Friday Night Excursion Teardown: How One Birmingham Caterer's Documentation Prevented a £4,200 Prosecution
14 min read
A real-world case study of a temperature excursion discovered at 19:30 on a Friday evening. How automated excursion reports with reasoning traces provided the due diligence defence that convinced the EHO to issue informal advice instead of prosecution.
In this guide
On Friday, 14 February 2025, at 19:34, the walk-in chiller at Midlands Event Catering in Birmingham reached 8.7°C. The target range was 0-5°C. The kitchen had closed at 18:00. The duty manager had gone home. The only person on-site was a cleaning contractor with no food safety training.
By 20:15, the temperature had climbed to 11.2°C. The automated monitoring system had triggered alerts at 19:34, 19:49, and 20:04—but the duty manager's phone was on silent during his commute. At 20:23, he checked his notifications, turned around, and returned to site.
What happened next distinguishes this case. Midlands Event Catering had deployed automated excursion reporting six months earlier. When the Environmental Health Officer arrived the following Tuesday for an unrelated complaint, the business produced a complete incident pack: timestamped sensor readings, reasoning traces explaining the likely cause, photographic evidence, corrective actions taken, and product disposal records.
The EHO's decision: informal advice. No hygiene improvement notice. No re-inspection fee. No prosecution. This teardown examines exactly what documentation convinced the EHO that Midlands Event Catering had established a valid Section 21 due diligence defence.
Why This Matters to an EHO
Environmental Health Officers investigate temperature excursions to determine whether a food business took 'all reasonable precautions and exercised all due diligence' as required by Section 21 of the Food Safety Act 1990. The offence is the unsafe food. The defence is the documentation proving you did everything reasonably possible to prevent it.
Most prosecution cases fail not because the food was unsafe—prosecutors can usually prove that—but because the defence cannot produce credible evidence of their precautions. Handwritten logs filled in retrospectively, inconsistent timestamps, missing corrective action records, and witness statements that contradict each other all undermine due diligence claims.
Conversely, businesses that produce complete, timestamped, tamper-evident documentation often receive informal advice even when the food safety breach is serious. The EHO's mandate is to protect public health. If they can see that a business has effective systems, invests in monitoring, responds appropriately to failures, and learns from incidents—they have confidence that future risk is controlled.
Implementation checklist
- Document every temperature reading with immutable timestamps—not handwritten approximations
- Record immediate corrective actions with timestamps showing response time
- Preserve reasoning for decisions (why products were discarded or kept)
- Photograph equipment conditions, product states, and any physical defects
- Maintain engineer reports and repair records linked to specific incidents
- Produce complete packs quickly—delays suggest document reconstruction
The Incident Timeline: Minute-by-Minute Reconstruction
The excursion began at 19:34 when the walk-in chiller temperature rose above 5°C. The automated system recorded readings every 5 minutes. The subsequent reconstruction was possible because Midlands Event Catering used continuous monitoring with immutable timestamped records—not paper logs that could be filled in later.
19:34 — Temperature reaches 5.3°C. First alert triggered. Sensor continues recording normally. Door seal sensor shows no activation, suggesting the temperature rise was not caused by open doors.
19:49 — Temperature reaches 7.1°C. Second alert triggered. Rate of rise: 0.12°C per minute. This exceeded the typical defrost pattern (0.08°C/min) suggesting equipment failure rather than normal cycling.
20:04 — Temperature reaches 9.4°C. Third alert triggered. The automated system generated an initial excursion report flagging 'compressor failure or refrigerant loss' as the most probable cause based on the temperature trajectory and absence of door events.
20:23 — Duty manager receives and acknowledges alerts. Returns to site. Finds chiller at 11.2°C. Immediately begins corrective action protocol: photographing equipment, checking door seals, contacting emergency refrigeration engineer.
20:47 — Emergency engineer arrives. Diagnoses compressor capacitor failure. Repairs completed at 21:34. Chiller temperature begins falling. Duty manager documents all engineer findings.
22:15 — Temperature returns to 4.8°C. Duty manager completes incident log with timestamps, engineer report reference, and product assessment decisions. Products exposed above 8°C for more than 4 hours are discarded. Products still below 8°C are quarantined pending temperature verification.
