Repair vs. replace windows: which is right for your situation?

The short answer

For most homeowners with one or two specific issues on otherwise-functional windows, repair is the right call. A failed seal on an insulated glass unit, a broken sash cord, weatherstripping that needs replacing, a cracked pane on a single window — these are $75–$650 fixes that buy you another 5–15 years of useful life from a window that's otherwise doing its job. Don't replace a $600 window because $250 of it has failed.

Replacement is the right call when the failure is structural or systemic: rotted frames (the wood has lost integrity, repair is cosmetic only), warped vinyl that no longer seals against the frame, multiple windows showing the same failure mode at once (which usually means the whole installation is at end-of-life), single-pane windows in a climate where heating/cooling cost is a meaningful budget line, or any window pre-1980 that's still original. Whole-home replacement runs $8,000–$15,000 for an 8-window project, with per-window economics improving as project size grows. The full cost picture is in the window replacement cost guide.

The genuinely ambiguous middle case is when you have 3–6 windows with moderate failures (failed seals on multiple windows, rot starting in 2–3 frames, weatherstripping gone everywhere) on a 15–25 year-old vinyl installation. The math rarely points cleanly in either direction; the decision depends on planned tenure and energy-bill trajectory. The decision framework below walks through how to resolve this case for your specific situation rather than guessing.

These cost numbers come from the same sources as the other cost guides on this site: Bureau of Labor Statistics construction-wage data, the National Association of Home Builders annual cost report, and aggregated installer pricing across 12,400+ verified projects. Repair-cost numbers also draw on per-issue glazier and handyman pricing data. The full data inventory and calculator formulas live on the methodology page; the broader replacement cost picture is in the window replacement cost guide; and the material-tier comparison (vinyl vs. fiberglass) is in the vinyl vs. fiberglass cost guide.

For a project-specific replacement number based on your state, window count, and material — to compare against the repair cost — the window replacement cost calculator produces a low–high range with line items in about 30 seconds. The rest of this guide walks through the cost reality of each path, the 5-question decision framework, and when each path is clearly right.

Cost reality of repair

Repair costs depend almost entirely on what's failing. Windows are component assemblies, and most failure modes affect one component at a time — which is why repair is usually dramatically cheaper than replacement. Per-issue cost bands for the most common failures, in 2026 dollars:

• Failed insulated glass unit (IGU) — $150–$400 per IGU. The most common failure on dual-pane windows. Glazier removes the failed unit, installs a new sealed double-pane (or triple-pane) IGU into the existing sash. Per-window cost depends on size and glass package; 30+ year-old windows may have non-standard sizing that pushes toward the high end.
• Sash replacement — $250–$650 per window. Replaces the operable glass + frame portion while keeping the exterior frame, jamb, and trim. Useful when the IGU has failed AND the sash itself shows wear (warped, broken hardware, weatherstripping degraded). Cheaper than full replacement; preserves all the trim work that adds visible cost to a full replacement.
• Weatherstripping replacement — $75–$200 per window. The cheapest meaningful repair. Improves draft, often improves measured U-factor by 5–15% on an otherwise-functional window. Worth doing across the whole house every 8–12 years even if you're not chasing a specific failure.
• Hardware replacement (sash cords, locks, balances, cranks) — $100–$300 per window. Old double-hung windows with broken sash cords or balances usually need this; casement windows with failed crank mechanisms need crank-assembly replacement. Specialized hardware may push higher; common hardware is at the low end.
• Cracked or broken pane (single-pane or one pane of a dual-pane) — $100–$350 per pane. Single-pane is cheap and quick (~$100–$150). Dual-pane requires IGU replacement (see above) since the broken pane compromises the sealed unit.
• Frame caulking and exterior re-sealing — $50–$150 per window. Not a structural repair; addresses air infiltration around the frame perimeter. Often combined with weatherstripping for a "tighten the existing windows" pass. Doesn't address IGU or hardware failures.
• Spot frame repair (small rot patch, minor damage) — $200–$500 per window. The judgment call: small rot patches can be cleaned out and filled with epoxy + repaint; substantial rot means the frame is structurally compromised and the window needs replacement. A glazier or window contractor's site visit will tell you which case you have.

