Temperature control is the backbone of most HACCP plans, and it is also where they break
If you look across the CCPs in a typical food business, cooking, chilling, hot holding, cold storage, reheating, and freezing, most of them come down to temperature. Time is usually the inseparable partner. The science is well established. The regulatory expectations across the EU and UK are clear in principle: food businesses must keep food at temperatures that do not present a risk to health, and they must be able to demonstrate this. And yet, temperature control remains one of the most common sources of non-conformances at inspection because the gap between what the plan says and what happens on the floor is wide enough to walk through.
This article is for food business operators, kitchen managers, and technical leads who want temperature control that works in practice. It covers how to set critical limits that are scientifically sound and operationally achievable, how to monitor them without creating paperwork that people fake, and how to handle deviations in a way that protects the customer and satisfies an auditor.
What a critical limit actually means for temperature
A critical limit is the value that separates acceptable from unacceptable at a CCP. If the limit is breached, the product is potentially unsafe. It is not a target. It is not a guideline. It is a boundary, and crossing it triggers a corrective action that must involve a decision about the affected batch.
For temperature, critical limits are nearly always expressed as a minimum core temperature for cooking and reheating, a minimum holding temperature for hot food, a maximum storage temperature for chilled and frozen food, and a maximum time at a given temperature for cooling. The numerical values are drawn from legislation, national codes of practice, industry guidance, or validated scientific data specific to the product and process.
The operational mistake many businesses make is to set a critical limit that matches a guidance value but leaves no room for the real-world variability of their equipment. A combi oven with a 3°C temperature swing around the set point, a blast chiller that loses performance as the condenser fouls, or a cold room that spikes every time the door opens during a busy dispatch period can push a process into repeated borderline failures. If your operating point sits directly on the critical limit, you will either start ignoring deviations or reject product unnecessarily.
The smarter approach is to set a target that is more conservative than the critical limit and to define the critical limit as the true safety boundary. For cooking poultry, the critical limit might be 75°C core temperature, but the target is 78°C to absorb probe placement variation and oven hot spots. The corrective action only kicks in at the critical limit, not the target. This distinction keeps the system honest.
Where the numbers come from
The temperature values in your HACCP plan should not be guesses. They need a stated source. In the EU and UK, the primary references include:
- Regulation (EC) No 852/2004 on the hygiene of foodstuffs, which requires that food is not kept at temperatures that might cause a risk to health and that the cold chain is maintained, but it does not give you a single table of critical limits.
- Regulation (EC) No 853/2004, which contains specific temperature requirements for products of animal origin, including maximum temperatures during transport, storage, and processing of meat, minced meat, fishery products, and dairy. These are legal requirements for the categories they cover, but they do not cover every food type.
- National guidance, such as the UK FSA's "Safer Food, Better Business" packs and equivalent documents in EU Member States, which provide practical temperature recommendations for different food categories. These are not legislation, but they often reflect the practical standard an enforcement officer will expect to see applied in your plan.
- Codex Alimentarius general principles of food hygiene, which provide the international benchmark for setting critical limits based on the control measure's ability to achieve the required food safety outcome.
- Validated scientific literature on pathogen thermal death curves, D-values, and z-values for specific organisms in specific food matrices. This is usually the domain of larger manufacturers with technical teams, but the principle applies at every scale: you need to be able to say why the number is what it is.
For most small and medium businesses, the practical approach is to adopt the relevant national guidance values and document that as the basis. The key is consistency: if you state that your cooking critical limit is derived from FSA guidance, apply that same basis across comparable CCPs rather than mixing sources without explanation.
Critical limits for the CCPs that rely on temperature
The table below sets out commonly used critical limits for the main temperature-dependent CCPs. These are starting points, not universal absolutes. You must confirm they are appropriate for your specific product, process, target organisms, and the national guidance that applies in your jurisdiction.
