N NoBullshit.wiki 中文

Fast food field supervisor - daypart-organized store visit checklist: open verifies records against reality, peak watches throughput and positioning, close checks execution - and every item is three-state with "no record" counting as a fail

Works with:Claude / GPT-4 classCategories:Fast Food RestaurantLevel:Intermediate Updated

You are a field supervisor in a fast food chain,
designing the store visit checklist. Produce a
single-store visit checklist organized by daypart,
for me to carry and execute on site.

Three dayparts; every item has three columns: what
to check / pass criteria (an observable behavior or
number) / evidence (what physical object or record
to look at).

Daypart one: arrive before open (about 30 minutes
early)
Focus - verify the records are true, not that they
exist:
- Temperature log: yesterday's cooler and freezer
  readings by interval, present, with no
  backfilling traces (several days in one
  handwriting at once = backfilled) - pass:
  interval-by-interval entries, no backfill
- Date labels: sample [5] opened/prepped items;
  label date consistent with the item's actual
  condition - pass: label present, date logic
  sound (open date + shelf life coherent), no
  signs of spoilage
- Thaw log: yesterday's thaw batches registered,
  thawed volume plausibly matching the day's usage
- Staffing: roster vs. actually present, and a
  coverage plan for any gap

Daypart two: stay through the lunch peak
(11:30-13:30, on the floor the whole time)
Focus - watch the floor, not the binder:
- Throughput: time [5] orders from ticket to hand-
  off; pass is this store's standard [X minutes];
  late orders get annotated with the stuck stage
  (prep / cooking / packing / pickup call)
- Positioning: every station staffed at peak, and
  where the manager stands (command position, not
  covering the register - covering means the
  schedule has a hole)
- Incident handling: when a sell-out, refund, or
  equipment fault happens, who takes it and how
  long to recover - record elapsed time
- Food-safety red-line behaviors: glove discipline,
  cross-contamination moves - record the behavior,
  do not erupt on the spot

Daypart three: the close
- Closing execution: equipment cleaning, leftover
  labeling and sealing, waste out - against the
  store's closing SOP key items
- Cash-up: over/short status, system day-close
  done

Verdict rules (uniform across the checklist):
- Three states per item: pass / fail / no record -
  "no record" always counts as a fail, noted as
  "no record" rather than "fail": the two mean
  different things in the debrief - the process
  never ran vs. it ran and missed
- Food-safety items are marked "red-line": one
  fail fails the whole visit, with no averaging
  against other items

Fixed ending:
- Findings table: daypart / item / verdict /
  evidence note - only fails and no-records; passing
  items do not enter the table
- The no-touching rule: in-store, the supervisor
  does not scold staff directly, does not remediate
  on the spot, does not bypass the manager - every
  issue waits for the formal debrief with the
  manager (a separate conversation)

Constraints:
- Cap the checklist at [30] items - beyond that
  the supervisor is doing the manager's job, and a
  visit becomes a residency
- No "store atmosphere" or "team spirit" items
  with no evidence basis
- Cite no regulation clause numbers - food-safety
  items reference this store's SOP and brand
  standards; statutory text belongs to compliance
  audits

Why it is written this way

Each daypart carries one focus, aimed at the most common waste in store visits: most supervisors arrive at random hours and avoid the peak entirely (fearing they'd be in the way), so they audit binders during the trough and miss the ninety minutes where store management actually shows itself. "Stay the whole peak" is written as a hard requirement, and the throughput sampling (five timed orders, stuck-stage annotated) is what keeps those ninety minutes occupied - otherwise standing in a rush looks like watching traffic.

"No record counts as a fail" is the heaviest rule here, and it targets a real, universal behavior: the cheapest way for a store to pass an inspection is not doing the work but completing the paperwork afterwards. A temperature log filled in one sitting with one pen, date labels present with casually filled-in dates - backfilled records are more dangerous than no records, because paper compliance buries physical risk. Splitting "the process never ran" from "it ran and missed" also feeds the remediation direction: the first is a discipline problem, the second a capability or resource problem.

"No-touching" is written as discipline, not courtesy, because a supervisor scolding staff on the spot destroys two management lines at once: the manager's authority (their team just got publicly overruled) and the staff's willingness to be seen (who shows a supervisor a problem next time?). The supervisor's value is finding issues and making the manager own the fix - and this discipline pairs with the formal debrief scripts that take over once the checklist leaves the store.

Variables

Variable Notes
Throughput standard The brand's passing minutes - varies hugely by menu; use the operations manual number, or establish a baseline from this store's median first
Red-line scope Whether anything beyond food safety (fire, cash) joins the zero-tolerance list - per brand policy; keep it short, too many red lines means none
Sample sizes Days of temperature logs, count of date labels, number of timed orders - scale with store size, but keep sampling below counting: that is what makes it sampling

store visitsupervisioninspection