Leak Source Diagnosis and Isolation Procedure
Purpose
Water travels before it drips. A stain on the soffit or a wet spot on the fascia tells you where water ended up, not where it got in, and repairing the visible spot without tracing the actual entry point leaves the real leak open. This procedure is the standing test that runs before any repair scope is chosen: it forces an active water test that names the true entry point, and it forces a check of the space behind the gutter, because a leak that looks like a gutter problem is sometimes a roof-to-fascia problem wearing a gutter's clothes.
The failure this prevents is a reseal or a rehang performed on the wrong joint while the water keeps coming in somewhere else, at the customer's expense twice.
Scope
Covers the physical isolation test, an interview review, a ground-level walk, a section-by-section hose test, and a behind-gutter check, used to trace any gutter-attributed leak to its physical source before a repair SOP starts. Feeds its finding into the reseal-vs-rehang-vs-replace matrix and hands off to the seam reseal or sagging section rehang standard procedure once the source is confirmed.
It does not translate symptom timing into a hypothesis; the soffit-stain, apron-versus-pitch, and single-observable stain-location decision trees own reading the clue before you climb. It does not perform the repair itself. Where the confirmed source sits behind the gutter, in the roof, apron, or drip edge, this procedure ends at a referral, it does not attempt roofing work.
Roles and responsibilities
| Role | Owns | Hands off |
|---|---|---|
| Office at booking | The reported location, timing, and weather conditions of the leak | That detail on the ticket; a leak reported only during ice or only after a hard freeze flags the tech to expect a possible non-reproducing test |
| Technician | The procedure below, the classification, the record | A confirmed source classified for repair, or a documented referral out of gutter scope |
| Office after the visit | Routing the classified finding | The repair booking if gutter-confirmed, or the referral contact if the source is roofing or fascia |
Before you climb
Ladder placement and overhead lines are the standing hazard for this whole visit. Set the ladder at a 4:1 ratio, tie off or foot it with a spotter, and stay at least 10 feet from any service drop.
A hose test runs water near soffit lighting and outdoor receptacles. Confirm any powered fixture near the test area before running water, and route the hose so the stream never crosses a live fixture; treat any fixture you can't confirm de-energized as live.
Pulling trim to check behind the gutter can open a wasp or hornet void. Look before reaching into a gap you've just opened, and back off rather than working through active activity.
The procedure
Record the reported symptom in the customer's own words, with location and conditions. Confirm where they see water, when, and under what weather (steady rain, wind-driven rain, only after ice, only after a hard freeze). Acceptance: a specific location and a specific triggering condition on the ticket, not "it leaks." Wrong looks like a report with no timing detail, which leaves you unable to tell later whether your test reproduced the actual complaint. Stop rule: none, incomplete information gets filled in by asking, not by guessing. Hazard: none, this is a conversation.
Confirm dry conditions before testing. Check that at least 24 hours have passed since the last rain and that trim, fascia, and soffit are dry to the touch. Acceptance: no residual moisture anywhere in the test area. Wrong looks like damp trim from yesterday's storm that you're about to re-wet with a test hose, making new water indistinguishable from old. Stop rule: fewer than 24 dry hours means wait or reschedule; a test run over residual moisture produces a finding nobody can trust. Hazard: none.
Walk the full run from the ground before climbing. Look for staining patterns, sag, moss lines, and discoloration along the entire run, not just at the reported spot, and photograph what you find. Acceptance: every visible clue on the run logged with a photo, including anything the customer didn't mention. Wrong looks like climbing straight to the reported location and missing a second source elsewhere on the same run. Stop rule: none, but any additional finding gets added to the section-by-section plan in step 5. Hazard: none, ground-level work.
Set up a controlled water source and let it reach steady flow before judging anything. Open the hose fully and run it about 30 seconds before starting to observe. Acceptance: a steady, consistent stream confirmed at the nozzle, not the surge and sputter of a line just turned on. Wrong looks like reading a result off the first few seconds of fluctuating pressure. Stop rule: none. Hazard: none beyond the standing ladder hazard.
Test section by section from the high point toward the downspout, watching the underside for the first point water actually emerges. Pour a measured amount at the top of each section in turn and track the stream along the underside rather than watching the puddle forming below. Acceptance: the leak named to a specific joint, hanger penetration, or pinhole, confirmed by a second pour at that same spot producing the same result. Wrong looks like a drip further downstream that traveled along the underside seam from an upstream origin, the classic false positive that sends a tech to reseal the wrong joint. Stop rule: if the first visible drip point doesn't reproduce cleanly on a repeat pour at that exact spot, back up one section and retest until the true first-emergence point is found. Hazard: none beyond the standing ladder hazard.
