The piece shows up with no history attached.
Maybe it's been in one family for sixty years and nobody remembers who built it. Maybe the client found it at an estate sale and liked the proportions. There's a repair somewhere, probably done in 1978 with contact cement and misplaced confidence, and now it's your problem.
Here's the common mistake: people treat it like a new build.
They get in there and want to strip it back, clean it up, make it look like something that just came out of their shop. They sand through the original surface, swap the hardware for something modern, fill the dents. The piece that comes out the other end is worse than what arrived. What they traded was irreplaceable.
The patina on a piece of cherry that's been aging seventy years — you cannot reproduce that. You can sand it off in twenty minutes. Hand-cut dovetails on the interior, still tight, a little irregular, a hundred years old: worth keeping. The layout lines on the inside of a case. The slight unevenness of a hand-planed surface. These aren't defects. They're the record of the person who built it.
Your obligation is to the piece, not to your preferences.
That said, you still have a job.
Structural failure gets repaired. A joint that's failing needs to be reglued or replaced. A leg that's split and could fail under load — fix it. A drawer jammed solid — fix it. Safety and function come first. Everything else is a matching problem, not an improvement problem.
If you're repairing a surface, the goal is to match the existing sheen and color, not to apply the "correct" product. A perfect patch with the wrong sheen level is visible from across the room. Match what's there. On hardware: if you need to replace a broken knob on a Victorian sideboard, match the era. Brushed nickel from the hardware store is not the right answer.
Do the assessment before you touch anything — and look in both directions.
Most repair diagnostics focus on finding what's wrong. On an inherited piece, you also need to find what's right. Pull a drawer slide and look inside. You might find a hand-cut dovetail that's still perfectly fitted after sixty years. Old-growth Douglas Fir that hasn't been commercially available in decades. Proportions that actually stop you.
Write it down. The documentation you deliver at the end should include what you found: species if identifiable, construction method, hand versus machine work, estimated age, what made it worth keeping. The client who brought in what they thought was an old table leaves knowing it's genuinely well made.
That changes how they treat it.
One more thing on previous repairs: don't reflexively undo them. Evaluate each one on its own terms. Is it structurally sound? Is it causing a problem, or just aesthetically inconsistent? A sound prior repair that's slightly visible is better than an unnecessary intervention that damages what surrounds it. If you do have to take one out, document it first — photograph it, note the material.
The piece was made by someone who put time into it.
Your decisions will be visible to whoever works on it next. Make them decisions you'd be willing to explain.