Short answer: a fan spec sheet quotes one number — the top setting. Our own bench test on three units shows why that number is close to useless for a buying decision. Two of the three units were within 2% of each other at their top setting, and 47% apart on their average across all five settings. The lowest setting on the weakest unit delivered a quarter of what that same unit could do at full speed. The peak tells you what the product can do once, briefly. The average tells you what it does for the six hours someone is actually wearing it.
What a spec sheet measures, and what it should
Quoting a single maximum figure is not dishonest — it is simply the cheapest measurement to state and the most flattering to publish. It answers the question "how strong can this get?" It does not answer the question a buyer actually has, which is "what will this do at the setting I will leave it on?"
Those are different questions with different answers, and the gap between them is not a rounding error. On our bench it was the difference between two units looking equivalent and one of them being substantially weaker.
Three units, five settings each
Method: a handheld anemometer centred about 5 cm in front of the outlet, with each unit run through all five of its speed settings in turn. Air speed is used here as the comparable measurement — see why air speed alone is still only half a measurement for what to add to it.
| Setting | Unit A | Unit B | Unit C |
|---|---|---|---|
| 1 (lowest) | 4.43 m/s | 3.30 m/s | 1.87 m/s |
| 2 | 5.26 m/s | 4.12 m/s | 2.79 m/s |
| 3 | 6.07 m/s | 4.67 m/s | 3.43 m/s |
| 4 | 7.09 m/s | 5.56 m/s | 5.05 m/s |
| 5 (highest) | 7.76 m/s | 6.92 m/s | 7.60 m/s |
| Average of the five | 6.12 m/s | 4.91 m/s | 4.15 m/s |
Why the peak hides the gap
Read the bottom row of the table against the row above it:
- At setting 5, units A and C are 2% apart. 7.76 against 7.60 m/s. No buyer would treat that as a meaningful difference, and it sits inside normal measurement variation.
- Across all five settings, they are 47% apart. 6.12 against 4.15 m/s. That is not a rounding difference; it is a different class of product.
Unit B is the useful control case. It was the weakest unit at its top setting — 6.92 m/s, about 11% below A — yet its five-setting average sits comfortably above unit C's. Judged on the peak, B is the weakest fan. Judged on how it behaves across the range, C is.
What the shape of the curve adds
The average is not the only thing the full set of measurements gives you. Two further patterns are worth reading directly off the table:
- How strong the low settings are relative to the unit's own maximum. Unit A delivers 57% of its own top speed at setting 1. Unit B delivers 48%. Unit C delivers 25%. For a device worn for hours, that low-setting figure is closer to the real product than the headline number is.
- How evenly the settings are spaced. Units A and B rise in steps of roughly 0.6–1.4 m/s from one setting to the next. Unit C rises by 0.92, then 0.64, then 1.62, then 2.55. Uneven spacing is not a defect in itself, but it tells you where the design was optimised: most of this unit's usable output is concentrated in its last two settings, and its first two are largely decorative.
A "five speed settings" line on a box says nothing about either of these. Two products can both claim five settings while behaving quite differently at settings one, two and three.
Why this matters more on a wearable than on a desk fan
A desk fan is usually pointing at someone across a room, at whatever speed it was set to and then forgotten. A wearable fan has a battery and a wearer's comfort in the loop:
- The settings people leave running for hours are the low and middle ones. Those are the ones a peak figure conceals.
- The top setting is typically used in short bursts — a few minutes of strong airflow before noise and battery drain become the deciding factor.
- Battery life is quoted at the lowest setting on some products and the highest on others, so the peak figure often travels alongside a runtime that was measured somewhere else entirely.
The result is a product that looks competitive where it is measured and disappointing where it is used. That is a specification problem before it is ever a product problem.
How to compare two units properly
- Ask for the reading at every setting, not just the maximum. A single number is a marketing figure; five numbers are a performance profile.
- Average them, and look at the average. It is a crude summary, but it ranks units the way a wearer experiences them far better than the peak does.
- Check the lowest setting as a share of the unit's own maximum. Under 30% means the gentle settings exist mainly to fill out a specification.
- Look at the spacing between settings. Even steps mean the range is usable throughout; one enormous final step usually means the top setting is where the effort went.
- Ask for the runtime at each of those settings, so a strong setting is not being paid for with a runtime that only exists elsewhere.
This is the shape of our own requirement list for this category: measured air speed at every setting, the computed air volume that follows from it, runtime at each setting rather than a best case, a sustained run rather than a spot reading, and noise at a stated distance at the same settings. Every item can be checked on a sample, which is the only kind of specification worth writing into a purchase order.
Frequently asked questions
Is a higher top-speed figure ever useful?
Yes, as information about headroom. It tells you the unit can reach a certain output in a burst. What it cannot tell you is whether the settings you will actually use are worth having — which is what a five-setting measurement set is for.
Why do manufacturers quote only the maximum?
Because one number is cheaper to produce, easier to present and always larger than the alternatives. It is also the number that travels best in a marketplace listing. That does not make it false; it makes it incomplete.
Does a bigger number of settings mean better performance?
No. The count says nothing about spacing or about how strong the lower settings are. A three-setting unit with evenly spaced, usable steps can be a better product than a five-setting unit whose first three settings are decorative.
How many settings should I test a sample at?
All of them. The whole point of the exercise is that the settings differ, and testing a subset reintroduces the same problem you are trying to avoid. Five settings take a few minutes on a bench.
What noise and runtime figures should accompany a speed claim?
Both, at the same settings, with conditions stated. A speed figure at setting 1 paired with a runtime measured at setting 1 and noise measured at setting 5 is three unrelated numbers presented as one product description.
If you are sourcing in this category, our airflow and runtime requirements are written as testable figures inside our OEM/ODM programme. The same argument applied to a hair dryer is in why maximum airflow is the wrong buying metric, and the companion piece to this one is air speed is not air volume. Request a quotation with your target market, volume and target airflow window.
Related reading: