How deep to put a moisture meter in the pot
The first time I used a moisture meter, I jammed it maybe an inch into the topsoil of my fiddle leaf fig, right up against the trunk, and got a reading of 15%. Bone dry, I figured. I watered. Two days later I did the same thing — different spot, different depth, whatever felt natural — and got 38%. I remember standing there genuinely confused, like the plant had changed its mind about being thirsty in forty-eight hours. It hadn't. I had just measured two completely different parts of the pot and called them the same number.
That fiddle leaf went through about three weeks of me chasing readings before I noticed the pattern. Some days I'd poke near the rim where the mix was chunky and dry-looking. Other days I'd push in close to the stem where it always seemed to run wetter. I was, without realizing it, running an experiment on soil variability instead of checking on my plant. The meter wasn't lying to me. I was just asking it a different question every time and getting mad when the answers didn't match.
The part where I figured out why my numbers never agreed
Here's the thing nobody tells you when you buy your first probe: a soil moisture meter is an unpowered galvanic probe. It's reading electrical conductivity between two dissimilar metals and printing that on a scale — it is not a lab instrument, and it doesn't average the whole pot for you. Move it a few inches in the same pot and it will read differently, because the soil a few inches over genuinely holds a different amount of water. Topsoil dries first and fastest. It's exposed to room air, it loses moisture before anything deeper does, and it is, by a wide margin, the least informative part of the pot to measure.
The roots that actually matter — the ones deciding whether your plant is thriving or slowly drowning — aren't up there. They're lower down, out past the immediate root ball near the stem, in the zone the plant has actually grown into. So a shallow reading right next to the stem measures leftover surface moisture and tells you almost nothing about what the root system is sitting in.
How deep to put a moisture meter in the pot
The answer that finally stopped my numbers from fighting each other: about two thirds of the way down the pot, midway between the stem and the rim. Not skimming the surface, not scraping the drainage hole. That mid-depth, mid-radius spot is where the roots that matter actually live, and it's far enough from both the drying-out topsoil and any weird edge effects at the very bottom.
The depth matters less than the consistency of it, though. Since the meter measures conductivity, not a calibrated moisture percentage, the number it gives you is only meaningful in comparison — to itself, in that same pot, at that same depth and spot, over time. Two different meters in the same pot will often disagree with each other. The same meter moved a few inches will disagree with itself. That's not the meter malfunctioning; that's just what an unpowered probe does. So pick your spot — two thirds down, halfway between stem and rim — and use it every single time. That's what turns a random number into a reading you can trust.
This is also why I stopped trying to water my plants on a calendar once I actually understood what the meter was telling me. Every pot dries at its own rate depending on species, pot size, mix, and season — a monstera might need water twice a week in July and once every three weeks in January. A fixed schedule can't know that. A consistent reading, taken the same way each time, can. If you want to see how that percentage maps to what a specific plant like a monstera or a pothos actually wants, the moisture meter chart is worth a look — most common tropical houseplants want to see that number drop into roughly the 20–40% range before they're watered again, though the exact target varies by species.
Where it still gets weird
Chunky mixes throw a wrench in even a well-placed reading. If your plant is in bark, orchid mix, or a chunky aroid blend, the probe can land against an air pocket or a piece of bark and read far drier than the pot actually is. The reading is still useful — you just have to judge it against that same pot's own history rather than any general chart. Succulents are their own case entirely; they want to dry out much further, down near 0–10%, before the next drink, so don't panic at a low number there the way you might on a peace lily. And if a pot never seems to read low no matter how long it's been since watering, that's usually fertilizer salts raising the conductivity, not a drainage problem — worth ruling out before you blame the plant.
Where the guessing actually stopped
What changed things for me wasn't just finding the right spot — it was writing the numbers down. I'd take a reading at that same two-thirds depth every week, walking the same probe from the fiddle leaf to the pothos to the peace lily, and start noticing shape instead of single data points: this one used to dry out in five days and now it's taking ten, that one's drinking way faster since I moved it to a brighter window. That's the kind of thing a leaf never tells you until it's too late — droopy and yellow looks the same whether a plant is drowning or thirsty, and the soil is the only honest witness. Once I started logging readings instead of just glancing at a number and shrugging, Plant Tender started catching those trends for me automatically — I just point my phone at the meter's screen each week and it tracks the direction each pot is moving, which is the actual signal, far more than any single reading ever was. You can set it up at plant-tender.com or jump straight to signing up if you've already got a meter sitting in a drawer somewhere.
The fiddle leaf is doing fine now. No more chasing contradictory numbers around the pot — just the same spot, two thirds down, halfway between stem and rim, every single week.