What PoolBoss Says
Draw the sample elbow-deep, at least 3 feet from a return line and away from the skimmer, before you add anything to the pool. Test free chlorine first because it changes fastest, then pH, alkalinity, and the slower readings. Use reagents that are inside their date, and write the numbers down at the pool.
Almost every guide to testing pool water is written for somebody with one pool in their own backyard, where the sample gets drawn wherever the cup happens to reach and it does not matter much. On a route it matters a great deal. You are producing forty numbers a day that land in forty different records, and those records are what you use later to explain a green pool, defend a chemical bill, or decide whether an account is worth keeping.
The kit gets most of the attention in this argument and it deserves the least of it. Two techs carrying identical kits will hand you different numbers on the same pool if one of them is dipping at the surface beside a return. What follows is the procedure itself: where the sample comes from, what order the tests go in, how old your reagents are allowed to get, and why the number has to be written down while you are still standing at the pool.
At a glance
Key takeaways
- Draw the sample before you brush, empty baskets, or add anything, because everything you do to the pool first changes what you are about to measure.
- Sample elbow-deep, around 18 inches down, and at least 3 feet from any return fitting or the skimmer throat.
- Test free chlorine first: it degrades fastest in a warm bottle, so testing it last is what produces low readings on perfectly healthy pools.
- Fix one test order for the whole crew, or two techs' numbers on the same pool are not comparable to each other.
- Date every reagent bottle on arrival and replace on that date, because expired reagents drift quietly rather than failing visibly.
- Keep kits in the cab rather than an uncovered truck bed, and budget roughly $150-$250 per truck per year for reagents and strips.
- Write the readings down at the pool, because numbers reconstructed at the end of the day come out smoother than the water ever was.
How should I test pool water on a service route?
The whole procedure is six moves and it takes about 90 seconds once it is habit. It is worth being rigid about, because every step exists to remove a source of variation that would otherwise show up in the pool's history as a change in the water that never happened.
- Test before you touch anything. Draw the sample as your first act at the pool, before you brush, before you empty baskets, and before you add a single chemical. Anything you do to the pool first changes what you are about to measure, and a sample drawn after a shot of acid tells you about the acid rather than about the water.
- Go elbow-deep, not surface. Push the bottle in upside down to about 18 inches, then turn it upright to fill. Surface water carries whatever has blown onto the pool and sits warmer and more chemically weathered than the body of the water below it. The top few inches are not the pool, and on a hot afternoon they can read noticeably differently.
- Get away from the returns and the skimmer. Stand at least 3 feet from any return fitting, and sample from the deep end or the middle of a long wall rather than at the skimmer throat. Water leaving a return is freshly circulated and, on a pool with an in-floor system, has just been mixed with whatever the equipment last pushed through it.
- Test free chlorine first. Chlorine is the most time-sensitive reading on the list, because it degrades in a warm sample bottle in direct sun. Testing it last is what produces mysteriously low chlorine on healthy pools. Run it immediately, then pH, then total alkalinity, then the slower readings that do not move on a 90-second timescale.
- Check the date on what you are testing with. Reagents and strips both drift as they age, and neither one fails in a way you can see. If the bottle is undated or past its date, the number it produces is not a reading. Replacing it costs less than a single wasted return trip to a pool that was fine all along.
- Write the numbers down before you move. Log the readings at the pool, while the bottle is still in your hand, rather than reconstructing them at the next stop or at the end of the day. This is the step most often skipped, and the one that does the most damage to a record you will need in three months.
Where you draw the sample changes the reading more than which kit you use
Sampling location is the largest controllable source of error in route testing, and it is the one almost nobody writes down. A pool is not chemically uniform. The top few inches are weathered by sun and wind, the water within a few feet of a return has just come through the equipment, and the skimmer throat is collecting whatever happens to be floating. Draw from any of those three places and you have measured something that is not the pool.
An operator running 175 pools across Chandler and Gilbert had one tech whose pH readings ran consistently about 0.2 units high on the same pools his colleague serviced. Same kit, same reagents, same week. The difference was that one of them was reaching in at the surface beside the return on pools with in-floor cleaning systems, so he was sampling water that had just been pushed past the fitting rather than water that represented the pool. Moving the sample to elbow depth at the deep-end corner closed the gap inside a week, and a laminated one-page procedure in each truck kept it closed.
