Roof recover vs. full tear-off on a Vancouver low-slope roof: when an overlay is honest and when it hides a problem

How a roof recover actually works

A recover leaves the old roof in place and installs a new membrane over it. In practice it is not just gluing a new sheet onto the old one. A proper recover starts by removing loose gravel, blisters, and failed flashing, then laying a cover board (usually a firm, moisture-resistant board) so the new membrane has a smooth, dry, uniform surface to bond to. On some assemblies an insulation overlay board goes down first to add thermal value and smooth out the old surface further. Then the new membrane goes on: mechanically fastened, adhered, or torch-applied, depending on the system.

On paper this is attractive. No tear-off labour, no bins of old roofing hauled through the strata loading bay, no deck sitting exposed to a Vancouver rain cell overnight. The catch is that you never see what is underneath. The new membrane follows the shape of the existing roof, and any wet insulation, rotted wood deck, or corroded metal decking stays sealed in, out of sight, still working against the building. A recover is a good roof only when what it covers is already a good roof.

How a full tear-off works

A tear-off strips everything above the structural deck. The old membrane, the old insulation, the old cover board, and the failed flashings all come off and get hauled away. With the deck open, the crew inspects it: they look for corrosion on a metal deck, rot or delamination in a wood deck, fastener pull-out, and low spots that pond water. Anything wet or failed is repaired or replaced while it is visible. Then a full new assembly is built from the deck up: a vapour or air barrier, insulation, a cover board, and the new membrane, all installed to the manufacturer specification.

The value of a tear-off is not the new membrane, which you get either way. It is the inspection and the fresh, dry base. You start the new roof over a known condition instead of an unknown one, and you document the deck so the next council knows what they have. That is why a tear-off is the honest answer whenever there is real doubt about what the old roof is hiding.

The layer limit is the first thing to confirm

General building code practice in Canada, following the International Building Code and most Canadian jurisdictions, caps a low-slope roof at two total coverings. The original roof, plus one overlay. Once a roof carries two layers, a recover is off the table and the roof has to come off to the deck. The limit exists for good reasons: stacked layers add dead load the deck was never sized for, they trap heat and moisture between membranes, and a multi-layer roof is very hard to diagnose when it starts to leak.

So before anyone talks price, confirm how many layers the roof already has. It is not always obvious from the surface, and old work orders are not always accurate, especially on buildings from the leaky-condo era where roofs were sometimes patched and overlaid more than the paperwork shows. A test cut through the assembly settles it. Because this is a hard limit rather than a preference, verify the specific requirement with your local building authority and the membrane manufacturer, not with a rule of thumb. In BC, roofing provisions for smaller Part 9 buildings live in Section 9.26 of the BC Building Code, and RCABC roofing practice standards govern most commercial and strata work. If you are weighing this against a broader repair scope, our page on envelope repair vs. recladding covers how a roof decision fits into a larger building plan.

Trapped moisture is the risk that gets people

A recover seals in whatever water is already in the assembly. If the existing insulation is wet and you overlay a new membrane, that moisture has nowhere to escape. It keeps degrading the insulation and attacking the deck from below while the roof looks perfect from above. Wet insulation also loses most of its thermal value, so the building pays more to heat and cool while the hidden damage grows. By the time it shows up as a leak or a soft deck, you are looking at a full tear-off you thought you had deferred for years, plus deck repair you could have caught earlier.

On a metal deck the risk is sharper. Water held against steel under wet insulation corrodes it, and corroded decking loses structural capacity you cannot see from the roof surface. On a wood deck the same trapped water rots the sheathing. Either way, the overlay that was supposed to save money becomes the reason a roof project turns into a structural repair. This is not a rare edge case in Metro Vancouver. It is the normal outcome of overlaying a wet roof in a wet climate.

Why the moisture survey is the deciding test

You cannot judge a recover by eye. The deciding test is a roof moisture survey, done before any decision. There are three common methods, and a good roofer uses them together. An infrared scan reads temperature differences on the roof surface after the sun sets, because wet areas hold heat differently than dry insulation, so they show up as warm patterns. It has limits: shaded areas and cool overcast evenings can hide the pattern, which is why it is verified. A nuclear gauge or a capacitance meter reads moisture directly, the nuclear gauge by sensing hydrogen in trapped water and the capacitance meter by reading how easily current passes through wet versus dry material. Core cuts, small sections cut out of the roof, confirm what the meters found and also reveal the true layer count. ASTM C1153 is the standard that governs how infrared survey results get verified with cores.

