What PoolBoss Says
Float a filled bucket on the pool step, mark the water line inside and outside, leave the pump on its normal schedule, and compare after 24-48 hours. If the pool drops more than the bucket, it is leaking; equal loss is evaporation, which runs roughly a quarter inch a day in Sunbelt summer heat.
You have topped off the same pool three visits running. The customer has noticed the hose, and sooner or later they are going to ask whether they have a leak - which is really a question about a repair bill, so the answer needs to be better than a shrug. Meanwhile it is August, the pool sits in full sun, and every pool on the route is down some water.
The bucket test settles it, but only if you know what number it has to beat. That is the part most answers skip: the expected evaporation for your conditions, the chemistry in the pool's own history that usually flags a leak weeks before anyone floats a bucket, and the autofill that can hide the whole thing for a season.
At a glance
Key takeaways
- Run the bucket test with the bucket on the pool step, the pump on its normal schedule, and the autofill off - most inconclusive tests are a control that got skipped.
- Know your baseline before you test: typical summer evaporation is about a quarter inch a day, or roughly 1.5-2 inches a week, and wind, heat, and water features push it higher.
- Compare the gap between the two marks, not the total drop. A gap under an eighth of an inch proves nothing; a half inch over 48 hours is a leak.
- Watch for stabilizer falling while calcium hardness climbs - that pairing is the chemical fingerprint of constant refilling, and it shows up weeks before the water level does.
- An autofill hides a leak completely, so on those pools use the water bill and the chemistry instead of the level, and shut the autofill off to test.
- Translate the loss into gallons before you talk to the customer: an inch off a typical 400-square-foot pool is about 250 gallons.
- Report what you measured and route the diagnosis to a licensed leak detection company - your job is the finding and the record, not the repair.
How do I tell if a pool is leaking or just evaporating?
Run a bucket test, which measures the pool against itself: the bucket loses water only to evaporation, the pool loses water to evaporation plus any leak, so the difference between them is the leak. It costs five minutes on a stop and a follow-up reading on the next visit, and it converts an argument into a number you can hand the customer.
Run it in this order. The controls matter more than the test does - almost every inconclusive bucket test comes from something on this list being skipped:
- Fill a clean 5-gallon bucket with pool water and set it on the second step, so it sits partly submerged and stays at the same temperature as the pool. A bucket on the deck bakes in the sun and evaporates faster than the pool, which fakes a leak.
- Mark both levels on the inside wall of the bucket: the water level inside, and the pool's level outside. Tape or a grease pencil both work. Two marks on one surface is what makes the comparison readable at a glance next visit.
- Leave the pump on its normal schedule and change nothing else. The test has to run under the same conditions the pool normally lives in, or the result describes a pool that does not exist.
- Shut off the autofill and tell the customer not to swim, backwash, or drain during the test. An autofill quietly topping the pool up mid-test is the single most common way this comes back inconclusive.
- Read it at 24 hours and confirm at 48. Rain voids the test - start over. A gap under about an eighth of an inch is inside the error of the test and proves nothing; a half inch or more over 48 hours is a leak.
- Compare the gaps, not the totals. Both levels dropping together is evaporation, however alarming the total looks. The pool dropping noticeably below the bucket is a leak, and the size of the gap is roughly the size of the leak.
Know what normal evaporation looks like before you test
A pool losing water is not news; a pool losing more than its conditions explain is. Typical summer evaporation runs about a quarter inch a day, or roughly 1.5-2 inches a week, and heat, wind, low humidity, and an uncovered surface all push that up. Without that baseline in your head, every pool in July looks like it is leaking.
Scale matters as much as the rate. An inch of water off a typical 400-square-foot pool is about 250 gallons, so a pool running 2 inches a week over baseline is losing roughly 500 gallons a week - the number worth saying out loud to a customer who thinks a leak is a slow drip. Treat the figures below as typical ranges for planning, not as a specification; your own route will settle into its own numbers within a season.
| Conditions | Typical loss per week | What it means on the route |
|---|---|---|
| Mild, shaded, or kept covered | About 0.5-1 in | A cover cuts evaporation sharply; a covered pool losing 2 in a week is a real signal |
| Typical warm summer week | About 1.5-2 in | The default baseline for most pools on a Sunbelt route |
| Hot, dry, full sun (July-August) | About 2-2.5 in | Desert peak; a hose every visit here is often just the season |
| Windy or fully exposed | Add about 0.5-1 in | Wind is the most under-rated driver; an exposed corner lot loses more than the pool next door |
| Heated pool, spa, or water feature running | Add about 1-2 in | A running waterfall can double the loss on its own - ask before you diagnose |
- Compare the pool to its neighbors on the same route and the same day, not to a national average. Two pools four streets apart in the same heat are the cleanest control you will ever get.
- A pool that only started needing the hose after a water feature was installed or a cover stopped being used is explaining itself.