For most single-issue repairs, expect $75–$650 per window. Multiple issues on the same window can stack — a window that needs a new IGU + new weatherstripping + new sash hardware can hit $500–$900 in repair, at which point you're approaching the cost of a single-window replacement and the math starts to argue for replacement instead. The decision framework below addresses this stacking case explicitly.

Cost reality of replacement

Replacement costs are documented in detail in the window replacement cost guide; the summary for the repair-vs-replace decision:

• Per-window installed: $400–$1,200 depending on material (vinyl baseline; fiberglass +25–60%; wood +50–100%) and installation type (insert vs. full-frame; full-frame is ~30% more labor).
• Single-window replacement (1 window): $600–$1,200 installed. Per-window setup costs (crew show-up minimum, permit, disposal) make this the worst per-window economics. Almost always the wrong call when a $250 repair would resolve the issue.
• Small-project replacement (2–4 windows): $1,200–$4,000. Better economics than single-window but still inefficient. Repair-vs-replace math gets close; usually still favors repair unless multiple windows have stacking failures.
• Whole-home replacement (8+ windows): $8,000–$15,000 for an 8-window project; $13,000–$25,000+ for 14-window or premium-tier projects. Per-window economics improve substantially at this scope; the cost difference between repairing 8+ windows with stacking failures and replacing them narrows to where the longer-term economics (energy savings + lifespan + reduced future maintenance) start to favor replacement.

The replacement cost crossover with repair is not a single number; it's a function of scope (how many windows) and failure depth (how many components per window). The decision framework below resolves the crossover for your specific situation. For material-tier-specific replacement comparisons (vinyl vs. fiberglass), see the vinyl vs. fiberglass cost guide.

The decision framework: 5 questions to answer

Most online "repair vs. replace" content gives generic advice and stops short of an actionable decision tool. This framework is the operator-grade version: 5 specific questions you answer about your situation, with the decision logic that follows from the answers. Run it for your project before you book any contractor visits.

1. How old is the window?

   • Under 10 years: The window is well within its functional life. Repair the specific issue and move on. Replacement at this age is almost always the wrong call (unless the original installation was defective, in which case warranty is the right path, not paid replacement).
   • 10–25 years: The middle zone. Repair is usually correct for isolated issues; replacement starts to make sense if multiple windows are failing at once or the whole installation was a low-grade tier originally.
   • 25–40 years (post-1980): The window is approaching but not yet past its expected lifespan for most materials. Replacement starts to make sense if multiple failures are stacking; isolated issues are still repair-correct. Begin treating future failures as data points about the installation's overall trajectory.
   • Pre-1980 (40+ years): The window is at or past expected lifespan for most materials AND likely single-pane or early-multi-pane technology. Repair is a stopgap, not a strategy. Plan for replacement within the next 2–5 years even if you repair the immediate issue now.

2. What's actually failing?

   • Single-component failure (broken sash cord, failed seal on one IGU, cracked pane, broken hardware, weatherstripping gone): repair. Almost always cheaper and the rest of the window is fine.
   • Frame integrity failure (rotted wood, warped vinyl that won't reseal, separated welds): replace. Frame is structural; once it goes, repair is cosmetic.
   • Performance failure (drafty even with intact weatherstripping, condensation between panes on multiple windows, single-pane windows in a multi-pane era): replacement is usually correct because the window is doing what it was designed to do — that design is just inadequate for current standards.

3. How many windows are currently failing? Count windows with active issues, not the whole installation — unless multiple windows show the same failure mode at once, which is the 7+ case.

