Skip to main content

Prove It

Prove It is SafetyBrik's live food safety proving log. It sits at /safety/prove-it and captures the temperature checks, calibration tests and chilling records you would otherwise scribble onto an SFBB diary page.

What Prove It is for​

The Safer Food Better Business pack asks you to prove that core safety controls are happening - that food is being cooked to a safe core temperature, hot-held above 63°C, chilled down quickly enough, and that your probe thermometer is actually accurate. Prove It is the digital version of those proving records.

When an EHO arrives, you can hand over your device and walk them through the log: every cooking check, every chilling-down readings sequence, every probe calibration, with timestamps and pass/fail outcomes. No printed folder, no missing pages.

Recording categories​

Prove It supports the categories the SFBB pack uses:

  • Cooking - core temperature check after cooking
  • Reheating - core temperature on a reheat
  • Hot holding - hot-held food sampled during service
  • Chilling down - a sequence of readings as a hot product cools
  • Chilled storage - fridge or display readings
  • Freezing - freezer or blast-chilled readings
  • Probe calibration - ice-water and boiling-water test for your probe

Each category has its own pass mark, shown on the form as a hint, so you can see straight away whether a reading passes:

  • Cooking and Reheating - 75°C or above, from one reading. SFBB also accepts 70°C held for 2 minutes, but one reading can't show the hold, so that can only be recorded by an owner or admin as an override, with a reason (see Overriding a failed reading). In Scotland, reheated food must reach 82°C or above (see Food Standards Scotland's CookSafe). SafetyBrik's pass mark follows SFBB's 75°C, so a business in Scotland should check for 82°C when reheating.
  • Hot holding - 63°C or above.
  • Chilling down - 8°C or below on the last reading, ideally within 90 minutes. Take the first reading as the food comes off the heat, and add the time of each reading. A chill that takes longer than 90 minutes still passes, with a warning (see Chilling down).
  • Chilled storage - 8°C or below, with 5°C or below recommended. A reading between 5°C and 8°C passes, with an amber "Pass. Above the recommended 5°C".
  • Freezing - -18°C or below.
  • Probe calibration - iced water -1°C to 1°C and boiling water 99°C to 101°C.

Summary cards​

At the top of the page three cards give you the at-a-glance picture for the selected location:

  • Last passed calibration - when your probe last passed a Probe calibration check, and whether it is overdue (calibration is recommended every 4 weeks). Green if up to date, amber if approaching due, red if overdue or never done. With none it reads "No passed calibration on record". A failed calibration never counts, here or anywhere else: not on the Overview's alert, the safety score, the weekly reminder, the 4-weekly review's prompt or the EHO Report Pack. It only counts as overdue once you are proving a temperature safe method: until then there is nothing for the probe to prove.
  • Records this month - the total number of proving records logged this month on the site's calendar (your business's time zone when no site is chosen), broken down by category.
  • Methods due - temperature safe methods due a Prove It check. Only the seven temperature-control methods count: Cooking Safely, Reheating, Hot Holding, Chilling Down Hot Food, Chilled Storage & Displaying Chilled Food, Freezing and Defrosting. A method counts only once it has been started or completed at a site, each site's copy counts once, and each one is due when it has never been proved or its last passed check is older than its own interval (28 days unless changed). Methods you haven't started never count.

Adding a new record​

Click New record and the dialog adapts to the category you pick.

  • For single-temperature categories (cooking, reheating, hot holding, chilled storage, freezing) you enter one Temperature in °C.
  • For Chilling down you add a sequence of readings, each with a temperature and a time, so you can show the curve from hot to cold. Every reading needs its time: until each one has it, the form says "Add the time of each reading." and the record can't be saved.
  • For Probe calibration you enter an iced-water reading and a boiling-water reading. To pass, iced water must read -1°C to 1°C and boiling water 99°C to 101°C. Outside those, replace the probe or have it recalibrated.

Readings from -60°C to 250°C can be entered. Pressing save with something missing, or a reading outside that range, saves nothing: each problem shows under its field, for example "Say what you proved.", "Enter the temperature." or "Enter a reading between -60 and 250°C.", and you are taken to the first one.

