I've pulled drinking water out of places I had no business trusting. Backcountry streams, a questionable spigot in a country where the tap was a gamble, a rain barrel during a stretch when the power was out and the town water notice said "boil." Every one of those moments taught me the same thing: the time to figure out water is not when you're already thirsty and staring at a source you don't trust.
So this is the guide I wish someone had handed me years ago. Not a spec sheet, not a fear pitch. Just how to think about filtering water so you can make one good decision now instead of a panicked one later. I carry the products I'll mention because they earned it, but the thinking matters more than the brand, and you'll be able to judge any filter on the shelf by the time we're done.
Start with the thing nobody tells you: not all "germs" are the same size
Here's the single idea that makes everything else make sense. Waterborne threats come in three sizes, and they're wildly different. The CDC puts it plainly: parasites are larger than bacteria, which are larger than viruses.
Think of it in scale. Protozoa, the Giardia and Cryptosporidium that ruin backcountry trips and follow floods into tap water, are the giants, roughly 5 to 15 microns across. Bacteria like E. coli and Salmonella are much smaller. Viruses like norovirus and Hepatitis A are smaller still, small enough to slip through gaps that stop everything else.
This matters because a filter is a physical net. It catches anything bigger than its holes and lets anything smaller through. So the whole game is matching the size of the net to the size of the threat you're actually likely to face. Once you see it that way, the marketing noise falls away and you can read any filter honestly.
The number that actually matters: absolute pore size
Every filter has a pore size, measured in microns. But there's a catch that trips up almost everyone, and it's worth slowing down for, because it's the difference between water that's safe and water that just looks safe.
Pore size comes in two flavors: "absolute" and "nominal." The CDC's own guidance spells out why that word matters. An absolute 1-micron filter means every pore is 1 micron or smaller, so nothing bigger gets through. A nominal 1-micron filter means the pores average 1 micron, which means some are bigger. And the CDC is blunt about the consequence: some nominal filters let 20 to 30 percent of 1-micron particles, like Cryptosporidium, pass through.
Read that again. A "1-micron" filter can still leak a third of the parasites you bought it to stop, if it's the nominal kind. This is the fine print that separates real protection from the appearance of it.
So the rule, straight from the CDC and EPA and the one thing to burn into memory: for parasites, you want an absolute pore size of 1 micron or smaller. That's the floor. Anything vaguer than that word "absolute" on the packaging, and you're guessing.
What each level of filtration actually buys you
Now that size makes sense, here's what you're really getting at each tier. I'm pulling these ranges straight from the CDC's technical guidance, because I'd rather you trust the source than trust me.
Microfiltration, around 0.1 micron. This is the sweet spot for most people, most of the time. It has very high effectiveness against protozoa and bacteria. It handles the Giardia, the Crypto, the E. coli, the things that actually make people sick in the field and after disasters. What it does not do is stop viruses, they're too small. The good news, and the CDC says this too, is that in remote areas with little human or animal activity, protozoa and bacteria are the primary threats, so a good microfilter is genuinely enough.
Ultrafiltration, around 0.01 micron. A finer net. Now you're catching some viruses on top of everything else. This is where you go if your water source is downstream of people, farms, or poor sanitation, the situations where viruses become a real concern.
Reverse osmosis and nanofiltration, down to 0.0001 micron. The extreme end. These stop essentially everything, protozoa, bacteria, viruses, even many chemicals. They're household and specialized systems, not something you throw in a pack, but worth knowing they exist when the goal is a home's long-term water security.
The honest takeaway: for a go-bag or a backcountry trip in normal terrain, a 0.1-micron microfilter is the right tool and anything more is weight you don't need. For a questionable water supply near population or agriculture, step up to ultrafiltration or add a second method. Match the tool to the actual threat, not to the scariest one you can imagine.
The virus gap, and the cheap fix for it
Here's where most filter marketing goes quiet, and I won't. A standard microfilter does not remove viruses. The manufacturers who proudly advertise bacterial removal often stay silent on that point, because size is a hard physical limit and there's no way around it with a bigger net.
For most preparedness situations in developed areas, this is not the crisis it sounds like, your bigger risks are the protozoa and bacteria a good filter already handles. But if viruses are a real concern, the answer isn't necessarily a fancier filter. It's a second, complementary method, and it's cheap.
