Water Safety Device Audit Card

A buying list goes out of date the week a new product ships, so this card tests the devices you already own instead. Four questions sort any alarm, cover, camera, wearable or app into a layer of protection or a layer of false confidence. The short version: almost everything electronic announces rather than prevents, and all of it fails silently. An alarm nobody has deliberately triggered this month is an assumption, not a layer.

WaterWiseKids.com — Free water safety education for families

🧪 The Four Questions — Audit Any Device in Two Minutes

  • Question 1 — does it prevent, or does it only announce? A fence stops a child reaching the water. An alarm tells you a child already has. Both are worth owning; only one buys you time. Almost everything with a battery in it announces.
  • Question 2 — what has to already be true for it to fire? Armed, charged, in range, worn, gate fully shut, phone on and unmuted. Write the list out. Every item is a way the device can be quietly out of service.
  • Question 3 — who hears it, and how far are they from the water? Set it off, then listen from where you actually are when the pool is unattended. Time the walk to the water’s edge. That number is your real capability.
  • Question 4 — does owning it make the adults look less? Ask it out loud, of everyone who supervises: since we installed this, are we watching more or less? This is the only question on the card about people rather than equipment.
  • Score it. Fails Q1 → normal, but it must sit behind a barrier, never instead of one. Three or more silent failures in Q2 → it needs a written monthly test. Fails Q3 → move the receiver before you buy anything else. Fails Q4 → net loss until the habit changes.
Why test the device rather than shop for one. A recommended-products list only covers devices somebody has already reviewed, and it cannot help with the alarm that came with the house, the hand-me-down band in the pool bag, or the app on a phone now switched off in a drawer. These four questions work on anything, including something released next year.

🚧 Prevent vs. Announce — Sort Every Device You Own

  • Prevention column: four-sided isolation fencing at least four feet high separating the pool from house and yard; a self-closing, self-latching gate with the latch out of a small child’s reach; high locks on doors leading out; a power safety cover; an above-ground ladder secured or removed.
  • Announcement column: every pool alarm type, door and gate alarms, wearable immersion bands, cameras, monitoring apps, trackers. All useful. None of them keeps a child out of the water.
  • If the prevention column is thin, stop buying from the announcement column. More detection does not add up to a barrier, and the spend usually feels like progress while the actual risk has not moved.
  • Know which alarm type you own — the failure modes differ. Type A surface: floats and senses waves; prone to wind false alarms, and cannot be used under a cover. Type B subsurface: senses disturbance below the waterline; works with a cover. Type C perimeter: detects movement at the pool edge, before entry; can be blocked by furniture or a robot cleaner. Type D personal immersion: worn by the child; only protects the child wearing it today.
  • Door and gate alarms are a separate category and belong specifically where the house forms one side of the pool barrier. They are on the barrier, not on the pool.
This is not a house opinion — it is what the standard and the pediatricians both say. ASTM F2208, the safety specification residential pool alarms are built to, states it is not intended to replace adult supervision, fences, gates and locks. The American Academy of Pediatrics’ 2026 drowning prevention policy statement names four-sided isolation fencing with working self-closing, self-latching gates among the strategies with evidence behind them; alarms are not on that list. Keep the alarm. Keep it behind the fence.

🔋 The Silent Failure List — What Has to Be True Today

  • Batteries are live in every device — replaced on a schedule, not when the low-battery chirp starts. A chirp in an empty yard is not a warning anybody hears.
  • The system is armed right now, not switched off after the last time it woke the baby. Check the panel, not your memory of the panel.
  • The gate actually closes to the point the sensor reads shut, and the magnet is still aligned after a season of being knocked about. A gate that latches is not the same as a gate the alarm agrees is latched.
  • The floating alarm is in the water and not beached on a step, and nobody has muted it because it kept going off in the wind.
  • The wearable is on the child — which means somebody remembered, this afternoon, for this child, including the visiting one.
  • The app is running, notifications are on, and the phone is charged, in range and in somebody’s hand. Treat a phone alert as the least reliable link in the chain, always.
The reason this list exists. A device that failed and a device that saw nothing are indistinguishable from the kitchen. Silence is the normal, reassuring output of every one of these products, and it is also exactly what a flat battery produces. There is no other household safety equipment we accept that from.