The survey is the difference between a defensible recover and a gamble. A clean, dry scan means an overlay is a reasonable use of strata funds. A scan showing saturation means the recover would just be hiding a problem. This one step, more than any other, separates a recover that lasts from one that fails in a few years. If water is already tracking into the building, our waterproofing team can help find the source before any roofing approach is chosen.

Added dead load and the structural check

A recover is not weightless. You add a cover board, sometimes an insulation overlay board, and a new membrane on top of the roof that is already there. On a newer building designed with headroom in the structure, that added dead load is usually fine. On an older deck sized decades ago for a single roof, it may not be, and it needs an engineer to confirm the deck can carry it before you overlay. This check matters most on the buildings people most want to recover: older strata and commercial roofs where the budget is tight and the structure was never generous to begin with.

A tear-off sidesteps this problem, because you remove one roof and install one roof, so the deck carries roughly what it always did. That is another reason a roof already at two layers has to go to the deck: a third layer piles even more weight onto a structure that has already taken on more than it was designed for. When a proposal recommends a recover, ask whether the deck load was checked. If the answer is a shrug, the scope is not finished.

Why the BC coast changes the answer

Metro Vancouver sees over 1,100 mm of rain a year, and low-slope roofs sit under that rain for a large share of the calendar. Aging assemblies here take on water in the insulation more often than owners expect, especially where drainage is poor or flashings have failed. That raises the odds that any given old roof is holding water, which is exactly why the moisture survey is not optional on the coast. The same climate shrinks the window for a tear-off, because the deck cannot sit open through a rain event, so tear-off work has to be staged by section and timed around dry days.

Elevation adds freeze-thaw. Buildings higher up on the North Shore and in the eastern suburbs freeze overnight in winter. Water trapped in a roof at those elevations freezes, expands, and cracks the assembly, then thaws and drives deeper, doing more damage each cycle. On a higher-elevation building, trapped moisture is not just a slow rot problem, it is an active freeze-thaw problem, and that tilts the honest answer toward a tear-off even when the layer count alone would allow an overlay. The wetter and colder the exposure, the less forgiving a recover over a marginal roof becomes.

How each option fails early

A recover fails from two directions. From below, trapped moisture in wet insulation keeps working on the deck and shortens the life of the new membrane, so a roof that should last decades starts leaking in a few years. From above, any detail left unresolved, a failed parapet flashing, a cracked penetration, a bad drain, leaks the same as it did before, and because the water now runs under a fresh membrane, it travels sideways and is harder to trace. Most failed recovers in Metro Vancouver trace back to a skipped moisture survey or a detail that was papered over instead of repaired.

A tear-off fails mainly from workmanship, not hidden conditions. The common failures are bad detailing at parapets and penetrations, seams that are not properly welded or torched, a deck repair skipped to save a day, or a new roof laid over the same ponding without correcting the slope. The difference is that these are visible during the work. A roof consultant or a diligent building envelope contractor inspecting the job can catch them before the roof closes up, which is why third-party inspection on a tear-off pays for itself.

What the cost actually depends on

On a recover, the cost drivers are the cover board and membrane system, the amount of failed flashing and detail work that has to be redone, and the moisture survey and structural check. Tear-off is stripped out, so labour and disposal are lower. The number that quietly decides everything is whether the survey was done, because an overlay on a wet roof is not cheap at all once you add the eventual tear-off it forces.

On a tear-off, the drivers are the tear-off labour and disposal, the new assembly, the detail work, and a deck-and-insulation contingency for whatever the crew finds once the roof is open. That contingency is the honest part of a tear-off quote. A proposal with no allowance for deck repair is either optimistic or hiding the number. We do not publish dollar figures because they swing too much with roof size, access, membrane type, and what the deck turns out to need, but the drivers above are what separate a real quote from a low-ball one. For a fuller picture of how these scopes are priced and staged, see our building envelope repair service.

