Spot repair
Fits when: One identifiable defect — a cracked length, a separated joint, a section crushed by a root or by settlement — in a line whose remaining footage looks sound.
Trade-off: By far the smallest job, and the right answer more often than the quoting habits of the trade suggest. It rests entirely on the rest of the line genuinely being in good condition, which is a judgement about material and age as much as about the footage — and it is the wrong choice if the same defect appears three times in the same run.
Cured-in-place lining
Fits when: A run with defects distributed along it, where the host pipe still holds its shape and its route, and where what is above it is expensive or disruptive to open.
Trade-off: Rehabilitates a length of pipe with far less excavation, and closes the joints roots were using. It is not universal: it needs a host pipe that has not collapsed, lost its alignment, or sagged, it slightly reduces the internal diameter, and connections along the run have to be reinstated. Where the line has failed structurally, lining is not an option that was rejected — it is one that does not apply.
Open-cut excavation and repair
Fits when: The defect is shallow, in accessible ground, or the pipe has collapsed or lost alignment badly enough that nothing can be worked through it.
Trade-off: The most direct and often the least expensive way to reach a single point, with the cost sitting in the digging and the reinstatement rather than the plumbing. What is above the line decides everything: a lawn, a mature planting bed, a driveway, and a public sidewalk are four very different reinstatement problems.
Full replacement of the run
Fits when: A pipe failing by age and material along its whole length, a run with a sag that ponds waste every time it is used, or repeated defects that make spot repairs a subscription.
Trade-off: The largest and most disruptive job, and the only one that resets the clock rather than extending it. Replacement can be done by excavation or by pipe-bursting a new line along the old route where conditions allow. It is worth arguing against when one defect explains everything found, and worth arguing for plainly when it does not.