Rule out a behind-gutter path at the confirmed point. Pull back the trim or use a mirror or borescope to check the space directly behind the gutter at the point named in step 5. Acceptance: that space confirmed dry during an active hose test at the same point, meaning the gutter joint itself is the source; or confirmed wet, meaning water is entering from behind the gutter, at the apron, drip edge, or roof-to-fascia interface, not from the gutter. Wrong looks like assuming a joint failure without checking behind it, since a behind-gutter leak can drip and pool at the exact same visible spot as a gutter joint. Stop rule: any confirmed wet path behind the gutter routes this leak out of gutter repair scope to a roofing or fascia referral, not to a reseal. Hazard: a hand goes into a freshly opened gap here; per the entry block, look before reaching in.
Where the leak won't reproduce under a clean, dry-start hose test, stop testing and document the trigger condition instead. This applies to a leak reported only during ice-dam conditions or only after a hard freeze. Acceptance: the reported condition matches a known temperature-triggered pattern, logged with a note routing to the applicable decision tree rather than a repair. Wrong looks like repeating the same hose test several times at the same mild ambient temperature hoping for a different result. Stop rule: after one clean, properly dry-started test with no reproduction, stop and document; do not guess a repair scope from an unreproduced symptom. Hazard: none.
Classify the confirmed source and hand off. Record the finding as reseal, rehang, replace, or non-gutter referral, with the supporting photos attached. Acceptance: a written classification on the ticket, not left in the tech's memory. Wrong looks like a verbal handoff with no record, which means the next visit repeats this entire procedure from scratch. Stop rule: none. Hazard: none.
The record this produces
A source-isolation record: the customer's reported symptom and conditions, the dry-window confirmation, the ground-level findings with photos, the section-by-section test log showing where water was and wasn't reproduced, the behind-gutter finding at the confirmed point, and the final classification. Where the leak didn't reproduce, the record states the trigger condition instead of a false classification.
The behind-gutter finding matters most. It is the one field that stops a reseal being billed and performed on a leak that was never coming from the gutter at all.
One run of this procedure, filled in
A customer reports water at the soffit near one corner, worse in heavy rain, no mention of ice or cold.
Step 1: location and "worse in heavy rain" logged. Step 2: last rain was two days prior, trim dry to the touch. Step 3: ground walk shows a faint stain at the reported corner and nothing else on the run. Step 4: hose opened and run 30 seconds to steady flow. Step 5: pour at the corner section produces weep at a hanger penetration two feet from the corner on the first pass; a repeat pour at that exact spot reproduces the same weep. Confirmed.
Step 6: trim pulled back at the hanger penetration. The space behind the gutter is wet during the same active pour, tracking up toward the roof edge rather than staying at the hanger. Fail against the gutter-source hypothesis. Stop rule taken: the confirmed path is behind the gutter, not through the hanger penetration itself; classification routes to a fascia and drip-edge referral rather than the seam reseal procedure.
Step 7: not applicable, the leak reproduced cleanly. Step 8: classified as non-gutter, roofing and drip-edge referral, with before photos of the wet path behind the trim attached to the ticket.
The job turned on step 6. Without pulling the trim, this call would have been booked and billed as a hanger-penetration reseal, and the customer would have called back after the next heavy rain with the exact same stain, because the seam was never the actual entry point.
When the site doesn't match the happy path
No dry window is available before the appointment ends. Reschedule the hose test rather than testing over residual moisture; a finding from a compromised test is worse than no finding, because it gets acted on with the same confidence as a clean one.
The customer reports the leak only during an ice dam or a hard cold snap. Do not spend the visit repeating a hose test that cannot reproduce a freeze-triggered condition; document what was found and route to the applicable temperature-trigger decision tree.
Two separate sources are found on the same run. Classify and record each independently; a run with a confirmed gutter-source joint and a confirmed behind-gutter path is two jobs, not one, and billing or repairing only one leaves the other leak live.
References
- See related: Reseal vs Rehang vs Replace Section on a Chronic Leak: Method Decision Matrix (the scope decision this procedure's finding feeds into).
- See related: Seam Reseal Standard Procedure and Sagging Section Rehang Standard Procedure (the repair procedures this diagnosis hands off to).
- See related: Soffit Water Stain Versus Active Leak Decision Tree; Customer Reports Dripping Behind the Gutter: Apron vs Pitch vs Backsplash Translation Decision Tree; Water Stain Location Tells the Leak Source: Seam vs End vs Hanger Single-Observable Decision Tree (own reading the clue before this procedure's active test confirms it).
- See related: Sectional Gutter Seam Leaks Only During Ice Dam Conditions: Temperature-Trigger Decision Tree; Gutter Leaks at a Seam Only After a Cold Snap: Decision Tree (own the condition this procedure documents rather than chases in step 7).
- SMACNA Architectural Sheet Metal Manual, gutter and drip-edge interface detailing.