TAP. TRACK. THRIVE.
KeeperOS
Feeding night is a route, not a form. The tracker is built for the route, and the freezer is part of it.
Feeding day has a shape that software usually ignores. You thaw on a count decided the night before, you work a rack from top to bottom because that is how the tubs are arranged, and you make three or four judgement calls on the way — this one refused again, this one has outgrown its size, that one is in blue and should be skipped. Then, once a month, you discover the freezer is emptier than you thought.
A feeding tracker that only records what was eaten misses most of that. The record is the easy part and it is worth almost nothing on its own. What matters is whether the logging keeps up with your hands, whether the patterns surface before they become problems, and whether the stock in the freezer is connected to the animals that will eat it.
KeeperOS treats the feeding round as the unit of work rather than the individual meal, and treats the freezer as inventory with a burn rate rather than a number you update when you remember.
You prepare before you thaw. The Prepare Today view rolls up what is due into a count by feeder size, so you take the right number of the right thing out of the freezer once, instead of going back twice. A thaw strip tracks what is out and warming.
Then you work the rack, not the list. Open the rack and the tubs are laid out as they physically sit; a session round steps down them in one pass so logging keeps pace with the route your hands are already taking. What was offered is recorded on the enclosure path — the actual item, not just a tick.
The judgement calls are surfaced rather than remembered. Two consecutive refusals since the last accepted meal tags a problem feeder, so a run is flagged while it is still two rather than after it has become six. A size-up hint raises when an animal has taken the same size for enough consecutive meals to have outgrown it — the threshold is your account's, not a fixed rule.
The freezer is a ledger, not a guess. Feeder stock is append-only: you restock by packs with a per-unit cost, consumption is joined to the feedings you actually recorded, and a 120-day burn-down projects a run-out date per size. That is the number that tells you to order, and it is derived from your real demand rather than an average.
Colonies, if you breed your own, get their own attention view, and scent, method and feeder-size vocabularies are yours to define so the log reads in your words rather than ours.
The round is fast because the tubs are laid out where they physically sit, which is really a property of the room view rather than of feeding — rack management explains that half. Everything the log feeds into afterwards, from growth curves to care scheduling, sits under husbandry.
The one product built around the opposite answer to ours is Reptile Scan, whose whole feeding loop is a QR scan at the tub; if that appeals more than tapping a screen, our comparison with them says so plainly and sends you there.
Actively built: 47 dated updates since June 2026, most recently Choose which KeeperOS emails you get on Oct 2. See everything that has shipped →
Whether it survives contact with a feeding night. If logging forty animals takes longer than feeding them, the log stops getting written and the data is worthless within a month. Speed on the round is the whole product; the charts are a by-product of having actually captured the data.
Yes. When an animal has accepted the same size for enough consecutive meals, a size-up hint is raised against it. The threshold is configurable per account, because a hognose and a boa do not deserve the same rule.
A refusal is logged like any other outcome, and two in a row since the last accepted meal tags the animal as a problem feeder so the run is visible early. Nothing is hidden or auto-resolved — a snake that goes off feed in winter is normal, and the app's job is to show you the pattern, not to panic about it.
That is what the burn-down is for. Stock is an append-only ledger, consumption comes from the feedings you actually recorded, and the projection returns a run-out date per feeder size over a 120-day horizon. It reflects your real demand, so it moves when your collection does.
Yes. Weights and feedings share a timeline in Insights, with shed and clutch events overlaid, so a growth curve that flattens can be read against the meals that did or did not happen around it.
No. Logging feedings is free and uncapped; the feeder stock ledger and its burn-down are part of the paid collection-level layer, along with the Calendar and Insights.
The demo needs no signup. The free tier needs no card, and we don't sell your data.