Food Temperature Logs: What to Record and How Often

A food temperature log records the actual temperatures of stored, cooked, held, cooled, and delivered food, with a time and a name attached to every reading. At minimum, log every fridge and freezer twice daily (fridges at or below 5°C / 41°F, freezers at −18°C / 0°F), core cooking temperatures for high-risk items, hot-holding above 60°C (140°F) every couple of hours, and cooling progress for anything cooked and chilled. The log is only as good as what happens when a reading is wrong — every out-of-range entry needs a recorded corrective action.

Why the log exists at all

Temperature is the main lever a kitchen has against bacterial growth. Between 5°C and 60°C (41–140°F) — the danger zone — bacteria multiply rapidly; below and above it, they slow or die. Every part of a temperature log maps to keeping food out of that window: cold storage keeps it below, cooking pushes it through quickly to a safe core temperature, hot-holding keeps it above, and cooling gets it back below fast.

There's a second reason, just as practical: proof. When an inspector, auditor, or brand-standards reviewer asks how you know the walk-in was cold last Tuesday, the log is the only answer that counts. It is also the first document examined after any suspected food-poisoning complaint. A kitchen with honest, complete logs is in a completely different conversation than a kitchen without them.

Temperature monitoring is also where most HACCP plans put their critical control points — the log is the monitoring record Principle 4 requires. If you're building the surrounding system, the plain-language HACCP guide covers how logs fit into the full plan.

What should you record, and how often?

Five categories cover a typical kitchen. The table shows the common practice for each; your own HACCP plan or local rules may set stricter limits, and those win.

A few notes on the ones teams get wrong:

• Cooking temperatures are per-item, not per-day. The reading is the core temperature of the thickest part of the actual batch, taken with a clean, calibrated probe — not a generic "cooked properly ✓".
• Cooling is a time series, not a single reading. One number can't show that a stockpot passed through the danger zone quickly. Log the start temperature and time, then checkpoints until it's below 5°C.
• Deliveries are a rejection decision. The point of probing an incoming chilled delivery is that you can still refuse it. Log the reading before signing, and record any rejection. The wider system for goods that live and die by temperature is covered in cold chain temperature monitoring.

Where the first readings of the day belong

The opening fridge check is the single most important entry, because it's the only one that can reveal an overnight failure while the food is still salvageable — or provably not. If a walk-in reads 9°C at 7 a.m., you need to know how long it's been warm, which is exactly what last night's closing reading tells you. Pairing an opening and a closing reading brackets the overnight window.

For that reason, temperature checks should be the first block of the kitchen opening checklist, before prep begins — a bad reading changes the whole morning's plan.

What to do when a reading is out of range

An out-of-range reading without a recorded response is worse than no log at all — it documents that you knew and did nothing. Every kitchen needs a short, pre-agreed playbook:

1. Recheck with a second thermometer to rule out instrument error.
2. Assess the food. How long has it plausibly been out of range? A fridge at 7°C found an hour after a busy delivery is a different situation from one that's been warm all night.
3. Act on the food. Move it to working refrigeration, use it immediately under the relevant time rule, or discard it. When in doubt, throw it out — the cost of stock is always lower than the cost of a sick guest.
4. Act on the equipment. Log a maintenance issue; don't return food to a unit that hasn't been fixed.
5. Record all of it — the reading, the decision, the action, who made it, when.

This corrective loop is the part inspectors read most carefully. A log with the occasional 6.5°C reading followed by "moved stock to walk-in 2, engineer called, rechecked 4°C at 14:00" reads as a kitchen in control. A log with nothing but 3.0°C for 365 consecutive days reads as fiction.

Calibration: the log is only as good as the probe

A probe thermometer that reads two degrees low makes every entry in the log meaningless. Calibrate weekly with the ice-point method: a glass of crushed ice topped with water should read 0°C (32°F) within your tolerance, typically ±1°C. Log the calibration check itself — date, thermometer ID, reading, initials — and take any failing probe out of service immediately. Keep at least one spare.

Also separate probes by use, or sanitise between uses: a probe that goes from raw chicken to a rested roast is a cross-contamination vector with a handle.

Paper logs vs digital logs

Paper temperature sheets fail in three specific ways: they get filled in retrospectively in one handwriting at the end of the week; out-of-range readings get "rounded" to passing because a failure means paperwork; and nobody at head office sees any of it until an audit. None of this is a character flaw — it's what happens when recording is tedious and failure is punished with forms.

Digital logs change the incentives. Each reading is timestamped at entry, so retro-filling is impossible. Limits are built in, so a 7°C entry automatically flags rather than relying on someone to notice. And the corrective action can be attached on the spot — a photo of the moved stock, a note, an assigned follow-up. See how digital checklists handle numeric readings on the product overview.

Task10x supports this directly: temperature checks are numeric checklist items with min/max limits, out-of-range readings can auto-create corrective actions assigned to a named person and tracked to closure, and every entry lands in a timestamped, attributed audit trail that exports to CSV when the inspector wants history.

The bottom line

Log five things — storage, cooking, holding, cooling, deliveries — with real readings, real times, and real names. Check fridges at least twice daily, treat every out-of-range reading as an event with a recorded response, and calibrate the probes that produce the numbers. The goal isn't a perfect-looking log; it's a truthful one that shows food stayed out of the 5–60°C danger zone, and what you did on the days it didn't.