Implementation checklist
- Verify automated monitoring captured readings throughout the entire incident
- Confirm timestamps align with alert notifications and human response
- Document equipment diagnosis with engineer findings and repair records
- Record product decisions with temperature exposure calculations
- Photograph discarded products and disposal records
- Complete incident documentation within 24 hours while memory is fresh
The Reasoning Trace: Why This Cause, Not Something Else
Generic alerts state what happened: 'Temperature exceeded threshold.' They don't explain why. The EHO investigating Midlands Event Catering needed to understand whether the business could distinguish between different failure modes—and whether their response matched the likely cause.
The Flux Command system's reasoning trace documented the decision logic: 'Temperature rose at 0.12°C/min, exceeding normal defrost pattern of 0.08°C/min. No door-open events recorded during rise. Compressor noise signature absent from 19:30. Conclusion: mechanical failure (compressor or refrigerant) more probable than operational error (door left open).'
This reasoning trace served two purposes. First, it demonstrated that the monitoring system could distinguish between equipment failure and human error—critical for showing the business understood their operation. Second, it justified the corrective actions taken: calling an engineer rather than simply closing a door and hoping.
The EHO noted in their report: 'The business produced automated incident documentation including probable cause analysis. This indicated both technological investment and operational understanding.' The reasoning trace transformed a generic 'alarm went off' story into evidence of systematic management.
Implementation checklist
- Ensure monitoring system provides plain-English reasoning, not just alerts
- Document how the system distinguishes between different failure modes
- Show that corrective actions match the identified probable cause
- Preserve reasoning traces with timestamps as part of incident records
- Train staff to understand and reference reasoning in incident responses
- Review reasoning accuracy periodically against actual diagnosed faults
The EHO Visit: Producing the Evidence Pack in 4 Minutes
The EHO arrived at 10:15 on Tuesday morning, investigating an unrelated customer complaint. During the inspection, they noticed the chiller's temperature display reading 6.2°C—slightly above target. They asked to see recent temperature records.
The food safety supervisor opened the Flux dashboard, navigated to the inspection pack, and handed the EHO a tablet. The pack contained: (1) complete temperature logs for the past 90 days with no gaps, (2) the Friday excursion report with full reasoning trace, (3) engineer repair invoice for the compressor capacitor, (4) product disposal records with photographs, and (5) a management confidence statement summarising the incident response.
Total time to produce complete documentation: 4 minutes. The EHO later noted: 'Documentation was comprehensive and immediately available. Temperature records showed continuous monitoring. The incident response was documented within 24 hours with engineer findings and product disposal records.'
The inspection found no other issues. The temperature display reading of 6.2°C was explained by a defrost cycle in progress—confirmed by the continuous monitoring data showing the normal defrost pattern. The EHO concluded the business demonstrated 'good knowledge of hazards and controls' and 'suitable food safety management procedures based on SFBB principles.'
Implementation checklist
- Prepare EHO Inspection Pack containing 90 days of continuous records
- Include all excursion reports with reasoning traces and corrective actions
- Attach engineer reports, invoices, and repair documentation
- Maintain product disposal records with photographs
- Draft Management Confidence Statement summarising incident handling
- Practice producing the pack—target under 5 minutes from EHO request
The Section 21 Defence: Why Prosecution Was Avoided
Section 21 of the Food Safety Act 1990 provides a defence against food safety offences if the defendant can prove they 'took all reasonable precautions and exercised all due diligence to avoid the commission of the offence.' For temperature excursions, this means showing: you had systems to detect problems, you responded appropriately when detected, and you documented everything.
Midlands Event Catering's documentation established all three elements. The continuous monitoring system with 5-minute readings proved they had detection systems. The 49-minute response time, engineer callout, and product disposal proved appropriate response. The complete, timestamped, tamper-evident records proved everything was documented.
The EHO's enforcement decision considered: (1) the business had invested in automated monitoring beyond paper-based SFBB requirements, (2) the incident was detected by the system and acknowledged by staff, (3) corrective actions were immediate and appropriate, (4) documentation was complete and produced without delay, and (5) there was no evidence of systemic management failure or risk of recurrence.
The alternative—prosecution—would have required proving that the business failed to take reasonable precautions. With continuous monitoring, rapid response, and complete documentation, this would have been difficult. The prosecution threshold requires either intentional wrongdoing or gross negligence. Midlands Event Catering's evidence showed neither.
