Methodology

Air purifiers for particulate matter — dust, pollen, pet dander, smoke, PM2.5 — all work the same way: a fan pushes air through a filter.

Every purifier we track uses a HEPA or HEPA-equivalent filter. True HEPA (H13+) captures 99.97% of particles at the “most penetrating particle size.” That is, of the air that goes through the purifier, 0.03% of the “hardest to filter” particles make it through. Some products labeled "HEPA-type" may capture less — but since we rank by independently measured clean air delivery rate, filtration efficiency is already reflected in the numbers.

Once you're past that, the only variable that really matters is how much air passes through the filter. That's what CADR (Clean Air Delivery Rate) measures: cubic feet of clean air delivered per minute. 200 CFM cleans a room twice as fast as 100 CFM.

Ionizers, UV-C lights, plasma generators, "nano" coatings, photocatalytic oxidation — these add cost and complexity, sometimes produce ozone, and don't meaningfully improve particulate removal beyond what the HEPA filter already does. We ignore them. What matters is how much air moves through the filter, how quietly, and at what price.

Most purifiers quote their max CADR — at the loudest fan speed. Nobody runs a purifier at max speed in a bedroom. Quiet CADR is the CADR at 41 dBA or below — roughly the ambient level of a quiet room. This is our default ranking metric.

A purifier with 400 CFM max but 80 CFM quiet is far less useful for daily living than one with 250 CFM max and 180 CFM quiet. The first one sounds great on paper; the second one actually cleans your air.

How we compute it

Most purifiers don't publish CADR at specific noise levels — they only publish max CADR and sometimes a noise range. So we estimate quiet CADR from test data where available.

Where independent reviewers have measured noise (dBA) and power draw (watts) at each discrete fan speed, we align those measurements into a noise-power curve. From that curve:

1. Find the highest fan speed where noise is ≤41 dBA.
2. Take the power draw at that speed and divide by the power draw at max speed to get a power ratio.
3. Apply the formula:

quiet CADR = max CADR × (quiet power / max power)^0.344

Products whose lowest fan speed exceeds 41 dBA are excluded from quiet rankings entirely — they simply can't run quietly.

When per-speed data isn't available

Not every purifier has detailed per-speed measurements from independent reviewers. For some, we only have the noise and power at the minimum and maximum fan speeds — no intermediate points.

In those cases we use the minimum-speed power draw as a conservative estimate for quiet power. This is the lowest speed, not necessarily the highest speed at or below 41 dBA — so it underestimates quiet CADR. These purifiers may perform better in practice than our numbers suggest.

For a small number of purifiers we have no per-speed data at all and cannot estimate quiet CADR. These are excluded from quiet rankings but still appear when you search by max CADR.

Data collection process

For each purifier, we compile data from multiple sources, including AHAM-verified ratings, independent lab tests, manufacturers, and reviewers like HouseFresh and AirPurifierFirst. Where sources disagree, we prefer independently measured values over manufacturer claims.

We then align the per-speed measurements into a noise-power-airflow model for each purifier, identify the quiet operating point, and apply the formula above. The result is a comparable quiet CADR across all purifiers in the database — even though the actual fan speed it corresponds to varies by model.

This tool recommends purifiers for particulate matter — dust, pollen, pet dander, smoke, PM2.5. A HEPA filter handles all of these.

Every purifier we recommend includes some carbon filtration — typically a thin pre-filter layer. This helps with light VOCs and everyday odors, and is fine for most people.

But volatile organic compounds (VOCs), formaldehyde, and chemical off-gassing at higher concentrations are a different problem. Effective VOC removal requires 2–15+ lbs of granular activated carbon — the thin carbon sheets in most consumer purifiers saturate within weeks. A thick carbon bed also creates airflow resistance, which means lower CADR for the same fan, and louder units.

For serious chemical concerns, our recommendation is a HEPA purifier for particles (what this tool finds) plus a dedicated, carbon-centric purifier for VOCs. The right carbon media depends on your specific pollutants — some concerns need specialty carbon formulations. See our VOC recommendations.

