Affiliate disclosure: The Pet Docket earns a commission from some links on this site. That never changes what we publish or how we rank a product. How we make money
The Pet Docket
Methodology The Pet Docket Disclosure

Find the right feeder

Set your constraints and the dataset filters as you go. Every row links to the sourced breakdown behind it.

Filter the five feeders

Set the constraints

Five automatic feeder records, filtered as you change each control.

Every control reads a field the source published: the food the feeder handles, the hopper capacity, the number of meals a day, the price we read, the Wi-Fi band, the warranty length, app control and the noise figure. A record that does not publish the value you are filtering on drops out rather than being guessed at.

The count under the list tells you how many records are still standing. Each row keeps its source and the date it was read, and the button takes you to the breakdown behind it.

Automatic feeders

5 of 5 shown

What the dataset holds

Seventeen products across four families, thirty four to forty fields each.

Five of the records are automatic feeders, eight are self-cleaning litter boxes, two are water fountains and two are indoor dog potties. Prices are dated on fifteen of the seventeen records, and every one of the seventeen carries a cited source. We have published no first-party measurements yet.

A cat's paw dipping into a bowl on a dark surface, lit from one side.
5
feeder records, each one with a cited source and the date it was read
0
first-party measurements published, because our own bench is not running

What it cannot tell you

The finder filters published facts. It cannot tell you how a feeder behaves in your home.

  • Whether the feeder keeps time when the power does not. Backup is published where the source stated it.
  • Whether it tells your pets apart. Recognition is published where the source named a method.
  • Whether the price will hold. Every price is an observation at a named source on the date it was read.
  • Which feeder is best. The list filters, it does not rank, and we publish no score we have not earned.

The same data, no interface

Filtering happens on your side. The dataset itself never changes.

The same records are available as JSON at /api/catalog.json, described in the OpenAPI document, and reachable by agents through the public lookup service.

Fetched live from the public API. Your filter selections never leave your browser.

Before you filter

Does the finder change for dogs? There is an animal control at the top of the filter. Four of the five feeders are sold for both cats and dogs, and one, the PETKIT YumShare dual hopper, is sold for cats only. Switching the control narrows the list to the records the maker sells for that animal. The animal a product is sold for is a property of the product, so nothing else on the page changes.

What does the filter read? Every control reads a field the source published: the food the feeder handles, the hopper capacity, the number of meals a day, the price we read, the Wi-Fi band, the warranty length, app control and the noise figure. Nothing is estimated. A record that does not publish the value you are filtering on drops out of the list.

Can it feed wet food? The food control separates dry only feeders from the models the source says handle wet food. Where a source did not say, the record does not claim it, and the specification table on each review says so.

What happens in a power cut? Battery backup is published where the manufacturer stated it. Where the source did not state one, the record says the source did not state it rather than guessing.

Can I use the data myself? Yes. The dataset is published as JSON with a documented schema and an open endpoint, and the endpoints are described in an OpenAPI document. Use it, cite it, and tell us what you found.

Read the record, not the summary

Every figure on a review page carries the page it came from and the date we read it.

See the breakdowns