Can My Roof Be Repaired, or Does It Need Replacing?

Five things decide it. The age of the surface. How much of it is affected. How many repairs it has had. Whether there is one leak source or several. Whether the deck underneath is sound. Lean toward repair when the answers are young, localised, few, one, and yes. Material matters as much as age: twenty years is halfway through a tile roof and near the end of a shingle one.

Rancho Roofing is a licensed C-39 contractor in Encinitas, CSLB #1153860. This page gives you criteria you can apply yourself, rather than a conclusion.

That is deliberate. This is the question where a contractor’s incentive and a homeowner’s interest point in different directions, so the useful thing to hand over is the framework, not the verdict.

What are the five criteria?

Age, extent, repair history, number of leak sources, and structural condition.

Cert-A-Roof’s May 2026 framework sets them out clearly, and they work as a set rather than individually. One answer on the replacement side does not settle it. Four of them usually do.

Two of the five you can answer yourself today. You know roughly how old the surface is, and you know how many times somebody has been out to fix it. Those two alone narrow the question a long way before anyone climbs a ladder.

Decision framework from Cert-A-Roof’s repair-versus-replacement guide (4 May 2026). Factor	Lean toward repair	Lean toward replacement
Age	Under 15 years	Within 5 years of end of life
Extent of damage	Localised, under 25%	Widespread, over 30%
Number of repairs	First or second	Third or more in 5 years
Leak source	Single, identifiable	Multiple or unknown
Structure	Surface damage only	Sagging, rotted decking

Why does the material change the answer?

Because the same number of years means something different on each one.

Cert-A-Roof’s May 2026 lifespans put asphalt shingle at 20 to 30 years and clay or concrete tile at 50 or more. Metal runs 40 to 70 years, and flat or built-up 15 to 25. So age alone decides nothing until you know what is up there.

The worked consequence is worth stating plainly. A twenty-year-old asphalt surface is near the end of its life and major repairs may not be worth it. A twenty-year-old tile surface has decades left, which makes repair the obvious call.

Tile has a second twist. Premium Roofing’s March 2026 figures put the underlayment beneath tile at 20 to 35 years, well short of the tile itself. A tile house often needs the sheet replaced and the tile kept, which is a relay rather than a replacement. More on that in what a relay costs against a replacement.

Where does the cost rule fit?

It is a rule of thumb, and the sources disagree about where the line sits.

Cert-A-Roof’s May 2026 guide puts it at 30 per cent: if repair costs exceed 30 per cent of a full replacement, replacement is the better investment. Build-Folio’s July 2026 guide and Angi both put the same rule at 50 per cent.

So the honest version is a range rather than a number. Somewhere between 30 and 50 per cent of replacement cost, depending which source you follow, a repair stops being the economical choice.

The rule also applies cumulatively, which people miss. Three repairs at the top of Cert-A-Roof’s $300 to $1,500 band is most of a fourth one already spent, and it counts against the replacement figure even though no single invoice looked like a decision.

One more thing about the rule. It is about the whole surface, not a single invoice. A repair at the bottom of that $300 to $1,500 band, against a replacement in the $8,000 to $25,000 range, is a small fraction and obviously worth doing. The rule bites when somebody quotes near the top of the repair band to keep an old surface going.

Every figure on this page is a published typical range with its source and date. None is a Rancho Roofing price. Cert-A-Roof puts typical Southern California repairs at $300 to $1,500 and full replacement at $8,000 to $25,000 or more.

When is repair genuinely the right answer?

More often than the industry’s incentives would suggest.

Cert-A-Roof’s own worked example makes the point better than an argument would. Six cracked concrete tiles on an eighteen-year-old surface, no leaks yet. The recommendation is repair, because concrete tile lasts 50 years or more. Replacing individual tiles now prevents leaks and extends the life of the roof.

The same logic covers a handful of missing shingles after a windstorm on an eight-year-old surface, or a single failed vent collar on a roof with a decade left. Those are repairs, and quoting them as replacements is the thing that makes people distrust roofers.

If the answers come back young, localised, and one leak source, the honest recommendation is to fix it and keep your money. For independent guidance rather than a contractor’s opinion, the National Roofing Contractors Association publishes standards, and we do repair work when a repair is what the surface needs.

Can one bad leak mean a whole new roof?

It can, but only when the leak reveals something larger, such as a soft deck or a failed underlayment across an area. A single leak from an identifiable source on a sound surface is a repair. The leak is a symptom, not the diagnosis.

How many repairs are too many?

Cert-A-Roof’s threshold is three or more in five years, which it treats as a signal of systemic failure rather than isolated problems. The pattern matters as much as the count: three patches clustered in one area says something different from three spread across the house.

Is it worth repairing before selling?

Often, because a sound surface removes a negotiating point at inspection. A full replacement rarely returns its cost in the sale price if the existing surface has years left. Which applies depends on where the roof sits in its life.

Should I get a second opinion?

On a replacement recommendation, yes. Ask both contractors for the five answers above in writing, not just a price. Two quotes that disagree on the extent of damage are describing two different roofs, and one of them is wrong.