Is a 'tested to NSF standards' filter unsafe?

Not necessarily. It means the reduction claims have not been confirmed by an accredited certifier through ongoing testing and facility audits. The lab numbers may be accurate, but you are relying on the seller's own documentation rather than a public, verifiable certification. Look up the model in a certifier database to see what, if anything, is actually certified.

Where do I check whether a filter is genuinely certified?

Use the certifier's public database. NSF is at info.nsf.org/Certified/DWTU, WQA Gold Seal at find.wqa.org/find-products, and IAPMO R&T at pld.iapmo.org. Search by brand, then confirm the exact model number, the standard, and the specific contaminant all appear together.

Are NSF, WQA, and IAPMO equally valid?

Yes. All three are accredited to certify water treatment units to the same NSF/ANSI performance standards. A product certified by any of them is a real certification, and each maintains its own searchable listing of certified models.

Why would a company avoid certification?

Common stated reasons are cost and the ongoing obligations of an audited program, or a preference to test against more contaminants than a single standard covers. Those can be reasonable choices, but they do not turn a self-arranged test into a certification, and the buyer should know which one they are looking at.

A product is certified to Standard 42. Does that cover lead?

No. Standard 42 covers aesthetic effects like taste, odor, and chlorine. Health-related contaminants such as lead fall under Standard 53. A model can be certified to one standard and make no certified claim under another, which is why you check the specific contaminant in the listing, not just the standard number.

How scoring works

"Tested to NSF Standards" Is Not the Same as "NSF Certified"

'Tested to NSF standards' describes a one-time lab result a brand controls, while 'NSF certified' is an ongoing third-party program you can verify in a public database.

The two phrases look alike and mean different things

Read enough water-filter marketing and you will see two phrases used almost interchangeably: 'tested to NSF standards' and 'NSF certified.' They sound like the same claim dressed in different words. They are not. One is a result a company arranged once; the other is a program a third party runs continuously and publishes.

This distinction is the reason FilterScored exists. A filter can carry impressive contaminant-reduction numbers and still have no certification behind them that anyone can check. Our job is to tell you which claims are verifiable and which ones rest on the manufacturer's own paperwork.

What 'NSF certified' actually involves

NSF/ANSI standards (for example, Standard 42 for taste and odor, Standard 53 for health-related contaminants like lead, and Standard 58 for reverse osmosis) are written specifications. Certification is the process of an accredited certifier confirming that a specific product model meets a specific standard, then keeping that confirmation current.

According to NSF, certification is not a single test. It includes evaluating the product, testing it against the standard's contaminant-reduction and structural requirements, auditing the manufacturing facility, and re-testing products from the market on a recurring basis. The certifier also reviews the product's marketing and labeling so the claims match what was actually certified.

Three accredited bodies certify water treatment units to these same NSF/ANSI standards: NSF itself, the Water Quality Association (WQA, under its Gold Seal program), and IAPMO R&T. A product certified by any of the three is listed in that certifier's public database against named model numbers and named contaminant claims.

What 'tested to' usually means

'Tested to NSF standards,' 'NSF-tested,' 'tested to NSF/ANSI 53,' and similar phrases describe something narrower. They typically mean a lab ran a product against some part of a standard's test method on some occasion. NSF's own guidance is direct about this: phrases like these are not the same as certification and may not indicate independent third-party verification at all.

A 'tested to' result can be perfectly real and still leave open questions a certification answers. Was the whole standard applied, or only the contaminant lines the brand wanted to show? Was the test done at the rated capacity of the filter, or on a fresh cartridge? Is the facility audited? Is the product re-tested over time? Are the marketing claims policed against the data? With a one-time test arranged and published by the seller, you are taking those answers on trust.

There is also a cost and structure angle worth naming plainly. Certification carries fees and ongoing obligations. Some companies decide the independent-lab route is cheaper or lets them test against more contaminants than a standard covers. That can be a defensible business choice. It is still not certification, and a buyer deserves to know which one they are looking at.