Lifecycle cost, not just this year's price

The right way to compare the two is over the full life of the roof, not on the quote for this year. A recover over a sound, dry, single-layer roof can deliver a full new-membrane service life at a lower price, which makes it genuinely cheaper over the cycle for that building. A recover over a compromised assembly is the opposite: you pay for the overlay, the trapped moisture cuts its life short, and then you pay for the full tear-off you were trying to avoid, plus the extra deck damage the trapped water caused in the meantime. That is why a cheap recover over a wet or two-layer roof is a false economy, not a saving.

There is also the layer math. Every defensible recover still uses up the one overlay the code allows, so it locks in a mandatory tear-off as the next roofing event. A strata council doing lifecycle planning should treat the recover and that future tear-off as two line items, both in the depreciation report, so the reserve fund is sized for the real remaining cost, not just the low number this year. Our sibling guide on parkade traffic membrane vs. sealer walks through the same lifecycle logic for a different assembly.

Warranty reality

A warranty on a recover is only as good as the substrate it sits on. Membrane manufacturers set conditions for a recover, and a common one is that the existing assembly must be dry and sound. The RCABC RoofStar Guarantee, which covers a lot of commercial and strata roofing in BC, requires the existing roof to be demonstrably dry and free of wet material, so wet insulation has to be removed before the guarantee applies. Overlay a wet roof to save the removal cost and you can end up with a new membrane that carries no meaningful warranty, because the substrate never met the manufacturer condition.

A tear-off makes the warranty simple. With a known dry deck and a new assembly built to specification, a full material and labour warranty, including a RoofStar Guarantee, is straightforward to obtain. So if the warranty coverage you want depends on a documented dry substrate, that requirement by itself can push the honest answer toward a tear-off. Get the manufacturer warranty terms in writing before you approve a recover, and confirm the installer actually meets the substrate condition, not just the membrane spec.

Where a recover is genuinely the right call

None of this means recover is the lazy answer. On the right roof it is the smart one. When the roof is a single covering, a moisture survey confirms it is dry, the deck is sound and has the load capacity, and drainage works, a recover gives you a full new-membrane life at lower cost and far less disruption to residents. That is a legitimate use of strata funds, especially when the reserve is not ready for a full replacement and the depreciation report supports buying planned years. The honest requirement is only that the council budget the future tear-off as a second event, because the overlay uses the one allowed layer.

The mistake is not choosing a recover. The mistake is choosing it without the two checks, or choosing a tear-off on a roof that did not need one. Both errors come from deciding on price instead of on the moisture condition and the layer count. Get those facts, and the right call for that specific roof follows from them.

What neither option changes

Neither approach fixes a roof that is leaking through its details rather than its field. Failed flashings, open parapet joints, and cracked penetrations still have to be addressed regardless of whether you recover or tear off, and a new membrane laid over an unresolved detail leak will disappoint fast. The same goes for drainage: a roof that ponds because of low spots or blocked drains needs the slope and the drains corrected, not just a new surface over the same puddles.

For a building where the roof problem is really one symptom of a wider envelope issue, the roof decision should be scoped alongside the walls, windows, and sealant, not in isolation. Our building envelope repair team assesses the roof, the details, and the rest of the assembly together, and our building restoration crews handle the larger resets, so the roofing budget is based on the whole picture instead of one surface. If your building envelope decision is really about walls, the sibling guides on rainscreen vs. face-sealed walls and torch-on vs. single-ply membranes cover the related choices.

How to read a roofing proposal for this scope

A sound proposal shows its work. It states the confirmed layer count and how it was verified (a test cut, not a guess), it includes or references a moisture survey with the method used, it describes the deck and drainage condition, and it names the cover board and membrane system with the manufacturer warranty terms in writing. It addresses the details by name: parapets, penetrations, drains, and flashings. On a recover it confirms the deck load was checked. On a tear-off it carries a contingency for deck and insulation repair found once the roof is open.

A weak proposal leads with a low number and leaves those items out. It states the layer count with no test cut, skips the survey, says nothing about the deck, glosses over the details, and treats the recover as a permanent fix rather than one that uses the allowed overlay. When two bids are far apart, the cheaper one has usually dropped the survey, the cover board, the detail work, or the deck contingency, which are the exact items that decide whether the roof lasts on the coast. A proposal that omits them is not cheaper, it just moves the cost to a future emergency.

Decision framework: ten questions that point to the right approach