What makes that error expensive is that it is invisible. A 0.2 pH difference is enough to move an acid dose on a 20,000 gallon pool by a real fraction of a bottle, and a free chlorine reading taken beside a return on a pool that was dosed the previous visit can read a full point high. Neither one looks like a mistake. Both produce entirely plausible numbers, which is exactly why they survive for months. None of this makes the equipment irrelevant, and choosing what each truck carries is a real decision with real consequences. It is just the second-order one. Get the sample right first.
Test in an order, and always the same order
Test order is not arbitrary. Some readings decay inside the sample, some are only interpretable next to another reading, and some are stable enough to leave until last or skip entirely this week. Fixing one order across the whole crew removes another source of variation and is what makes two techs' numbers on the same pool comparable.
| Reading | When to test | Why it sits there |
|---|---|---|
| Free chlorine | First, within a minute of drawing | Degrades fastest in a warm sample, especially in direct sun |
| Combined chlorine | Immediately after free chlorine | Comes off the same sample, and the gap between the two is what you act on |
| pH | Second | Stable for a few minutes, and needed before alkalinity means anything |
| Total alkalinity | Third | Read next to pH, because each one explains the other |
| Calcium hardness | Monthly or quarterly | Moves slowly, so weekly testing adds no information |
| Cyanuric acid | Monthly, or after adding stabilizer | Changes only when stabilizer goes in or water is replaced |
| Salt | Monthly on salt pools | Drifts slowly; test seasonally or when the cell reports low |
| Water temperature | Anywhere in the sequence | Does not decay, and it explains the behaviour of everything above |
Reagents and strips have a date, and heat shortens it
Reagents are a consumable with a shelf life, and a service truck in a Sunbelt summer is a poor place to keep them. Liquid reagents are generally good for about a year unopened and meaningfully less once in use. Strips degrade faster still once the bottle has been opened, commonly within 3-6 months in heat and humidity, and the desiccant packet in the bottle is doing real work rather than being packaging.
The failure mode is what makes this worth an actual rule. Expired reagents do not stop working in a way you would notice. They drift, which is worse, because they keep producing numbers that look completely normal while every reading on the route moves quietly in one direction. A month on a drifted chlorine reagent looks exactly like a month of pools trending down, and the natural response, which is to dose harder, is the wrong one.
So write the date on every bottle the day it arrives rather than relying on a lot code you cannot decode at the pool, and replace on that date instead of waiting for a reading to look wrong. Keep the kit in the cab rather than an uncovered bed, because a closed truck bed in Phoenix in July runs well above the storage range printed on most reagent labels. Budget roughly $150-$250 per truck per year for reagents and strips together, which is less than the chemicals burned on one green-pool recovery caused by a chlorine target that was wrong all month.
Log at the pool, not from the seat of the truck
Readings written down three stops later are the largest source of bad history in a service business, and nobody codes it as an error because the numbers still get recorded. Memory smooths. A tech who has tested eleven pools before writing anything down is not recalling eleven distinct sets of readings; he is recalling a general impression and reconstructing plausible numbers around it. The record that comes out the other end is smoother and tidier than the water ever was.
This is the part a phone in a pocket genuinely solves. Entering the readings on the spot means the number that lands in the record is the number that came off the test, and a per-pool history built that way is worth something months later. In PoolBoss the tech records the visit on the mobile app while standing at the pool, so free chlorine, combined chlorine, pH, total alkalinity, calcium hardness, cyanuric acid, salt, and temperature all land on that pool's own record as they are taken. Being able to log the readings at the pool while you are standing there is the difference between a history you can read and one you are guessing at.
Which readings belong in the record at every stop, and which ones are monthly, is a separate decision worth making deliberately rather than by habit, and which readings belong at every visit works through it. What a good sample buys you is that any of those numbers landing outside its band is a real signal rather than noise, so acting on a reading that is out of range starts from something you can trust.
The payoff shows up the day somebody questions the service. A customer who says the pool has been cloudy for a month gets answered with a month of dated readings rather than a recollection, and a tech new to the route inherits numbers he has reason to believe. That is also why the procedure belongs on paper rather than in the head of whoever has been doing it longest: the standard you hand a new technician is what keeps sampling consistent after the person who set it up has moved on.
