Wood Filler Vs Epoxy Vs Dutchman Vs Replace: Rot Repair By Extent Decision Matrix

Why this matters

Wood rot repair is a sizing decision before it is a materials decision. The same rotted sill, jamb, column base, or trim board can be correctly fixed four different ways, and the line between them is the extent of sound wood left and where the load goes. Use cosmetic filler on a structural post base and it fails the first season. Tear out and replace a board that had two repairable inches of punky face and you doubled the labor for nothing. Match the method to the extent and the structural role, and the repair lasts as long as the surrounding wood.

This matrix sequences the four options from least invasive to most, with the conditions where each one wins.

The options

• Wood filler (cosmetic): A lightweight or hardening filler for shallow surface defects: nail holes, dents, small voids, hairline checks in non-structural, painted trim. It bonds to sound wood and is sanded flush. It is not a structural material and does not tolerate movement or moisture cycling well.
• Penetrating epoxy + epoxy filler (consolidate and rebuild): A two-part system. A thin penetrating consolidant soaks into and re-hardens punky-but-present fibers; a thicker structural epoxy paste rebuilds missing sections. Used where there is rot but the member's core and footprint are still mostly there and you want to keep the original wood. Strong, waterproof, machinable, paintable.
• Dutchman (patch a sound member): Cut out the rotted section cleanly and let in a matching new piece of wood (the "dutchman"), glued and/or fastened. Used when rot is localized to one area of an otherwise sound, often visible or profiled member you want to preserve (a column, a profiled casing, a stair part).
• Full replacement: Remove the member and install new. Used when rot is extensive, the member is structural and compromised, the cost to repair approaches replacement, or the wood is so far gone that consolidation has nothing to grip.

When wood filler wins

Filler is correct only when ALL of these hold:

• The defect is shallow (surface only) and small in area.
• The wood around it is fully sound and dry, with no soft, punky fibers.
• The member is non-structural (decorative trim, a painted surface) and does not flex.
• There is no active moisture source, and the repair will be sealed and painted.

If you can push a probe (an awl or screwdriver) into the wood with light pressure, it is past filler. Filler bridges a hole; it does not restore strength or stop rot.

When epoxy wins

Penetrating epoxy plus filler is correct when:

• There is real rot (soft, punky fibers) but the member's overall shape and a sound core remain, and you want to keep the original wood (historic trim, a profiled element, an in-place sill or jamb that is hard to remove).
• The repair area can be more than filler-deep but the member is not so far gone that it crumbles.
• The location sees weather/moisture; epoxy is waterproof and moves reasonably with the wood.
• You can fully dry the wood first; epoxy will not bond to or harden saturated fibers.

Epoxy shines on window sills, jamb bottoms, column bases, and exterior trim where removal is disruptive. Drill relief/keying holes into sound wood so the consolidant penetrates and the fill keys in.

When a dutchman wins

A dutchman is correct when:

• Rot is localized to a defined area of an otherwise sound member.
• The member is visible or profiled and you want a wood-to-wood repair that takes stain or paint identically and reads as original.
• The surrounding wood is strong enough to host a let-in patch and carry any load.
• You can make a clean, square cut to good wood and mill or source matching stock.

A dutchman is often the right call where epoxy would be a large, visible mass on a stained surface, or on a part that must keep a precise profile (a turned column, a stair tread nosing, a crown return).

When full replacement wins

Replace when:

• Rot is extensive (more than roughly a third of the cross-section, or running the length of the member).
• The member is structural and its strength is compromised (a load-bearing post, a beam, a rim that carries a deck).
• Probing finds soft wood deep into the core with no sound material to consolidate or host a patch.
• The labor to consolidate or dutchman approaches the labor to replace, or matching/repairing would be unreliable.
• There is an active moisture/structural cause that demands the member come out to address (failed flashing behind it, hidden rot extending into adjacent framing).

For any structural member, default toward replacement unless a structural professional signs off on a repair, and always find and fix the water that caused the rot.

Field decision flow

1. Probe the full extent. Push an awl in at multiple points. Map sound vs. soft and how deep. This single test drives the choice.
2. Identify the role. Structural or cosmetic? Load-bearing tips you toward replacement or an engineered repair.
3. Find the water. No repair survives if the source is live. Fix flashing/drainage first.
4. Apply the matrix. Shallow + sound + non-structural = filler. Real rot + sound core + keep-the-wood = epoxy. Localized + sound host + visible profile = dutchman. Extensive or structural or fully punky = replace.
5. Dry before you bond. Filler and epoxy both need dry wood. Replacement lets you start fresh.
6. Seal and finish all faces, including end grain and back, so the repair does not become the next rot site.

References

• ASTM D2017 / D2899 (wood decay evaluation) and Forest Products Laboratory, "Wood Handbook" (FPL-GTR-282), on decay, moisture, and repair.
• International Residential Code (IRC) Chapter R3/R5 (R317 decay and termite protection, structural members) for when a member must be replaced rather than repaired.
• Epoxy consolidant manufacturer technical data sheets for moisture limits, penetration, and cure; preservation guidance (e.g., National Park Service Preservation Briefs on epoxy and dutchman repairs).