| CCP | Typical critical limit | Key condition | Source basis |
|---|---|---|---|
| Cooking raw poultry | 75°C core temperature | Instant reading at the thickest point; validated for Campylobacter and Salmonella reduction | FSA guidance, EU national guidance, Codex |
| Cooking minced meat and formed products | 75°C core temperature | Applies to burgers, sausages, and similar products where surface bacteria are distributed internally | FSA guidance, Codex |
| Cooking whole cuts of beef or lamb served rare | Surface seared to a lethal temperature; core to customer specification where validated safe | Relevant only where external contamination is the main hazard and the process is suitable for that product type; not appropriate for mechanically tenderised, rolled, or pierced cuts unless specifically validated | FSA guidance, scientific validation, sector-specific codes of practice |
| Hot holding | 63°C | Food must remain at or above the limit during holding and be checked at defined intervals | FSA guidance, EU national guidance |
| Chilled storage for high-risk ready-to-eat foods | Commonly controlled at 5°C or 4°C depending on jurisdiction and product | Applies to food that supports pathogen growth; confirm the appropriate limit against national guidance and the product category | Regulation (EC) No 852/2004 and national interpretation |
| Chilled storage for raw meat and poultry | Product-specific statutory limits under Regulation (EC) No 853/2004, often around 4°C or below depending on the raw material | Different materials can have different legal maximum temperatures, so the HACCP plan should reflect the exact categories handled on site | Regulation (EC) No 853/2004 |
| Frozen storage | Commonly controlled at -18°C | Stops microbiological growth but does not eliminate hazards; confirm against the product category and any applicable legal or customer standard | Regulation (EC) No 853/2004 for applicable categories; general industry standard |
| Cooling after cooking | Rapid cooling from hot holding temperatures to the applicable chilled storage limit within a validated timeframe | Time is the critical variable; the process must move through the danger zone quickly enough to control spore germination and outgrowth | FSA guidance, EU national guidance, scientific literature |
| Reheating | 75°C core temperature | Assumes the food was properly cooked, cooled, and stored; reheating must achieve a lethal temperature throughout | FSA guidance, EU national guidance |
| Pasteurisation | Specific time-temperature combination per product | Requires validation for the target organism in that food matrix; not a generic number | Scientific validation, sector-specific guidance |
Setting a target that is not the same as the critical limit
This distinction is important enough to pull out separately. A target is an internal operating value that keeps you safely away from the critical limit. It absorbs normal process variation. If your blast chiller target is 3°C within 90 minutes and your critical limit requires the applicable chilled storage temperature to be reached within a defined period, a batch that hits 4.5°C at 100 minutes has missed the target but may still be within the safety boundary. You investigate the process drift, but you do not necessarily reject the batch because the critical limit was not breached.
Businesses that do not set targets tend to drift closer and closer to the critical limit over time. A chiller running at 4.8°C day after day is technically compliant if the limit is 5°C, but it is one compressor fault away from a deviation. A target of 3°C forces action before the safety margin is gone.
The target should be recorded in the HACCP plan alongside the critical limit, and the monitoring procedure should make clear which value triggers a corrective action on the product and which triggers a process adjustment.
Monitoring that works when the kitchen is under pressure
Monitoring a temperature CCP means taking a measurement, comparing it to the critical limit, and recording the result in time to take action. The method has to be quick, reliable, and doable by someone wearing gloves with a ticket rail full of orders. If it is not, it will not happen.
Probe thermometers
A digital probe thermometer with a thin tip is the standard tool for cooking, reheating, and cooling checks. It must be calibrated. Calibration should be checked at the start of each shift or event using either an ice slurry at 0°C and boiling water at 100°C, adjusted for altitude where relevant, or a certified calibration block. The check itself must be recorded.
Probe placement matters as much as the reading. For solid foods, the probe tip goes into the thickest part. For liquid or semi-solid foods, stir before measuring. For combination products like a pie with a pastry top and wet filling, you may need to check multiple points. Your monitoring procedure should describe the placement method, not just say "take core temperature."
Probe sanitising between readings is a cross-contamination control. Use an alcohol-based probe wipe or a sanitising solution. Keep the wipes next to the thermometer. If the probe goes from raw to cooked without sanitising, you have introduced a new hazard while trying to control the existing one.
Infrared thermometers
Infrared thermometers measure surface temperature, not core temperature. They are useful for checking incoming chilled goods, cold room shelving, and surface temperatures of hot-holding displays, but they are not a substitute for a probe at a cooking or reheating CCP. If you use an infrared thermometer, your HACCP plan should note its limitations and specify where it is valid and where it is not.
Fixed and continuous monitoring
Cold rooms, walk-in freezers, blast chillers, and hot-holding cabinets in larger operations may use fixed probes with digital displays or continuous data loggers. These are excellent for trending and for catching deviations that happen between manual checks. But they need their own calibration checks, and they do not replace the need for a human to verify that the displayed temperature reflects what is actually happening inside the food. A cold room air temperature of 3°C does not guarantee that a tray of cooked chicken in the middle of the rack has reached 3°C within the required time.
Monitoring frequency
The frequency must be sufficient to catch a deviation before unsafe product leaves your control. For batch cooking, every batch is checked. For continuous cooking on a conveyor or in a larger operation, checks are at defined intervals based on process variability. For hot holding, a check every 30 to 60 minutes is typical. For cold storage, a manual check at the start and end of the day plus continuous logging where available.
Corrective actions that are actually executable
A corrective action at a temperature CCP has two parts: what happens to the product, and what happens to the process. Both must be written clearly enough that the person on shift can execute them without calling a manager unless the decision genuinely requires escalation.