Other useful fields:

  • What did you prove? - a short note, for example "Checked chicken breast core temperature after cooking"
  • Safe method (optional) - link the record to one of your safe methods. The list shows only the temperature methods at the site you picked that the chosen category proves, and picking a method sets the category to match. Only a passed check marks the method as proved; a failed one is recorded but doesn't count. A method from another site, or one that this category can't prove, is refused with a plain message. If the list can't be loaded, it says "We couldn't load this site's safe methods." with Try again
  • Photo (optional) - take or upload a photo as evidence: JPEG, PNG or WebP, up to 5 MB. A HEIC photo is refused with "This photo is in HEIC format. Save it as a JPEG and try again." Photos are private to your business, and Brikly opens them through links that work for an hour
  • Location - if you run multiple sites, pick which one this record belongs to

Once every reading is in, a banner shows the result: a green "Pass. Within the safe limit", an amber "Pass. Above the recommended 5°C" (chilled storage), or a red "Fail. Outside the safe limit". A passed chill shows how long it took instead, in green or amber (see Chilling down). The button says what will go on file: Record result until the readings decide it, then Record pass or Record fail. A failed check can still be recorded with Record fail, and the failure stays on the record. It records with "Prove It check recorded" (or "Prove It check recorded as a fail"). If it can't be recorded, the dialog stays open and says why, for example "We couldn't record this check. Try again."

Recorded checks can't be changed. If something is wrong, record a new check: the wrong one stays visible in the log.

When a check fails​

A failed check asks What did you do about it?, for example "Cooked it for 5 more minutes and checked again". The box shows only on a fail, and it is optional: "Leave this blank and we'll add a follow-up so it isn't forgotten."

  • Filled in - what you typed is saved with the check, and shows on it as "What was done: Cooked it for 5 more minutes and checked again". The check records with "Prove It check recorded as a fail".
  • Left blank - the check records with "Recorded as a fail. A follow-up is on Today.", and a follow-up is added for whoever recorded the check: "Say what was done about the failed cooking check". It is due by the end of the next day at the check's site, and it shows on the site's Today and on the Follow-ups page. Closing it, with what happened, is how what was done gets recorded.

A failed check that an owner or admin saves as a pass (see Overriding a failed reading) never gets a follow-up, and the line under the box says so: "This is saved as a pass (manager override), so no follow-up is added." Anything typed in the box is still saved with the check.

If your device is offline, the check isn't sent: "You're offline, so this wasn't sent. What you typed is still here. Try again when you're back online." The dialog stays open with what you typed.

Owners, admins and members can record checks and add photos. Viewers can see the log but can't record checks.

Chilling down​

A chill passes when its last reading is 8°C or below. SafetyBrik also works out how long it took, from the first reading (taken as the food comes off the heat) to the first reading at 8°C or below:

  • Within 90 minutes - a green "Pass. Reached 8°C in 85 min".
  • Longer than 90 minutes - an amber "Pass. Took 6 h 30 min to reach 8°C."
  • Not known - with one reading, or when the first reading is already 8°C or below, the time can't be worked out: an amber "Pass. How long it took to reach 8°C isn't known. Take the first reading as the food comes off the heat."

Either amber warning is still a Pass: the check proves its safe method and needs no override. But it is never a clean pass, so the warning stays with the record: on its own amber line in the log ("Took longer than 90 minutes." or "How long it took to reach 8°C isn't known."), in the diary and in the EHO Report Pack. A chill whose last reading is above 8°C is a Fail, shown without a time or a warning.

Overriding a failed reading​

Sometimes food is safe in a way one reading can't show, such as a dish held at 70°C for 2 minutes. For a failed reading, owners and admins see Override with a reason (optional). With a reason written, the button reads Record pass (override), and the check is recorded as Pass (manager override), with the reason shown:

  • on the log, as "Manager override: (reason)"
  • in the diary, with an amber badge
  • in the EHO Report Pack, as "Manager override: ..."

The override is recorded in the audit log as Prove It manager override. Members and viewers can't override a check. An overridden check is a pass, so it gets no follow-up (see When a check fails).

A failed probe calibration can never be overridden, by anyone: "A probe that fails calibration can't be overridden. Replace it or have it recalibrated, then check it again."

Proving a method from its page​

On a temperature safe method's page, Prove this method opens the New record form with the category, site and method already filled in. The method's Last proved badge shows when it last passed a check. Other safe methods don't have the button or the badge, because there is no temperature to prove.

Closed sites​

A site that has been switched off or removed keeps its Prove It records under a read-only "Station Road (closed)" tab, with "Station Road is closed. Its records can be read and printed but not changed." and no New record. The New record form offers open sites only, and a check sent for a closed site from a page left open is refused with "This site is closed, so its records can be read and printed but not changed." A Prove this method link opened after its site has closed shows that site's checks, read only, rather than the form.