The chemical log usually says it first
The readings you already take flag a leak weeks before anyone floats a bucket, because evaporation and leaking leave opposite chemical signatures. Evaporation removes pure water and leaves every dissolved mineral behind, so calcium hardness and salt climb while cyanuric acid holds steady. A leak carries dissolved solids out with the water, and the fresh fill replacing it contains no stabilizer at all - so cyanuric acid drifts down while hardness tracks toward whatever your fill water runs.
Calcium hardness climbing is the expected half, and why calcium hardness climbs in refilled pools is the same mechanism read from the other end. The tell is the pairing: hardness up while stabilizer slides is the fingerprint of a pool being topped off far more than the weather explains. Hardness up while stabilizer holds is just a hot month.
Here is what that looks like on a real route. A tech running 48 pools across Peoria and Surprise, Arizona has needed the hose at one Peoria pool every visit for three weeks. Local August evaporation runs about a quarter inch a day, so roughly 1.75 inches a week is expected - this pool is down closer to 4. The log had already said so: calcium hardness climbed from 310 to 425 ppm over six weeks on fill water that runs hard, while cyanuric acid drifted down from 60 to 41 ppm instead of holding. That is dilution, not evaporation. The bucket test confirmed it in 24 hours, and the customer got a documented finding instead of a guess.
None of that works from a single visit - it needs the same readings taken the same way over a couple of months, which is the ordinary argument for logging chemistry rather than remembering it. Any system where you can log readings at every visit and see the history per pool will show you this pattern; in PoolBoss the readings sit on the pool record so you can read a pool's history across visits rather than paging through one visit at a time.
Two honest caveats, because this is a signal and not a proof. Cyanuric acid degrades on its own over a season, so a slow drift is weak evidence by itself - what you want is a slide that is faster than the rest of the route is seeing. And heavy rain, a big splash-out week, or a partial drain dilutes stabilizer the same way a leak does. The chemistry tells you where to point the bucket; the bucket is what settles it.
An autofill can hide a leak for months
When an autofill is doing its job, the water level tells you nothing at all - the pool sits at the tile line every week while a leak runs continuously underneath. This is the case where a technician sees no symptom for an entire season, and the first sign is the customer's water bill or a chemical reading that stopped making sense.
The numbers are worth stating plainly, because customers underestimate them. A leak of roughly a quarter inch a day on a typical 400-square-foot pool is about 62 gallons a day, near 1,900 gallons a month, on top of normal evaporation the autofill was already covering. On a metered supply that is a visible jump in the bill, and it is usually the first thing that gets anyone's attention.
With the level pinned, you are down to three signals. Chemistry is the strongest: stabilizer sliding while hardness climbs, on a pool nobody drained. The water bill is the bluntest, and worth asking about directly rather than hinting at. The third is the autofill valve itself - if it runs far more often than the pools around it, or the customer mentions hearing it at night, that is the same information the level would have given you.
To bucket-test a pool on an autofill, shut the autofill off for the 24-48 hours of the test, and make sure the customer knows it is off so nobody turns it back on halfway through. Turn it on again when you read the result. A test run with the autofill live measures nothing.
What to tell the customer, and where your responsibility ends
Hand over what you measured, not a diagnosis. You are qualified to say the pool is losing water beyond what evaporation explains, how much, and over what period. You are not - unless you are separately licensed for it - the person who determines whether that is a cracked skimmer throat, a failed light conduit, or a plumbing break under the deck, and saying so in the customer's driveway is how a service call becomes a liability.
A finding that lands well has four parts: the expected evaporation for the conditions, what the pool actually lost, the bucket test result, and the chemistry trend backing it. Something like - your pool is losing about 4 inches a week where the pools around it lose 1.75, the bucket test showed the pool dropping half an inch below the bucket in 24 hours, and your stabilizer has fallen from 60 to 41 ppm since June while hardness has climbed. That is a documented finding. It is also, unlike a hunch, something a leak detection company can start from.
Write it where it stays. Put the numbers and the bucket test result in the visit notes with photos of the water line and the marked bucket, so the record exists on the pool rather than in a text thread - the same reason the readings worth logging at every stop belong on the visit rather than in a tech's memory. If the repair is work you actually do, a quote off that visit turns the finding into a job; if it is not, the notes are what the specialist reads.
Then route it out and say so plainly. Recommend a leak detection company, tell the customer what you will keep doing in the meantime, and be specific about the water: a pool being topped off constantly will keep pulling stabilizer down and hardness up, so expect the chemistry to need correcting again after the repair. Charging for diagnostic time is a business decision, but do it as a named line item rather than burying it - customers accept paying for an answer far more readily than they accept a surprise.
FAQ
Frequently asked questions
How do I tell a leak from splash-out or backwashing?
