Blog
How to read a vacuum spec sheet
Most of the numbers on a vacuum listing are not comparable. Here are the ones that are, and the footnotes that change what the rest of them mean.
By Scooter M. · Published September 1, 2026
Read the qualifier before the number. "40 minutes" and "40 minutes, measured at the hand vac in standard mode" are the same figure describing different machines. The brands that publish the qualifier are being more honest, not worse — and the ones that publish a single unqualified number are the ones to be careful with.
Figures that are genuinely comparable
- Weight in pounds. Comparable, published by nearly everyone, and it decides whether the machine gets carried upstairs.
- Bin capacity in liters or quarts. Comparable once converted. Watch for gallons, which make a small bin sound large — 0.14 gallons is 0.53 liters.
- Noise in dB. Comparable, and published far less often than it should be. Remember it is logarithmic: 90 dB is not 'a bit louder' than 67 dB.
- Cord length in feet. Comparable and genuinely useful on corded machines.
- Tank capacity on wet machines. Comparable, and the number that decides how many trips to the sink a room costs.
- Warranty in years. Comparable, and the clearest single difference between price tiers.
Figures that are not comparable across brands
- Pascals (Pa). Sealed static pressure, measured differently by every brand, method never published. Useful only within one brand's range.
- Motor watts. Electricity consumption, not cleaning. A rough proxy inside one product line and close to meaningless across brands.
- Runtime, without its qualifier. Measured at the handheld, in the lowest mode, often with no powered head. Levoit's honest version of the same figure is 50 / 30 / 12 minutes depending on mode and tool.
- Percentage filtration claims with no standard named. '99.9%' means nothing without a test method. Shark's NV356E names ASTM F1977-22 and IBR Laboratories; that is a claim you could check.
- Suction claims against 'the leading competitor'. Bissell and Hoover both do this without naming who. Unfalsifiable by construction.
The qualifiers worth hunting for
| Published claim | The qualifier | What it changes |
|---|---|---|
| Up to 40 minutes of runtime | Measured at the hand vac in standard mode | No powered floor head. Real cleaning runtime is lower. |
| 70x more power-lifting suction | As compared to the Roomba 600 series | Compared to the manufacturer's own decade-old model, not a competitor. |
| 2x better stuck-on stain removal | Per IEC/TC SC 59F (mustard), tested against the S1000 | Names the standard AND the comparison model. Genuinely checkable. |
| Traps 99.9% of dust and allergens | Based on ASTM F1977, tested by IBR Laboratories | Names the standard and the lab. The most specific claim on this site. |
| Best suction in a cordless stick vacuum under $200 | No comparison set named | Unverifiable by construction. |
| Kills 99.9% of germs, bacteria and viruses | When used as directed | Requires the dwell time in the manual, not a brisk pass. |
What a missing figure tells you
Absence is information. A cordless vacuum listing with no published warranty is telling you something. A robot with no published bin volume is telling you something. A steam mop with no published heat-up time when its siblings publish 30 seconds is telling you something.
This is why every product page on this site writes "Not published" in the specification table rather than leaving the row out. A gap that has been marked is a fact about the product; a gap that has been quietly removed is a fact you were not given.
The one-minute version
- Convert every capacity to the same unit before comparing anything.
- Find the runtime qualifier. If there is not one, assume the most flattering configuration.
- Check whether any performance claim names a test standard. Most do not.
- Check whether a comparison names its comparison set. Most do not.
- Note what is missing — warranty, noise, bin volume — and treat each absence as a data point.
- Then, and only then, look at the price.
That is essentially the method we run on every page here. It is written out in full, including the parts where it stops working.
Common questions
What vacuum specs actually matter?
The comparable ones: weight, bin capacity converted to a common unit, noise in dB, cord length, tank capacity on wet machines, and warranty length. Those are published in the same units by everyone and together they predict satisfaction better than any suction figure.
Why do vacuum runtime claims vary so much in practice?
Because of the measurement condition. Shark publishes 40 minutes measured at the hand vac in standard mode, with no powered floor head. Levoit publishes the honest breakdown for its own machine: 50 minutes in Eco with a crevice tool, 30 with the floor head, 12 in Turbo.
What does 'not published' mean on your spec tables?
That the manufacturer does not state that figure anywhere on its listing. We write the row rather than deleting it, because a marked gap is a fact about the product and a removed one is a fact you were not given.
How can I tell a real performance claim from marketing?
A real claim names its test standard and its comparison set. Shark's 2x stain-removal claim names IEC/TC SC 59F and the model it was tested against. 'Best suction under $200' names neither and cannot be checked.
Sources
- Manufacturer-published specifications — Every runtime, wattage, capacity, weight, filtration and warranty figure quoted on this page is the manufacturer's own published number for that model, retrieved September 1, 2026. Where a manufacturer publishes no figure, the page says 'Not published' rather than estimating one.
- Sweep & Suction — How We Pick — Our published method, including the scoring rubric, the three-year running-cost formula and an explicit statement of what published specifications cannot tell you.