Almost every hard point in survey work is really an offset problem in disguise. The point you actually want — the dead center of a bolt, the true corner where two faces meet — is usually a place you can't set a prism. So you measure to something near it and derive the real point from a known relationship. Get that relationship right and the point is exact. Guess at it and everything built on that point inherits the guess.
An offset is only an error if you don't know it
A prism offset isn't a mistake by itself. Surveyors work with known offsets all day — a prism constant, a fixed rod height, a documented instrument offset. The reason those don't hurt you is that they're known: a fixed number you can subtract back out cleanly. The measurement is offset, but the result isn't, because the geometry was defined instead of estimated.
The damage starts the moment the offset becomes a guess. "About three inches off the corner." "Roughly centered on the bolt." Now there's a quantity in your result that nobody wrote down and nobody can remove later. And a guessed offset doesn't just add error — it adds inconsistent error, a different amount at every point and in a different direction, which is the kind you can't correct in the office because there's no single number to back out.
Why this is the whole game on bolts and corners
The two points that cause the most rework in as-built work — anchor bolt centers and outside corners — are hard for the same reason: you can't put a prism on the exact point, so you're always working off an offset. A bolt that's leaning splits the "top" from the true center. A corner is a theoretical line with no surface at all. In both cases, whether your data is any good comes down to a single question: was the offset known, or was it guessed?
Making the offset known instead of guessed
That's the shared idea behind both tools we build. The SDCJ Anchor Bolt Locating Jig drives the offset on a bolt to a definitive datum — it seats over the bolt and gives you true dead center regardless of lean, so the point is captured, not estimated. The Precision Corner Offset (PCO-1) does the same thing for a corner, holding your prism at a fixed, selectable 1", 2", or 3" offset so the true corner backs out by geometry. Different points, same principle: replace the number you were guessing with one you know.
The takeaway from the dirt
You will never get to put a prism exactly on the point that matters. That's fine — surveying has always been offset work. The line between a clean as-built and a soft one isn't whether you used an offset. It's whether that offset was a known quantity you could remove, or a guess you quietly baked into the result.
The SDCJ-1 and the Precision Corner Offset are both built to make the offset a known number instead of a field guess. Engineered and built in the USA from 6061-T6 aluminum. See the specs →
See the full Shields survey tools lineup.