Filtering the log​

The log shows your records newest first, 50 at a time, with Load more for the next 50. If the log can't be loaded, it says so with Try again. Filters at the top let you slice by:

  • Category - cooking, chilling, calibration, etc.
  • From / To - a date range
  • Result - all, passed, or failed

Each row shows:

  • the category and what you proved
  • the reading, for example "76°C", "Ice 0.2°C, boiling 99.8°C", "62°C to 6°C in 85 min (3 readings)", or "72°C to 4°C (3 readings), 8°C or below in 3 h 25 min" when the chill reached 8°C before its last reading. A chill's warning shows on its own amber line underneath ("Took longer than 90 minutes." or "How long it took to reach 8°C isn't known."). A failed chill shows only its readings, for example "72°C to 10°C (3 readings)"
  • the result: Pass, Fail or Pass (manager override), with "Manager override: (reason)" underneath an override
  • when it was recorded, and who recorded it by name ("Former team member" if they are no longer on file)
  • the site, in a business with more than one, and the photo if one was attached
  • for a failed check, what was done about it once it is known (see What was done), and its follow-up (see A failed check's follow-up)

Chilling down rows expand to show every reading in the sequence.

A failed check's follow-up​

A failed check carries its follow-up on its card, under Follow-up, with the same buttons as on the Follow-ups page:

  • Close follow-up - say what happened. What you write becomes what was done about the check.
  • Change - reword it, move it to another day or give it to someone else.
  • Void - for owners and admins, when it was added by mistake, with the reason.
  • Add follow-up - once a follow-up has been voided, another can be added. It starts with the same wording and is due the next day. Leave Who blank and it goes to whoever recorded the check: "Optional. Leave it blank and it goes to whoever recorded the check."

A check has one follow-up at a time. Once one is closed, or once what was done is typed with the check, there is no Add follow-up. Earlier follow-ups, closed or voided, stay listed under the current one, so the card keeps their history.

A failed check with nothing recorded about it and no follow-up, such as one recorded before this was added, shows Add follow-up too. Nothing can be added at a site that has been switched off or removed.

What was done​

Once it is known, what was done shows on the failed check's card:

  • Typed with the check - "What was done: Cooked it for 5 more minutes and checked again".
  • From its follow-up - the closing note, with the day it was added and who added it, for example "What was done: Binned the batch and cooked a fresh one, added 12 Nov by Sam Taylor". The day is on the check's site's clock, with the year when it isn't this year.

What you typed with the check comes first: when there is both, that is the line shown.

The same line is printed under the check in the EHO Report Pack and shown under it in the diary. A fail whose follow-up is still open has no line yet: the follow-up itself is in the pack's Follow-ups section.

Opening one check​

A follow-up's link to its check opens the Prove It page with that check at the top, under "The check you opened", highlighted, whatever the filters or page of the log. Dismiss takes it away. The page address carries the check (/safety/prove-it?record=...), so the link can be shared. If the check can't be found it says "This record can't be found. It may belong to another business.", and if it can't be loaded, "We couldn't load this record." with Try again. Either way, Show the log goes back to the log.

Weekly reminders​

Every Monday morning, SafetyBrik checks Prove It for each business:

  • Safe methods due a Prove It check - for example "1 temperature safe method is due a Prove It check. Record a probe reading to show each one works." It follows the same rule as the Methods due card.
  • Probe calibration overdue - sent only to businesses that have started or completed one of the temperature safe methods at an open site, when the last passed calibration is more than 4 weeks old (or there has never been one). A failed calibration doesn't reset it. A site that has been switched off or removed doesn't count.

The Overview shows the same two as alerts. The safety score uses the same calibration rule: calibration counts against a site only while it is overdue, so a failed calibration at a site that isn't proving a temperature method yet costs no points.

Why this replaces the printed pack​

The printed SFBB diary asks you to write the same things by hand - dates, temperatures, who did the check, whether it passed. Prove It captures all of it digitally with three advantages:

  • Auto pass/fail - pass marks are applied as you type, so you cannot accidentally log an unsafe reading without seeing the warning. A failed reading is recorded as Fail unless an owner or admin overrides it with a reason
  • Searchable - filter by date or category in seconds rather than thumbing through paper
  • Photo evidence - attach a probe reading photo straight from the camera

It also feeds into Reports and the EHO Report Pack PDF, so you never have to compile this information manually.

tip

Prove It is the snapshot in time view - what is happening right now and this month. For the printable EHO pack covering any dates you choose, head to Reports.