Pet water fountains: what to look for, what to ignore, and what nobody can prove

Almost every pet fountain sold anywhere is sold on one promise: that it will make your animal drink more. We sell one, and we are not going to make that promise, because three separate pieces of published work have gone looking for the effect at a group level and have not found it.

That is the first thing this guide has to say, and it comes first on purpose. Everything else here is about the things a fountain demonstrably does, which are not nothing: it holds a larger volume of water than a bowl, it moves that water, it covers it, and it filters out hair and grit. Those are real. Whether they change the number of millilitres your cat takes in is a different question, and the honest answer to it is that nobody has shown that they do.

The claim we are not going to make

Here is the evidence, such as it is. It is thin, and it points one way.

Pachel and Neilson, 2010. A pilot study in the Journal of Veterinary Behavior, nine individually housed cats, each measured over four consecutive 22 hour periods on a still source and a flowing source. The abstract and citation are on ScienceDirect. Mean consumption was 115.44 mL from the flowing source and 109.83 mL from the still one. The difference was not statistically significant. The authors note that individual cats appeared to prefer one source strongly, but that "a group preference was not established," and they caution against over-interpreting the result.

Robbins and colleagues, 2019. A larger and better controlled study in the Journal of Feline Medicine and Surgery, volume 21 issue 8, pages 682 to 690, and the full record is on the publisher's site. Sixteen cats enrolled, fourteen completed, randomised crossover, each cat its own control, two weeks on each of three presentations: still, circulating, and free falling. Daily intake came out at roughly 27.5, 26.3 and 26.1 mL per kg per day, with P equal to 0.942. Urine specific gravity was 1.050, 1.051 and 1.051, with P equal to 0.595. Three individual cats did show a significant preference for one bowl. The authors conclude that bowl type had no appreciable effect on water intake, and suggest that strategies other than special water bowls be used when a patient would benefit from drinking more.

Kosmal and colleagues, 2025. A scoping review in the Journal of Animal Science, covering 32 publications from 1975 to 2025, free in PubMed Central. Its summary of this exact question is one sentence: "The water intake of cats was not affected by fat content or water source type (static or flowing)." In the body it adds that "cats did not appear to prefer moving or flowing water sources," that a few individual cats showed a preference for moving water while intake was unaffected, and that the common belief about moving water "is not well supported by current scientific research."

Now the caveats that work in the other direction, because they are real too. These are small studies. Nine cats and fourteen cats are small numbers, mostly laboratory housed, and the review notes that across all 32 papers sample sizes ran from 2 to 55 animals and that intake was "highly variable within and across studies." A study that finds no difference in a small sample has not proved there is no difference. And in both of the primary studies, some individual cats clearly did prefer one presentation.

So the position we take is this. If your cat turns out to be one of the individuals who prefers moving water, that is a real thing about your cat and a fountain is a reasonable buy. What we will not do is sell you the group effect, because the group effect is not there in the published record.

What the guidelines do support, which is placement

There is one genuinely good, genuinely free reference on why cats under-drink, and it is not about fountains at all. The 2013 AAFP and ISFM Feline Environmental Needs Guidelines are published in full by the American Association of Feline Practitioners.

The second of their five pillars is "provide multiple and separated key environmental resources: food, water, toileting areas, scratching areas," plus play and resting areas. The guidelines are specific about what separated means. "Each key resource should be placed in its own location, separate from other resources." "Food and water resources should be separated from each other." "Toileting areas should always be located away from food and water dishes." On choice, they recommend "a minimum of two resting areas, two feeding areas and two toileting areas." On multi-cat homes: "expanding the habitat prevents cats in a multi-cat household from eating and drinking in close proximity to one another."

The single most common water setup in a house is a bowl sitting next to the food bowl. The guidelines say to separate them. That change costs nothing, has published guidance behind it, and is the thing we would do before buying any fountain at all.

The five decisions that are actually yours

Capacity, against household size

Ours holds 3 litres, which the supplier also writes as 105 oz. For one or two cats that is generous. For a large dog or four animals it is not, and you will be refilling often enough to wonder why you did not buy a plain bowl and a jug.

Bigger is not automatically better, and there is a measurement behind that. A 2021 study in Vet World sampled the drinking water of 55 cats in Thai households and found fecal coliform contamination in 67.27 percent of them, with a free full text in PubMed Central. Water that had sat for longer than 12 hours was worse: among tap water samples, 84.21 percent of those left over 12 hours were positive. The authors' recommendation is plain: "drinking water for cats should be changed every 12 h, especially for households using tap water." A reservoir you only empty weekly is not an advantage. Size the capacity to what gets changed, not to what fits.

Power, and what cordless costs you

Ours is cordless and rechargeable. The gain is placement freedom and no cable on the floor. The cost is a recharge cycle and a battery that is the part most likely to end the product's life. And on our unit there is a hole in the data: the supplier does not publish a runtime figure, so we cannot tell you how long a charge lasts. That is the single number we would most like to have and do not. A mains fountain tells you nothing about its battery because it does not have one, which in a household with a reliable socket in the right place is a genuine advantage.

How hard it is to clean

This is the decision that determines whether the thing is still in use in a year. A fountain is a pump, a housing, an impeller, a filter seat and a reservoir. A bowl is a bowl. Every additional part is a surface that biofilm can grow on, and biofilm is just what the CDC describes as germs living together in a group that "makes a slimy, glue-like substance" to stick to surfaces. Before you buy, look at how many parts come apart, and whether the pump opens without a tool.

How loud the pump is

A pump in a quiet kitchen at two in the morning is a real consideration, and it is the complaint that most often ends in a fountain in a cupboard. We have not measured ours. See below.

What the body is made of

Ours is plastic. The supplier's product title contains the words "304 stainless steel" and their own specification panel says the material is ABS and PC, which are plastics, and their material field says plastic. Two of their three answers say plastic, so plastic is what we sell it as. If you came to this category because you want a metal vessel, buy a metal one. We will not dress ours up.

The filter, and the running cost nobody quotes

Ours takes a multi layer cartridge. The replacement interval is not published and neither is the price of a spare, which means you are buying into a running cost of unknown size. We say that on the product page and we will say it here.

On what a filter does: the EPA's consumer sheet on home filtration is the clearest free source, and its key line is that "an activated carbon filter, by itself, is not designed to remove all disease-causing organisms". It also notes that carbon filters are rated for a set volume and should be replaced on schedule. A foam layer catches hair and grit. Carbon affects taste and odour. Neither is water treatment and neither is a substitute for taking the thing apart and washing it.

What we have not tested

We have not run a battery runtime test, a noise measurement or a cleaning teardown on this unit. We have not had one in our hands. When we do, the tests we intend to run are: charge to full, run in sensor mode in a normal kitchen, and record hours to cutoff; measure sound pressure at one metre at full and low reservoir level; and time a full disassembly and reassembly, listing every part that needs a brush. Those numbers get published here with the method, including if they are bad.

The short version

Separate the water from the food. Change it more often than you currently do. Buy the capacity that matches how often you will actually empty it, not the biggest number. Decide about cordless on whether you have a socket where you want the water, not on the marketing. Count the parts before you buy, because you will be washing them. And weigh the evidence honestly: at a group level, moving water has not been shown to change how much cats drink, so buy a fountain for the volume, the cover and the filtration, or for a specific cat you have watched prefer a running tap.

Our fountain is 3 litres, cordless, plastic, with a radar sensor mode and a timed mode, at $44. If what you want is a vessel that is easy to keep genuinely clean and nothing more, a wide metal bowl changed twice a day is a better buy and costs less.