   • 1–2 windows with localized issues: repair. Per-window setup costs make replacement uneconomical for small-scope failures.
   • 3–6 windows with moderate issues: the genuinely ambiguous case (covered below). Run the rest of the framework before deciding.
   • 7+ windows showing the same failure mode: the installation itself is failing. Replacement is almost always correct because the cost difference between repairing 7+ windows and replacing them narrows substantially.

4. What's your current glazing? The gap between your current glass technology and modern glass technology is one of the largest hidden drivers of the ambiguous-case decision — and it's the variable that anchors the energy-bill-trajectory logic in Q5.

   • Single-pane glass (typical of pre-1980 windows, some early-1980s low-grade installations, original windows in older historic homes): replacement is usually correct on its own merits. Single-pane in a 2026 climate is a continuous energy leak; the U-factor gap to modern double-pane low-E is roughly 4–5× worse, and to triple-pane in cold climates is 6–7× worse. The energy savings alone usually pay back within typical homeowner-tenure timelines.
   • Dual-pane non-low-E (typical of 1980–2000 installations, some low-grade installations into the early 2000s): the middle zone. Replacement makes sense if multiple windows are failing AND your climate makes thermal performance a meaningful budget line. Otherwise, repair-and-defer is reasonable.
   • Dual-pane low-E or triple-pane (typical of post-2000 installations and most 2010+ replacements): the glass technology is current. The decision is driven entirely by the failure-mode and project-scope answers (Q2 + Q3); glass tech is not pulling the decision either direction.

5. How long do you plan to stay in the home, and what's your energy-bill trajectory?

   • Planning to sell within 3–5 years: repair is almost always correct. Replacement's payback timelines (energy savings + lifespan advantage) require longer tenure to make economic sense. Don't pre-fund the next owner's energy savings.
   • Planning to stay 10+ years AND your heating/cooling bills are climbing or already disproportionate to home size: replacement starts to win even on isolated failures, because the energy-bill trajectory tells you the windows are leaking energy at a rate that compounds with each passing year. This is especially true if Q4 surfaced single-pane or non-low-E dual-pane glass.
   • Planning to stay 5–10 years with stable energy bills: repair the specific failure now; reassess in 3–5 years when more failures start surfacing and you have better data on the installation's overall trajectory.

How to use the framework's outputs: if 4 or 5 of the questions point toward repair, repair is clearly correct. If 4 or 5 point toward replacement, replacement is clearly correct. If the questions split 2-3 or 3-2, you're in the genuinely ambiguous middle case — read the "When repair wins" + "When replacement wins" + "Genuinely ambiguous case" sections below for scenario-specific guidance.

When repair wins (concrete scenarios)

Three specific scenarios where the framework points cleanly to repair, with the worked dollar math:

• Scenario A: 2018 vinyl windows, 2 with failed seals, owner staying long-term. Q1 = under 10 years (repair) · Q2 = single-component failure (repair) · Q3 = 1-2 windows (repair) · Q4 = dual-pane low-E (no decision pull) · Q5 = stable energy bills (repair). All 5 questions point to repair. Cost: 2 × $250 IGU replacement = $500. Replacement equivalent would be 2 windows × $800 installed = $1,600 — and would discard 18+ years of remaining frame life. Repair is correct.
• Scenario B: 1995 vinyl windows, 1 with broken sash hardware, owner planning to sell in 4 years. Q1 = 25–40 years (post-1980, ambiguous) · Q2 = single-component failure (repair) · Q3 = 1-2 windows (repair) · Q4 = dual-pane non-low-E (slight replace pull) · Q5 = selling in 3–5 years (repair). 4 of 5 point to repair. Cost: $200 hardware replacement. Replacement makes no economic sense for a single window with shorter tenure ahead. Repair is correct; future owner can decide whether to replace based on their own framework run.
• Scenario C: 2010 fiberglass windows, weatherstripping degraded across the whole house. Q1 = 10–25 years (slight repair pull) · Q2 = single-component failure across multiple windows, but cheap component (repair) · Q3 = 7+ same failure mode (slight replace pull, but the failure is weatherstripping, not structural) · Q4 = dual-pane low-E (no pull) · Q5 = any tenure (repair). Net: repair is clearly correct because the failure is the cheapest component and the windows themselves are otherwise fine. Cost: 14 × $100 weatherstripping = $1,400. Replacement equivalent would be ~$15,000. This is the case where the 7+ same-failure-mode rule has a legitimate exception: the failure mode matters, not just the count.