How to verify a certification yourself in under five minutes

You do not have to take anyone's word, including ours. Each accredited certifier publishes a searchable listing. For NSF, search the Drinking Water Treatment Units database at info.nsf.org/Certified/DWTU. For WQA Gold Seal products, use find.wqa.org/find-products. For IAPMO R&T, use the Product Listing Directory at pld.iapmo.org.

Search by brand or trade name, then confirm three things. First, the exact model number you intend to buy appears, not just a similar one. Second, the standard listed matches the claim that mattered to you (Standard 53 for lead, for example, not just Standard 42 for taste). Third, the specific contaminant is named in the listing's reduction claims. A product can be certified to Standard 53 for one contaminant and say nothing about another.

If the model is not in the certifier's database, it is not certified to that standard, regardless of what the box, the product page, or a review says. 'Tested to' language on a product that is absent from every certifier's database is the clearest signal that you are looking at a self-arranged test rather than a certification.

Named examples: who leans on 'tested to' and who certifies

Berkey is the most prominent example of the 'tested to' approach. Its own materials state that the Black Berkey elements are tested to applicable NSF/ANSI protocols by independent laboratories, and that the company has not pursued formal NSF/ANSI certification, citing cost and the volume of water a full certification test would require. The lab reports exist; the third-party certification does not.

Clearly Filtered publishes detailed performance data sheets showing reduction percentages for a long list of contaminants, generated by independent labs. This is test data the company commissions and presents. It is not the same as a model appearing in NSF's, WQA's, or IAPMO's certified-products databases, so check those listings directly before treating any specific claim as certified.

LifeStraw and Epic Water Filters are useful contrasts because they hold real certifications on specific models while also publishing broader 'tested to' data. The cleaner way to read either brand is to ignore the broad marketing and look up the exact model: confirm which standards and which contaminants are actually certified in the certifier database, and treat everything else as 'tested to.' The lesson is that a brand having some certified products does not make every claim on every product a certification.

How to read any filter claim from now on

Treat 'certified' and 'tested' as two separate words that you do not let collapse into each other. When a page says a filter 'reduces' or 'removes' something, ask the next question automatically: certified by whom, to which standard, for which contaminant, and is the model in the database?

None of this means a 'tested to' product is bad or that its numbers are fake. It means the verification burden sits on you instead of on an accredited third party. A certified product hands you a public, audited, re-tested record tied to a model number. A 'tested to' product hands you a document the seller prepared. Knowing which one you hold is the whole point.

FAQ

Is a 'tested to NSF standards' filter unsafe?: Not necessarily. It means the reduction claims have not been confirmed by an accredited certifier through ongoing testing and facility audits. The lab numbers may be accurate, but you are relying on the seller's own documentation rather than a public, verifiable certification. Look up the model in a certifier database to see what, if anything, is actually certified.
Where do I check whether a filter is genuinely certified?: Use the certifier's public database. NSF is at info.nsf.org/Certified/DWTU, WQA Gold Seal at find.wqa.org/find-products, and IAPMO R&T at pld.iapmo.org. Search by brand, then confirm the exact model number, the standard, and the specific contaminant all appear together.
Are NSF, WQA, and IAPMO equally valid?: Yes. All three are accredited to certify water treatment units to the same NSF/ANSI performance standards. A product certified by any of them is a real certification, and each maintains its own searchable listing of certified models.
Why would a company avoid certification?: Common stated reasons are cost and the ongoing obligations of an audited program, or a preference to test against more contaminants than a single standard covers. Those can be reasonable choices, but they do not turn a self-arranged test into a certification, and the buyer should know which one they are looking at.
A product is certified to Standard 42. Does that cover lead?: No. Standard 42 covers aesthetic effects like taste, odor, and chlorine. Health-related contaminants such as lead fall under Standard 53. A model can be certified to one standard and make no certified claim under another, which is why you check the specific contaminant in the listing, not just the standard number.

Sources

Browse air purifier scores or water filter scores.