📣 The Reach Test — Who Actually Hears It

  • Set the device off, then walk indoors and listen. Kitchen with the extractor on, laundry, far bedroom, front room with the television going and the door shut. Can you hear it? Can somebody who is not you hear it?
  • Use a remote receiver inside the house. The U.S. Consumer Product Safety Commission advises exactly this, so the alarm can be heard inside and in other places away from the pool area. A poolside siren announcing a fall to an empty deck did its job and saved nobody.
  • Time the walk from where you heard it to the water’s edge — through the door, down the hall, past the gate you have to unlatch. Write the number down. It is the honest measure of your detection layer.
  • If that number is long, the answer is not a louder alarm. It is a barrier, and an adult closer to the water. Volume cannot fix distance.
  • Re-run the reach test after anything changes in the house — a new appliance, a bedroom that moved, a door that now stays shut, a phone that changed hands.

⏱️ The Sixty-Second Arm Test — First Saturday of the Month

  • Step 1 — confirm armed and powered. Fresh batteries on schedule. Panel showing armed.
  • Step 2 — trigger it the way a child would. Open the gate. Open the house door. Drop a weighted object into the water. Wet the wearable. Do not test by pressing the test button — that only proves the speaker works.
  • Step 3 — confirm the alert reaches you where you normally are. Not at the poolside. Indoors, with the door shut.
  • Step 4 — write the date on this card. A test you remember doing and a test you did not do look identical a month later.
  • Do the same pass at pool opening and pool closing, and any time a device has been moved, serviced, replaced or silenced.
Four lines, once a month. That is the entire difference between a layer of protection and an assumption you have been carrying since the day it was installed. The first pass usually finds something.

👀 The Vigilance Check — Is the Device Costing You Attention?

  • Since we installed this, do we check the yard as often? Ask it out loud, of every adult who supervises, and accept the honest answer.
  • Have we ever said yes to a swim, or to a few minutes indoors, because the alarm was on? That is the trade happening silently. Name it.
  • Do visiting adults know what the devices do and do not do? A guest who hears “we have an alarm” will supervise less, not more, unless you say what it is for.
  • Is there still a named water watcher with no phone whenever children are in the water? Detection catches the moment supervision fails. It does not license the failure.
  • If the answer to the first question is “less”, the device is a net loss until the habit changes. That is an argument for changing the habit, not for removing the device.

🧮 Audit Every Device You Own

List each device — alarms, cover, locks, camera, wearable, app. Mark P or A for prevent or announce, count the silent failures, tick whether the reach test passed, and write the date you last triggered it on purpose.
Device / where it isPrevent or announce?Silent failures (count)Reach test passed?Last triggered on purpose
     
     
     
     
     
     
Reading the grid. A column of A’s with no P is the finding, not a failure of the card — it means the yard is well monitored and poorly fenced. Any row with a blank date is a device you are currently counting on without evidence.

✅ Your Family Device Plan

Fill this in once, then keep the card inside the pool shed door or on the utility board. Re-run the audit on the first Saturday of each month and at pool opening and closing.
Our prevention layer — what it is
Gaps in it we know about
Devices we own — how many
Where the indoor receiver lives
Seconds from alert to water’s edge
Whose job the monthly arm test is
Date of last full arm test
Named water watcher when kids swim
What we tell visiting adults
Date agreed

Read the Detail Behind This Card

Keep reading — the guides behind every question on this card.

The Only Layer That Goes Into the Water With Your Child

Every device on this card works on the approach. The instant a child is in the water, all of them have already done everything they are going to do — and none of them exist at the neighbour’s pool, the hotel, the lake or the party. The layer that travels is the taught sequence: reach the surface, roll onto the back, float and breathe, turn, and get to an edge.

Find Swim Lessons Near You