The CDC's recommendation is to filter first, then add a chemical disinfectant to the filtered water. And there's a specific detail here worth knowing, because not all chemicals are equal either. Chlorine dioxide tablets kill Cryptosporidium if you follow the directions, but iodine and plain chlorine tablets will not kill Crypto. So the belt-and-suspenders approach that actually covers all three threats is: a good microfilter for the protozoa and bacteria, plus chlorine dioxide tablets for the viruses the filter misses. Two lightweight tools that together close the whole gap.
This is why I keep chlorine dioxide tablets alongside every filter I own. Not because the filter fails, but because the filter and the tablets fail in different places, and the places they overlap are covered twice.
How to verify a claim instead of trusting it
Anyone can print "removes 99.9%" on a box. Here's how to check whether it means anything, using the same standards the professionals use. First, that word again: absolute, not nominal. If the packaging won't commit to "absolute," assume the worst.
Second, look for NSF/ANSI certification, specifically standards 53 or 58. NSF is an independent organization that develops public health standards for these products, which means a third party, not the manufacturer's own lab, verified the claim. The EPA and CDC point to NSF/ANSI 53 and 58 as the verification standard for parasite and cyst removal. A filter carrying that certification has been checked by someone with no incentive to lie.
Third, understand what "EPA purifier" does and doesn't mean. To use the word "purifier," a product runs testing to show it removes 99.9999% of bacteria, 99.99% of viruses, and 99.9% of protozoan cysts. That's a high bar. But here's the honest footnote the CDC includes: the EPA does not independently test the validity of those claims. So "purifier" is meaningful, but it's still a manufacturer's testing, which is exactly why the independent NSF certification carries more weight.
Put simply: "absolute" tells you the design is right, NSF certification tells you someone independent checked, and "purifier" tells you it targets viruses too. Look for those three things and you can cut through almost any marketing on almost any shelf.
Two failure modes that have nothing to do with the filter
The best filter in the world doesn't help if it dies in the field. Two things kill filters, and both are avoidable.
Freezing. This is the quiet one. A ceramic or hollow-fiber element that freezes can develop microscopic cracks, and those cracks are wide enough to let parasites straight through, on a filter that looks perfectly fine. If a filter has been used and there's any water left inside it, a hard freeze can compromise it invisibly. This is why I don't leave a filter in a cold garage or a winter vehicle without thinking about it. A filter that froze wet is a filter you can no longer trust, and it won't warn you.
Neglect. Filters clog, that's them doing their job. Flow slowing down is normal and usually fixable by cleaning as directed. But a filter you never maintain eventually stops, and a filter you can't maintain in the field is dead weight. This is where the cleanable designs, ceramic elements you can scrub, hollow-fiber cartridges you can backflush, earn their place over throwaway units. A filter you can service is a filter that comes home with you.
So what do I actually carry, and why
I'll tell you plainly, and I'll tell you the reasoning so you can disagree. For the pack, a 0.1-micron hollow-fiber filter in a squeeze or gravity design. Light, fast, absolute pore size that handles the protozoa and bacteria that are the real field threats, and cleanable by shaking or backflushing. For most people this is the whole answer, and I'd rather you own this and actually carry it than own something heavier that stays home.
For the house and the long haul, a ceramic-element filter, the buy-it-once kind. A good ceramic candle is field-cleanable, lasts through enormous volumes of water, and is built to be serviced rather than replaced. It's not the cheapest answer up front. It's the answer that's still working in ten years, which makes it the cheapest answer over time. The Katadyn Pocket I keep for this reason has a documented history of decades of hard use and a warranty that tells you the maker expects it to outlive the buyer.
And alongside both, always, chlorine dioxide tablets. Cheap, weightless, and they cover the viruses the filters don't. Redundancy on the one thing you can't improvise and can't un-drink.
That's the kit. A microfilter for volume, a ceramic filter for permanence, tablets for the gap. It's not the most gear. It's the least gear that actually covers all three threats, which is the only spec that matters.
The one thing to take away
If you remember nothing else: absolute 1 micron or smaller, verified by NSF, plus chlorine dioxide tablets for viruses. That single sentence will keep you from buying wrong, and it'll let you read any filter on any shelf without needing anyone, including me, to tell you if it's good.
Water is the thing you can't do without and can't fake your way around. Handle it once, correctly, and it stops being something you worry about. That's the whole point of doing this now, calmly, instead of later, in a hurry, over a source you don't trust.
We carry a curated set of water filtration built on exactly this reasoning, organized by how you'd actually use it. If you want a hand matching a filter to your situation, that's what we're here for.
This guide draws on public guidance from the CDC, EPA, and NSF/ANSI standards. It's meant to help you make an informed choice, not to replace the instructions that come with any filter you buy. Follow those.