Most articles on blue light tell you the same three things: screens are bad, blue light suppresses melatonin, wear amber glasses at night. All three are roughly true. Almost none of them tell you the parts that actually matter for a real evening — the dose, the timing, and the difference between a phone screen and an overhead light.

This is the version we wish we had read before we started building products in this space.

What "blue light" actually means

The visible light spectrum runs from roughly 380 nanometres (deep violet) to 700 nanometres (deep red). The wavelengths that suppress melatonin most strongly sit in a narrow band between 460 and 480 nm — the cyan-blue end. This range maps to the peak sensitivity of a class of retinal cells called intrinsically photosensitive ganglion cells (ipRGCs). They are not the cells you see with; they are the cells that tell your circadian clock what time of day it is.

The single best paper on this remains Brainard et al. (2001), which mapped the action spectrum for melatonin suppression in healthy adults. Subsequent work by Lockley, Cajochen, and others has refined the picture, but the headline holds: 466 nm light is roughly 100 times more melatonin-suppressing than 555 nm (yellow-green) light at the same intensity.

Not all light is equal. A bright kitchen ceiling at 6500K hits the melatonin-suppressing band hard. A dim warm bedside lamp at 2700K barely touches it.

How much does it actually matter?

This is where most of the internet gets the story wrong. Light suppresses melatonin in proportion to dose — wavelength × intensity × duration. A laptop screen at full brightness, 50 cm from your face, for two hours before bed, delivers a meaningful suppression. A phone you glance at for 30 seconds does not.

The key threshold is somewhere around 30 lux of melanopic illuminance at the eye — below that, suppression is minimal. Above it, you start losing minutes of sleep onset for every additional 30 minutes of exposure.

In a typical Western home in the evening:

  • Overhead LED ceiling lights: 100–300 melanopic lux
  • Laptop screen at full brightness: 60–150 melanopic lux
  • Phone in dark room at full brightness: 30–80 melanopic lux
  • Warm bedside lamp at 2700K: 5–20 melanopic lux
  • Candlelight: under 5 melanopic lux

The ceiling light is the bigger problem than the screen. That is the finding that almost nobody acts on.

The three interventions that actually work

1. Reduce ambient light, not just screen light

Swapping your evening overhead light for a warm 1800–2700K bulb does more than any screen filter. The amber sleep bulb in our range was built specifically for this — it sits at 1800K, which is below the threshold where ipRGCs activate meaningfully.

2. Filter the screen if you must use it

Software-based "night mode" reduces blue output by shifting the colour spectrum, but the effect is modest — typically 10–30% suppression reduction. Amber-lensed glasses worn over your normal vision block the relevant wavelengths regardless of source. The strongest version (red lens, blocking everything below 580 nm) is the one we recommend for the last hour before bed.

3. Time it correctly

The ipRGCs become more sensitive in the evening, peaking around 2 hours before your habitual bedtime. The hour from 90 to 30 minutes before bed is when light exposure does the most damage. This is also the window where most people are scrolling on a couch under bright overheads. Fix that hour and you fix most of the problem.

What the marketing gets wrong

A few things we have seen claimed that are not supported:

  • "Blue light damages your eyes." No good evidence at any normal exposure. Eye strain from screens is mostly about reduced blink rate and accommodation, not blue light per se.
  • "All screen time before bed disrupts sleep." Time-of-day matters more than total time. A bright morning screen is actively helpful for circadian alignment.
  • "Blue light blocking glasses worn all day improve focus." Studies are weak and conflict. The evidence for evening use is much stronger.

What we would tell a friend

If you have £15 to spend on sleep, the highest-leverage purchase is a warm bulb for the lamp closest to your bed. If you have £80, add a pair of properly tested amber lenses for the last hour of screen time. If you have already done both and still struggle, the issue is almost certainly something else — sleep apnoea, anxiety, alcohol, exercise timing — and you should talk to a doctor before buying anything else.

The wavelengths matter. The marketing around them often does not.