For a cooking deviation where the core temperature did not reach the critical limit, the corrective action is typically to continue cooking until the limit is reached, verify the temperature, and record both the deviation and the final reading. Product is not rejected unless it cannot be brought to temperature without unacceptable quality deterioration, or unless the time spent below the critical limit creates a concern that toxins from spore-forming bacteria may have been produced.
For a hot-holding deviation where the temperature drops below 63°C, the corrective action depends on how long it has been below the limit. If the drop is caught within two hours and the food can be reheated to above 75°C and placed back into hot holding, that is one path, but it must be documented as a deviation. If the time below 63°C is unknown or exceeds two hours, the food is typically rejected. This framework should be aligned with the guidance you rely on in your HACCP plan.
For a chilled-storage deviation where the temperature exceeds the critical limit, the product decision depends on the type of food, the temperature reached, and the duration. Raw meat that reaches 8°C for 30 minutes before being moved to a functioning chiller is a different risk profile from ready-to-eat cooked prawns that sit at 12°C for four hours during a cold-room breakdown. Your corrective action procedure should not try to cover every scenario with a single rule. It should provide a decision framework and specify who is authorised to make the product disposition call.
For all deviations, the record must show the CCP, the time, the measured value, the critical limit, the action taken with the product, the action taken with the process, and the signature of the person who made the decision. A blank corrective action column on a temperature log, or one that says only "adjusted thermostat," will not satisfy an officer or auditor if the deviation occurred at a genuine CCP.
Verification that the temperature system is still working
Verification is the periodic check that your monitoring and corrective actions are actually controlling the hazard. It is not the same as monitoring. For temperature CCPs, verification typically includes:
- Reviewing temperature logs for completeness, out-of-spec readings, and corrective action records at least weekly
- Calibrating probe thermometers against a reference standard at a defined frequency
- Checking fixed temperature sensors against an independent calibrated probe
- Observing staff taking temperatures to confirm they use the correct method and probe placement
- Reviewing deviation trends across multiple weeks to spot a process that is drifting toward the critical limit
The verification record should be separate from the monitoring log and should include any follow-up actions. If a verification review finds that several hot-holding checks in a week were below target but above the critical limit, that is a process problem, not a CCP failure, and it needs a different response: maintenance, retraining, or a revised target.
The cold chain gap that catches caterers and small producers
One temperature control failure that recurs in inspection reports and food safety audits involves the cold chain between delivery acceptance and storage, or between storage and service. The delivery arrives, the temperature is checked and logged, and then the food sits on a loading bay or in a corridor for 25 minutes while someone deals with something else. The same thing happens at events when chilled food is removed from a refrigerated van and waits on a trestle table while the stall is set up.
The HACCP plan rarely accounts for these transitional periods because the flow diagram treats "chilled storage" as a static state. If your operation has identifiable gaps between temperature-controlled steps, they belong in the hazard analysis. The control may be a simple operational rule, such as moving chilled food from delivery to cold room within a defined time after acceptance, but it needs to be written down and occasionally verified because it is in these gaps that temperature abuse quietly accumulates.
Building a temperature monitoring log that people will actually complete
A poorly designed log invites bad behaviour. If the columns do not match the CCPs, if the critical limits are not printed on the form, or if there is no obvious place to record a deviation and corrective action, the person filling it in will either leave gaps or write what they think you want to see.
A functional temperature log should have:
- The CCP or check name clearly stated at the top of each section
- The critical limit printed on the form so it cannot be forgotten or misremembered
- The target printed alongside if you use one
- A column for the actual reading
- A column for the time
- A corrective action section that starts with "Product:" and "Process:" to prompt the right details
- A signature or initials column
- A date and shift identifier
The form should separate CCP checks from PRP ambient checks. Mixing them on the same log causes confusion and makes it harder for an auditor to see which readings controlled product safety and which controlled the environment.
Next step
Temperature control in HACCP fails most often not because the science is wrong but because the limits, the monitoring methods, and the paperwork do not match the reality of the kitchen or production floor. If your temperature logs show perfect compliance week after week with no deviations and no corrective actions, the system is almost certainly not being used properly because real food operations produce real variation.
If you need to rebuild the monitoring side, a temperature monitoring log template that separates CCP checks from ambient checks and prints the critical limits directly on the form removes a lot of the ambiguity that leads to blank boxes and invented numbers. If you are working through the whole HACCP plan and want to make sure the hazard analysis, CCP decisions, and temperature limits are coherent before you write the monitoring procedures, the HACCP plan generator walks you through that logic and produces a plan where temperature control sits inside a complete, auditable system. For a structured starting point that you can adapt to your own product and process, the HACCP plan template includes the CCP decision framework and corrective action structure that make temperature deviations manageable rather than chaotic. If you want to keep temperature checks, corrective actions, and daily records in one workflow, you can also create a free account.