The homepage starts with defaults that reflect how most people actually use purifiers — in living spaces where noise matters, with filters they won't replace every month:

• Quiet mode on — ranked by quiet CADR (at 41 dBA or below), not max. Toggle "loud okay" per room if noise doesn't matter.
• Premium design — only well-designed models by default. Turn off to see all tiers including basic and budget.
• 5 Air Changes per Hour (ACH) — this is how many times the purifier will go through all the air in the room, assuming sufficient air movement. Switch to 2.5 ACH (Basic) or 8 ACH (Cooking / Wildfires) under advanced options.
• 12+ month filters — excludes short-lived filters. Adjustable. (All filter lifespans are averages — actual life depends on usage and air quality.)

All of these are adjustable. The algorithm works the same regardless — only the candidate pool changes. For more on why we chose these defaults, see what we focus on.

We calculate required CADR from room volume and air changes per hour (ACH):

The three air quality goals:

• Basic — 2.5 ACH. Light dust and pollen.
• Recommended — 5 ACH. Matches the AHAM two-thirds rule. Our default for everyday use, allergies, and pet owners.
• Cooking / Wildfires — 8 ACH. Maximum filtration for smoke and fumes.

Example: a 300 sqft bedroom with 8 ft ceilings at the Recommended goal needs (300 × 8 × 5) / 60 = 200 CFM.

Once we know the CADR a room needs, here's exactly what happens:

1. Filter candidates

Only purifiers whose CADR (quiet or max, depending on mode) is within 5% of the needed CADR — a purifier delivering 95% of the target still qualifies. CADR measurements vary by test protocol and lab conditions, so a small buffer prevents a good option from being excluded over measurement noise. Purifiers below any set filter lifespan minimum are excluded. If "premium design" is on, only premium-tier models qualify.

2. Best Value

Qualifying purifiers are sorted by price. The Best Value is the cheapest purifier that meets your room's CADR requirement. Anything over 3× the needed CADR is moved to "more options" as overkill.

3. Upgrade Pick

The Upgrade Pick is the highest-CADR purifier from the premium design tier, subject to the same 3× overkill cap. If the highest-CADR premium is the same purifier as the Best Value, no separate Upgrade Pick is shown — the single card gets both labels.

Sometimes two or three smaller purifiers can cover a room for less money — especially in quiet mode, where multiple units at very low speeds beat a single unit at a higher speed.

How combos work

We test same-model pairs (2× or 3× one purifier), mixed pairs (one of A + one of B), and mixed triples (2× A + 1× B). Each combo's total CADR is the sum of the individual units' CADR at reduced noise levels.

Noise in combos

Multiple air purifiers in a room make more noise. Taking some reasonable simplifying acoustic assumptions, 41 dBA total (our “quiet” limit) is outputted by two separate units each at 38 dBA; three units each at 36 dBA add up to the same. So in order for a combo to be eligible in a quiet room, its total noise needs to come in below the threshold. Having two or three units each at 41 dBA themselves will create more than 41 dBA of noise.

Combo scoring

Combos are compared against single-unit options using a value score:

The price is penalized for inconvenience: multiplied by 1.25× for 2 units, 1.50× for 3 units, accounting for extra filters and floor space. A combo is only recommended if it's strictly cheaper than the best single and its value score beats the 2nd-ranked single.

When a combo wins, the best qualifying premium single is promoted to Upgrade Pick so you always see a single-unit alternative.

We rely on data from AHAM (manufacturer-reported CADR ratings), manufacturer websites, HouseFresh (independent lab testing in a controlled environment), and AirPurifierFirst (per-speed noise and power measurements), and others.

Noise levels come from independent measurements where available. For a small number of purifiers, only manufacturer-reported noise specs are available — these tend to be significantly lower than independent measurements (manufacturers often test in anechoic chambers or at negligible airflow speeds) and should be interpreted cautiously. Where a related model has been independently tested, we use that data as a proxy.

Not every purifier can be included — some manufacturers don't publish comparable data. For more on what's excluded and why, see About.