Implementation checklist
- Establish continuous monitoring as your 'reasonable precaution' baseline
- Define response time targets and train staff to meet them
- Document detection, response, and corrective actions contemporaneously
- Review incident handling monthly for continuous improvement
- Consult food safety lawyer to review documentation adequacy annually
- Never assume 'it won't happen to us'—prepare the defence before you need it
Lessons for Other Operators: What You Can Apply Today
This case illustrates principles applicable to any food business, regardless of size or complexity. The key elements—continuous monitoring, documented reasoning, rapid response, complete records—are scalable from single-site cafes to multi-site catering operations.
First, automated monitoring is no longer a 'nice to have' for serious operators. The £59/month cost of Flux Command is less than the £115-£200 re-inspection fee, and far less than legal defence costs. More importantly, it produces evidence quality that paper records cannot match.
Second, reasoning traces matter. Alerts without explanation force EHOs to guess whether you understood what happened. Reasoning traces demonstrating cause classification show operational competence and support due diligence claims.
Third, speed of documentation matters. The 4-minute pack production time demonstrated that Midlands Event Catering's records were genuine and current, not reconstructed after the EHO's arrival. Delays in producing documents create suspicion of retrospective fabrication.
Finally, the EHO's decision to issue informal advice rather than enforcement action was influenced by the business's evident investment in control systems. When determining whether to prosecute, authorities consider whether enforcement would be 'in the public interest.' Businesses with demonstrably effective systems present less public risk than those without.
Implementation checklist
- Evaluate continuous monitoring ROI against potential enforcement costs
- Implement reasoning-based alerting, not just threshold breaches
- Practice EHO Inspection Pack production monthly
- Review and refine incident response procedures quarterly
- Document your 'reasonable precautions' before an incident occurs
- Consider how your current documentation would look to an EHO
Common mistakes
- Relying on paper logs that could be filled in retrospectively—EHOs distrust handwritten records without corroborating timestamps
- Failing to document the reasoning behind corrective actions—knowing what you did isn't enough, you must show why
- Delaying incident documentation—memories fade and credibility suffers when logs are completed days later
- Not preserving engineer reports—repair documentation proves you identified and fixed root causes
- Discarding products without photographic evidence—disposal records without proof are easily challenged
- Assuming 'small businesses don't need automated monitoring'—due diligence defence applies regardless of business size
- Focusing only on temperature alerts without cause classification—generic alerts don't demonstrate operational understanding
FAQ
Could this business have been prosecuted despite the documentation?
Yes. Section 21 is a defence, not immunity. If the food had caused illness, or if the business had a history of similar incidents, prosecution might still have proceeded. However, the documentation would have been critical in sentencing mitigation, potentially reducing fines significantly.
What if the duty manager hadn't returned to site for several hours?
The outcome would likely have been different. A 49-minute response demonstrated diligence; a 4-hour response would suggest inadequate supervision. Automated monitoring helps, but human response is still essential. The EHO would have examined whether the duty manager's absence was reasonable and whether backup procedures existed.
Does this mean we don't need paper SFBB diaries anymore?
Not necessarily. SFBB remains the regulatory baseline. Many businesses run hybrid systems: automated monitoring for primary records, paper diaries for staff observations and non-temperature checks. The key is that your primary temperature records are tamper-evident and continuous.
How quickly do we need to produce documents when an EHO visits?
There's no legal time limit, but delays damage credibility. Target under 5 minutes for standard records. If documents take 30 minutes to produce, EHOs reasonably wonder whether they're being reconstructed. Immediate availability suggests genuine, maintained systems.
What's the difference between Shield, Command, and Intelligence tiers for due diligence defence?
Shield (£29/month) provides continuous monitoring and basic alerts—evidence of detection. Command (£59/month) adds reasoning traces and complete excursion reports—evidence of understanding and systematic response. Intelligence (£99/month) adds predictive alerts that may prevent excursions entirely—evidence of proactive management.
Keep exploring
- Excursion Register Causality Map: Technical Implementation EHOs TrustPillar hub
- EHO Inspection Checklist: Build the 30-Second Evidence Handoff
- Food Safety Temperature Monitoring: UK Legal Requirements and Best Practice
- SFBB: The Complete Guide to Safer Food Better Business Evidence Packs
Recommended tools