FAQ
Frequently asked questions
Can I test the water right after adding chemicals?
No, and this is the most common way a good procedure gets undone. Test first, dose second, and if you need to verify a dose, do it on the next visit rather than a few minutes later. Most residential pools turn over in about 6-8 hours on a modern variable-speed schedule, so a same-visit retest is not measuring the dose, it is measuring the plume. Liquid chlorine and acid are the worst offenders because they are dense and sink, so a sample drawn near the deep end shortly after dosing can read wildly high while the rest of the pool has barely changed. If you genuinely need a post-dose number, brush to distribute first, wait as long as the stop allows, and mark the result as a spot check rather than the visit's reading. Otherwise trust the arithmetic of the dose and confirm it next week, which is exactly what a per-pool history is for.
Does it matter what I carry the water sample in?
More than most techs expect, because a sample bottle is a small volume with a lot of surface area. Rinse the bottle two or three times with the pool's own water before you fill it, every time, at every pool. A bottle carrying a film from the last pool, or a trace of reagent from a previous test, will move a reading measurably. Use a dedicated container rather than a cup that has held anything else, keep it capped between stops, and replace it once it is scratched or clouded, because a rough interior holds residue that rinsing does not clear. Never use a bottle that has held chlorine or acid concentrate, since the residue alone is enough to invalidate a chlorine test. If your kit came with a vial, that vial is the sample container, and the habit worth building is rinsing it in the pool at the moment you draw rather than back at the truck.
Can I collect the sample and test it back at the shop?
You can, but only for the slow readings, and you will get poor numbers on the ones that matter most. Free chlorine and pH both start moving the moment the sample leaves the pool: chlorine degrades in warmth and sunlight, and pH shifts as dissolved carbon dioxide comes out of solution. A bottle that has ridden in a hot cab for an hour is no longer the pool you sampled. If you have no choice, because the photometer lives at the shop or a sample is going to a lab, fill the bottle to the very top with no air gap, cap it, keep it dark and cool, and test within a couple of hours. Even then, treat chlorine and pH as indicative rather than exact. Cyanuric acid, calcium hardness, and salt are far more forgiving over a few hours, so those are the reasonable candidates for testing away from the pool.
Where should I take the sample on a pool with an attached spa?
Sample the pool and the spa separately, and take the pool sample with the spillover off where the system allows it. A raised spa that spills into the pool is effectively a second body of water with its own temperature, its own bather load, and often its own chemistry, and the water crossing the spillway belongs to neither. Draw the pool sample at the deep end, well away from the spillway, at your normal depth. If the spa is genuinely in use, it gets its own reading, because a spa held at 102 degrees consumes chlorine several times faster than the pool beneath it and a single blended number hides that completely. On shared-equipment systems where valves rotate between pool and spa, note which mode the system was in when you sampled: a pool that has been running in spa mode has not been circulating, and the sample will show it.
What should I do when a reading looks impossible?
Retest before you dose, using a fresh sample rather than the one already in the vial. A number that contradicts the pool in front of you, such as a clear pool reading zero chlorine or an alkalinity that has supposedly doubled in a week, is far more likely to be a testing problem than a water problem, and dosing against it can do real damage. Work the three usual causes in order: a bad sample, an expired or contaminated reagent, and a reading that has gone off the top of the test's range. Very high chlorine is the classic third case, because colour-based tests bleach out above roughly 10 ppm and start reading low again, so a heavily shocked pool can look as though it has almost none. If the retest agrees and the pool still does not match the number, record it, flag it to the office, and check it against that pool's own history before you act.
How do I train a new technician to test consistently?
Put the procedure on one laminated page in the truck, then verify it in the water rather than assuming it took. Written steps get read once and drifted from within a month, so the check that actually works is having the new tech and an experienced tech test the same pool from their own separate samples and compare on the spot. Do it on three or four pools in the first week, and again after a month. A gap of more than about 0.2 in pH or 0.5 ppm in free chlorine points at technique, and the cause is almost always sampling depth or distance from a return rather than anything to do with reading a colour. Watch them draw the sample rather than watching them read the result, because that is where the error gets introduced. Five minutes a month cross-checking two techs on one pool catches drift long before it becomes a history nobody can interpret.
