Care advice is full of light words and short on light numbers. This is the other way round: 204 readings taken with a light meter at 104 real spots in real homes, published as aggregates anyone can use.
Every reading here was taken by a person holding a light meter in a place where a plant might go. We keep them because our app scores spots, and a score is only worth anything if it is anchored to what rooms are really like. This page publishes the aggregate shape of that pile: how many readings, how they spread, and which light tiers they land in. Nothing here is modelled, and nothing here is a single person’s reading.
| Where | Readings | 25th pct | Median | 75th pct |
|---|---|---|---|---|
| Indoors | 49 | 420 | 2,349 | 6,200 |
| On a balcony | 25 | 3,756 | 8,173 | 24,208 |
| Not recorded | 127 | 1,718 | 3,142 | 7,270 |
| Outdoors, in the open (n=3) | 3 | 318 | 562 | 5,590 |
| Where | Below every tier under 270 | low 270–800 | medium 800–1,600 | high 1,600–10,800 | very_high 10,800+ |
|---|---|---|---|---|---|
| Indoors | 7 | 8 | 4 | 19 | 11 |
| On a balcony | 1 | 1 | 2 | 12 | 9 |
| Not recorded | 2 | 13 | 16 | 81 | 15 |
The word does almost no work. Indoor readings ran from 33 lux to the top of the scale, and the middle half of them, the part you would call typical, still spans 420 to 6,200 lux. That is a factor of fifteen inside one word, before anyone has argued about what bright indirect means.
The lower end matters more than the range does. Fifteen of the 49 indoor readings came in under 800 lux, and below 800 the only tier at home is low, which is the smallest of the four tiers in our species table. Seven of them were under 270 lux, which is below the floor of every tier we publish. Those are not readings from a cupboard. They are readings from spots in ordinary rooms where somebody was thinking about putting a plant.
The balcony column is the same point from the other side. Step outside the glass and the median goes from 2,349 to 8,173 lux, and the 75th percentile goes from 6,200 to 24,208. A wall of glass and a couple of metres cost more light than most people would guess, which is why a plant that was fine on a balcony sulks when it comes in for winter.
The readings above come from many homes, so they cannot show what happens inside a single room. This scan can. Six spots in one ordinary living room, all read within the same hour so the sun barely moved.
| Spot | Reading |
|---|---|
| At the bright window | 20,000 lux |
| 1 m from the window | 15,000 lux |
| 2–3 m away, in window shade | 3,500 lux |
| Middle of the room | 570 lux |
| Darker side of the room | 119 lux |
| The dark corner | 18 lux |
Both halves of this page say the same thing in different ways. A room is not one light level, and the label on the pot describes a spot rather than a room. If you want that answered for one spot of your own, the free spot checker runs in the browser.
It is not a controlled experiment. These are many spots on many days in many kinds of weather, not the same spot measured under two conditions, so the gap between the balcony and the indoor rows describes two populations rather than one before-and-after. It is not a census either: the outdoor row holds three readings, which is why we publish it for completeness and draw nothing from it. And it is not our light model. Our engine estimates a spot from window direction, distance and glazing; these numbers are what a meter said in a real room, and mixing the two in one table would let each borrow authority from the other.
Published under Creative Commons Attribution 4.0. Quote it, chart it, build on it, commercially or not. The one condition is attribution, and this line covers it:
The same two files are deposited where people look for data, so cite whichever suits you. Zenodo holds the archived version and the DOI, which is the one to use if you need a citation that will still resolve in ten years. Kaggle holds a copy you can open in a notebook without downloading anything. This page always carries the newest snapshot; the deposits are frozen at the date printed on them.