When replacement wins (concrete scenarios)

Three specific scenarios where the framework points cleanly to replacement:

• Scenario D: 1972 wood windows, single-pane, 9 windows with multiple failures each (rot in 4, hardware broken in 5, 3 with broken sash cords), owner staying 15+ years. Q1 = pre-1980 (replace) · Q2 = frame integrity failure (replace) · Q3 = 7+ same scope (replace) · Q4 = single-pane (replace) · Q5 = 10+ years tenure with almost-certainly-rising bills (replace). All 5 questions point to replacement. Cost: 9-window full-frame wood-or-fiberglass replacement, ~$11,000–$18,000 installed depending on material tier and state. Repair would be a $4,000+ stopgap on windows that are at end-of-life across the board. Replacement is correct; the energy savings + 30+ years of new lifespan + avoided ongoing repair spend make the math clean.
• Scenario E: 1988 aluminum windows, 8 windows with seal failures + draft + visible frame distortion in 3, owner staying long-term in cold climate. Q1 = 25–40 years (post-1980, ambiguous) · Q2 = mix of seal failure + frame integrity failure (replace) · Q3 = 7+ same/related scope (replace) · Q4 = dual-pane non-low-E with aluminum frame's poor thermal performance (replace) · Q5 = long tenure + cold climate (replace). 4 of 5 point to replacement. Cost: 8-window vinyl or fiberglass full-frame replacement, ~$8,000–$15,000 installed. Aluminum windows are the worst thermal performer in the comparison; even setting aside the seal failures, the frame thermal performance alone justifies replacement in a cold-climate long-tenure home.
• Scenario F: 2000 vinyl windows, no acute failures, but draft + climbing energy bills + owner staying 15+ years in Energy Star Northern climate zone. Q1 = 25 years exactly (slight replace pull) · Q2 = performance failure (replace) · Q3 = 7+ same failure mode (whole installation underperforming) (replace) · Q4 = dual-pane non-low-E (replace) · Q5 = 10+ years + climbing bills + cold climate (replace). 4 of 5 point to replacement. Cost: ~$13,000–$19,000 for a vinyl whole-home replacement. The subtle case: no individual window is acutely broken, but the installation as a whole is leaking energy at a rate that compounds. The energy-bill trajectory is the load-bearing signal; if bills were stable, the framework would split and the decision would shift to the ambiguous case.

The genuinely ambiguous case (and how to resolve it)

The middle case is the hardest one to write generic advice for, which is exactly why most online repair-vs-replace content stops at "it depends." Here's the operator-honest version: the case is genuinely ambiguous, AND there's a structured way to resolve it for your specific situation.

The classic ambiguous-case profile: 15–25 year-old vinyl installation, 3–6 windows with stacked moderate failures (failed seals + degraded weatherstripping + some hardware issues), owner planning 7–12 year tenure, energy bills slightly elevated but not dramatically so, dual-pane non-low-E glass. The framework typically splits 2-3 or 3-2 in this profile — neither path is clearly wrong.

