PictureThis or PlantSnap: which plant subscription is worth it?

These apps do the same thing in the same way, so the honest comparison is about magnitude rather than approach.

PictureThis holds 1.1 million ratings at 4.8; PlantSnap holds 68,000 at 4.6. Review counts are a rough proxy for user base, and user base in this category is a direct proxy for reference-set size, because both apps improve from the photographs people submit. An order of magnitude more users compounds into an order of magnitude more reference images.

That advantage shows up specifically on cultivars, houseplants and ornamentals — the long tail of named garden varieties where more data is the only thing that helps. On a common wild plant both will get it right.

PlantSnap is fussier about framing, and that is a design decision from an earlier era of computer vision rather than a defect. It works best with a single leaf or flower filling the frame against a plain background, because separating the subject from clutter was the hard part of the problem when it was built.

Newer apps handle busy scenes better. But the old discipline still improves results in both, and it costs nothing: one subject filling the frame, even light, no flash, and the most informative organ — flower first, then leaf, then fruit, then bark.

The habit almost nobody has is taking three photographs instead of one: leaf top, leaf underside, and the twig showing how the leaves attach. Arrangement is the first branch of every real botanical key, which is why our free leaf key asks for it before anything else.

On 2026-09-03, PictureThis listed 4 in-app purchase tiers between $3.99 and $49.99; PlantSnap listed 10 between $0.99 and $59.99.

Comparing the low numbers is a trap. Apple lists in-app purchase names and prices but not the billing period, so a low-priced tier is frequently weekly rather than one-off, and a weekly subscription is several hundred a year — routinely multiples of the annual tier of the same product.

The tables below print every tier verbatim. Read them as shapes rather than prices: several tiers sharing a name means one product billed over different terms, the lowest number is the shortest term, and the app will present that one first because it looks cheapest.

Both of these apps compete with something that costs nothing. On 2026-09-03, PlantNet listed no in-app purchases at all — no subscription, no unlock, nothing to buy. It is published by CTRE COOP INTERNAT RECHERCHE AGRO DEV.

On wild flora it is at least the equal of either app here, and it is more honest: it asks which organ you photographed, the way a botanical key does, and returns a ranked shortlist with similarity scores rather than one confident name.

What it does not have is cultivars or care guidance, which is exactly what a subscription buys. So the real decision is not PictureThis against PlantSnap; it is whether your plants come from a shop or from the ground.

Deleting an app does not cancel its subscription — subscriptions live with your Apple Account, not with the app.

iPhone or iPad: Settings → your name → Subscriptions. That list is the authority regardless of what an app's own settings screen says. Mac: App Store → your name → View Information → Manage. Android: Play Store → profile icon → Payments and subscriptions.

If you were charged for something you did not intend to buy — a weekly tier read as one-off, or a trial that converted — request a refund at reportaproblem.apple.com. Apple grants those more often than people expect, particularly for a first accidental subscription.

Every price, tier, rating, version, size and update date is read from each app's US App Store listing on 2026-09-03 through Apple's Lookup API and product pages. Nothing is typed from memory and the build fails if a price appears in the prose.

Comparing two apps that work the same way is unusually hard, because neither publishes accuracy figures and independent benchmarks for plant identification barely exist.

What is available is review count, and in this category it is a better proxy than it looks. Both apps improve from user submissions, so users compound into reference images, and reference images are most of what determines whether a rarely photographed cultivar is recognised. An order of magnitude difference in users is an order of magnitude difference in the long tail.

What review count does not measure is accuracy on any particular plant, satisfaction among the people who left, or how the free tier behaves. A high average is entirely compatible with a large number of people who felt misled about pricing, because those people leave and stop rating.

Which is why the tables here print tiers verbatim rather than summarising them: the pricing shape is checkable in a way the accuracy claim is not.