What resolution satellite imagery do you need to verify a building footprint?
A sponsor sends over a draw package with photos of a slab, a framed first floor, maybe a crane. The photos look right. But photos are whatever the sponsor chose to point a camera at, and a draw schedule tied to "60% complete" deserves more than a chosen angle. The question that actually matters before you sign off: can you see the footprint from above, and does it match what the pro forma says is there?
That's a resolution question, and it has a specific answer.
Ground sample distance, in terms that matter for a footprint
Ground sample distance (GSD) is the size of ground that one pixel covers. A 1m GSD image means each pixel represents a 1m by 1m patch of earth, so a building edge gets smeared across a pixel or two and the shape goes soft at the corners. A 0.5m GSD image cuts that patch in half, and a footprint that was a blur at 1m resolves into a clean rectangle with square corners you can measure against a site plan.
For footprint verification specifically, that 0.5m line isn't arbitrary. Confirming a footprint means confirming three things: the structure exists, its outer dimensions match the stated square footage or acreage, and its position sits where the site plan says it should. All three of those are edge-detection problems. You're tracing a perimeter, not reading a sign or counting cars in a lot. Coarser imagery, the 1 to 3m range you get from free public archives, gives you a general shape and a rough location, which is fine for confirming "yes, something got built here" but not for catching a footprint that's 15% smaller than the permit drawing, or a wing that got value-engineered out after the appraisal.
Anything finer than 0.5m, down toward 0.1-0.15m aerial, is overkill for this specific question. That tier is built for reading rooftop equipment, counting units, or spotting material stockpiles, not for tracing a perimeter you already have dimensioned on a survey. For a footprint check against a stated acreage, 0.5m VHR is the practical floor and roughly where the cost curve stops paying for itself.
Why the date matters as much as the pixel size
Resolution tells you whether you can see the edge of the structure. It says nothing about when the image was taken relative to the sponsor's claimed completion date. A 0.3m image from eighteen months ago proves nothing about where the project stands today, and an archive provider's next pass over a given parcel lands whenever the satellite happens to be overhead, which may or may not line up with a draw deadline.
Fine resolution paired with the wrong date is the gap that trips up a lot of verification requests built on stock archive imagery. A draw review needs imagery tasked for the specific day the question gets asked, so the "after" half of the pair shows the asset as it stands this week. Pair that with a baseline taken before ground was broken, and the dimensions, the position, and the construction stage all line up against the dates the deal team set for this draw.
What this looks like for a draw request
Put the two together and the spec is simple: VHR imagery at 0.5m GSD or finer, tasked on demand rather than pulled from whatever's sitting in an archive, delivered as a before/after RGB pair against the parcel's stated footprint. That combination answers the question a draw committee is asking, which isn't "does this look like progress" but "does the footprint on the ground match the footprint in the file, as of this week."
Site Verification builds exactly that pair on request, a before/after image set and a short written confirmation that the footprint, acreage, and construction stage line up with what the sponsor reported, at 0.5m resolution, for the specific dates a draw decision depends on.
If the next capital call is sitting in your inbox, order one before you sign it.