Account for both before you call anything a leak, because a busy pool loses real water without anything being wrong. A pool with kids in it every afternoon can lose an inch a week to splash-out alone, and a single backwash on a sand filter runs 250-500 gallons, which is roughly an inch to two inches off a typical residential pool in one shot. Ask what the pool's week actually looked like - a pool party, a filter cleaning, a partial drain to lower stabilizer - and check whether the loss tracks those events or runs steadily whether anyone swims or not. That steadiness is the real distinction. Splash-out is spiky and matches the calendar; a leak is boring and constant, including the weeks nobody touches the pool. The bucket test settles it either way, as long as nobody swims or backwashes while it runs.
Should I run the bucket test with the pump on or off?
Run it first with the pump on its normal schedule, because that is the condition the pool actually lives in and it is the result you can act on. If that test shows a leak, running it a second time with the pump off is genuinely useful, and it is the one variation worth the extra visit. A pool that loses water at the same rate whether the pump runs or not usually points to the shell, the skimmer, or a fitting - somewhere water escapes under simple standing pressure. A pool that loses noticeably more with the pump running points at the pressure side of the plumbing, where the return lines are under load. A pool that loses more with the pump off can point at the suction side. None of that tells you exactly where the break is, and it is not your job to say. But it is a genuinely useful thing to hand a leak detection company, because it narrows where they start and it can cut what the customer pays for the diagnosis.
What do I do if the pool only loses water when the pump is running?
Document the difference and pass it on - that pattern points at the pressure side of the plumbing, and it is a specialist's job from there. Run the bucket test twice, once with the pump on its normal schedule and once with it off for the same 24 hours, and write down both numbers. A pool that drops half an inch a day running and barely moves with the pump off is telling you the leak opens up under pressure, which typically means return lines, fittings, or something downstream of the filter rather than the shell itself. In the meantime you have a real decision to make with the customer about run time. Cutting the pump schedule back slows the water loss, but it also cuts filtration and circulation, and in summer heat that trades a water problem for an algae problem within a week. Usually the right call is to keep circulation adequate, keep topping off, and get the repair scheduled rather than nursing it for a month.
How do I document water loss so the customer believes me?
Show the comparison, not the conclusion. Three things make this convincing: a photo of the bucket with both marks visible and the date, the pool's own reading history showing the trend, and what comparable pools on your route lost in the same week. That last one does the most work, because it removes the weather as an explanation - if the four pools around it lost 1.75 inches and this one lost 4, the heat is not the difference. Put all of it in the visit record rather than a text thread, so it is still there in three months when a repair is being argued about. Keep the language to what you measured. Your pool lost 4 inches this week where similar pools lost under 2, and the bucket test showed half an inch of difference in 24 hours is a statement you can defend. You have a cracked return line is not, and it is the kind of guess that gets quoted back at you when the actual repair turns out to be something else.
Do I charge for the time spent diagnosing a leak?
Most operators fold the bucket test itself into the regular visit and charge for anything beyond it, which is a reasonable line. Setting a bucket on the step and reading it next week costs a few minutes on a stop you were already making, and eating that builds a lot of goodwill for very little. A dedicated return trip, a second test with the pump off, or an hour spent checking equipment pads is different - that is time off the route, and it should be a named line item at your hourly rate rather than absorbed. Say so before you do it, not after. The framing that works is simple: the quick check is part of your service, and a real investigation is billable work with a number attached. Customers rarely argue with that. What they do argue with is a surprise charge for something they assumed was included, so the conversation is worth more than the money.
Can constant refilling throw off my chemical dosing?
Yes, and it is one of the more expensive side effects of an undiagnosed leak. Every gallon of fill water dilutes stabilizer, alkalinity, and salt while adding whatever hardness and metals your local supply carries. On a pool losing several inches a week, that means chlorine burns off faster as cyanuric acid falls, salt cells start reading low and underproducing, and calcium hardness climbs steadily even though nobody added anything. You end up dosing more chemical for a pool that will not hold a reading, which is money going into the ground alongside the water. Treat the readings as real and correct them, but expect them to keep drifting until the leak is repaired rather than assuming a dose failed. After the repair the pool stabilizes, and it is worth a full retest at that point - the chemistry you were chasing for months was a symptom, and the balance you settle on afterward is the one that will actually hold.
When should I tell a customer to call a leak detection company?
As soon as a bucket test confirms a gap you can measure, rather than after months of watching it. The practical threshold most operators use is a pool losing roughly a half inch a day more than comparable pools, or any confirmed bucket-test gap that repeats on a second test - at that point you have evidence, and more waiting only adds to the water bill. Refer it earlier than that if the loss is accelerating, if the deck or the equipment pad is staying wet, or if the pool is on an autofill and the water bill has jumped, since that combination can run for a long time unnoticed. Make the referral cleanly. Hand over your numbers, say plainly that finding the break is outside what you do, and keep servicing the pool normally in the meantime. Operators who refer early rather than experimenting for a season tend to keep the account through the repair and the recovery afterward.
