When you’re called to a multi-tenant site, what’s your baseline for an audit — just cylinders, hinges, and closers, or do you fold in key control and after-hours access tests? I’m standardizing a 90-minute walkthrough for a 12-door storefront row and want to compare notes on what’s worth measuring to drive a tailored upgrade plan (rekey vs SFIC swap, audit-trail readers on rear exits).
On a ‘90-minute’ pass for a 12-door row, I always photo the hinge and door-edge labels plus strikes to confirm fire ratings and latch throw, because that can flip the plan between a quick rekey and SFIC/reader upgrades and keeps you from quoting non‑compliant hold‑opens. Would you bake in that extra 5 minutes up front?
Quick win for us: we turned on-deck stability into a passive task — pooled serum QC for the LC–MS steroid panel sat at 4°C on the autosampler with injections at 0/8/24/48 h during routine batches, plus three freeze–thaw aliquots the next morning — so we had plots by lunch and shaved about 10 hours off that “60–80 hours” bucket. Small caveat: log autosampler temperature with a cheap USB probe or the reviewer will question the claim.
I always build in a key-control snapshot, @OP: carry a couple blanks (Schlage C and Best A) and do quick insert-only tests at each cylinder to map keyways and flag uncontrolled duplication; that alone usually decides rekey vs SFIC consolidation. Small caveat: if tenants are open I skip any cross-try and just note storefront hardware (Adams Rite deadlatch vs hook) so I can price the right cams if we go SFIC. Okay to get PM approval for a single after-hours cross-try on shared entries?
Building on @mShiro85 , turn those messy clusters into two lean IA options and run a head‑to‑head Treejack on the clinic’s top tasks; winner keeps its labels, loser donates any clear paths.
I’d pull language from front‑desk scripts and insurance forms to rewrite labels, then run a tiny moderated Treejack on 7–8 tasks like “book an appointment” and “find urgent care hours.” Small caveat: cap it to two nav variants and listen for the why, not just success % — if three people miss the door, rename the sign. Do you have call logs or email FAQs you can mine?