and snagging lists, and close out without months of painful callbacks.
A good site diary has won more disputes than any lawyer. Here's what to record, how to record it, and why site diary software beats a notebook in adjudication.
Tier-one processes, priced for any size business. Try everything free for 30 days — no charge until your trial ends.