How to resolve it:

1. Get quotes on both paths. One repair-contractor visit (~$0–$150 for the visit itself) producing per-window repair quotes for the failing windows, plus one replacement-contractor visit producing a whole-home replacement quote. Most installers will do both visits; some specialize in one and don't quote the other. Two real numbers beats any framework.
2. Calculate the repair-then-replace-later total. If the repair cost is $X today, what's the likely replacement cost in 5–8 years (when the next round of failures forces the decision)? Use the cost calculator with a 3% annual cost-inflation assumption. Compare $X today + future replacement cost vs. replacement cost today.
3. Estimate the energy-savings delta. For dual-pane non-low-E installations being replaced with dual-pane low-E, expect $40–$120 per year savings per home in mild climates, $80–$180 per year in cold climates. Multiply by your planned tenure to get the cumulative savings on the replacement path.
4. Add the lifespan-replacement-avoidance value. If you replace now and live in the home another 25+ years, the new windows last your tenure. If you repair now and live another 15 years, you may face a forced replacement at year 12–15. The avoidance value of the second replacement, discounted to present value, is real money.
5. If the math is still close, default to repair. Two reasons: (a) repair preserves optionality (you can always replace later; you can't un-replace), and (b) replacement quotes typically come in 10–20% above the framework's expected range once contractors find the things the calculator doesn't see (hidden framing repair, oversized custom shapes, lead-paint surcharge for pre-1978 homes). Repair has fewer hidden costs.

The ambiguous-case resolution is not a formula — it's a structured way to gather the inputs and then make a judgment call. The framework converts a "I don't know" situation into a "here's what I'd need to know to decide" situation, which is the best any decision tool can honestly do for genuinely ambiguous cases.

Frequently asked questions

Is window repair always cheaper than replacement?

Per-window, almost always yes — a $250 IGU reseal is dramatically cheaper than an $800 replacement. The exception is when you're repairing 7+ windows showing the same failure mode at once. At that scope, per-window setup costs (glazier visits, scheduling, mobilization) start to push the repair total close enough to the replacement total that the longer-term economics flip. The decision framework's Q3 covers this threshold; below 7 windows, repair is genuinely cheaper; at 7+, run the full math.

Can I just replace the glass and keep the frame?

Yes — this is called sash replacement (replacing just the operable glass + frame portion) or IGU replacement (replacing only the insulated glass unit inside an existing sash). Sash replacement runs $250–$650 per window depending on size and material. IGU-only replacement runs $150–$400 per IGU. Both preserve the frame, exterior trim, and interior trim. The catch: this only works if the existing frame is structurally sound. For rotted wood frames, warped vinyl, or separated welds, glass-only replacement is throwing good money after bad — the new glass will outlast the failing frame.

How do I know if my window seal is actually failed?

Three reliable signs of a failed insulated glass unit (IGU) seal: (1) condensation or fogging visible between the panes that you can't wipe off (it's inside the sealed unit); (2) visible distortion of the glass surface — the panes have warped slightly because the inert-gas fill has escaped; (3) measurably worse thermal performance at the window (cold to the touch in winter, drafty even with intact weatherstripping). One sign is a strong indicator; two or three confirms it. Note that surface condensation on the room-facing side of an intact window is normal in cold climates and does NOT indicate seal failure — it indicates indoor humidity is high.

Does repairing windows reduce home value vs. replacing?

Generally no, with one caveat. In typical mid-market resale, buyers don't differentiate "recently repaired windows" from "recently replaced windows" — both read as "windows are not a problem." Functional, properly-maintained windows of any age don't hurt resale. The caveat: if the home is being marketed as a renovation-finished property in a market where buyers expect modern construction (high-end suburban, new-build comp markets), original or visibly-aged windows can get flagged in inspection comments and create negotiating leverage for the buyer. In that specific situation, replacement reads better than repair regardless of functional status. For typical resale, the energy-bill trajectory matters more than window age — buyers will notice high heating costs in a comp comparison.

When should I get a professional inspection vs. just deciding myself?

If the framework above produces a clear answer (4 or 5 questions pointing the same direction), you can skip the inspection and book either the repair contractor or the replacement quote directly. If the framework split 2-3 or 3-2 — the genuinely ambiguous case — paying $150–$300 for a window-specific home inspection is the cheapest tiebreaker available. The inspector will quantify frame integrity (rot vs. cosmetic damage), measure actual thermal performance, and identify failure modes you may have miscategorized. Use the inspection report as the input to a final framework run, not as the decision itself.