Honest limits

What this checker can’t tell you

The spot checker gives you a verdict in about a minute, free, without an account. Here is what it gives up to do that, and where it stops. If any of this changes how much weight you put on the score, it has done its job.

It estimates. Nothing here has seen your room.

The checker on this site never takes a reading. It works out the light from where you are, the time of year, how far the spot sits from the nearest window, and what you told us about the sun that reaches it. That is why every result on the page carries an estimated label rather than a measurement.

What it cannot know is everything between the sky and the plant: your blinds, the tree that leafs out in May, the balcony over your window, the building across the street. Any one of those can move a spot a whole tier, and none of them are in the answer.

The compass sets airflow, not light

This one catches people out often enough that it is written on the checker itself. Drag the needle to the way your window faces and the light result will not move. Facing feeds the airflow read — and indoors it does not even do that, because indoor airflow is not derived from outdoor wind.

So a facing that changes nothing is not a bug, and a north-facing and a south-facing spot at the same distance will give you the same light estimate. If the direction your window faces is the thing you actually want judged, this page is the wrong instrument for it.

A lamp is invisible to it

The estimate is built on daylight. If a grow light is doing the real work in that spot, nothing here can see it, and the spot will read darker than it is. Read a poor light verdict as daylight alone is not enough here, which is a different sentence from this plant will struggle, and add the lamp back yourself.

Two honest readings of the same spot can disagree by nearly double

This is a limit of phone light meters generally, ours included. In our own calibration rounds a single phone reading, taken twice at the same spot, sat on top of real values of 1,800 and 3,500 lux. Run both through the same engine and the same plant scores 65 and 93.

We wrote that up in full, including the part where being off by double changes nothing at all: are light meter apps accurate for plants? It is the least flattering page on this site and it stays up.

When too much is missing, it says so

Skip the location step and the checker can still work out light, but not warmth, humidity or airflow. The badge switches to light only, the stars come off, and the verdict narrows to the one dimension it can stand behind. Can’t assess yet is a designed answer here, not an error state.

The same rule runs through the plant data. Where a plant’s pet-safety record has not been verified, the result says Pet safety not verified instead of guessing from a related plant. Where the care details for a plant are AI-suggested and our team has not reviewed them yet, the app says so on the plant itself. We would rather show you a gap than paper over it.

What it is actually good for

Comparing. The estimate carries the same assumptions everywhere, so the difference between two spots survives even where the absolute number wobbles — which is why the checker will hold two spots side by side and tell you which dimensions separate them. Two spots one wall apart can be two different climates, and that gap is the part we can stand behind.

It is also honest about tiers. A dark corner and a bright wall are never confused; the doubt lives inside one narrow band, not across the range. Read the tier, act on the tier, and the estimate will carry you.