How we research
By Arina Nikoliukina · Reviewed · Updated TailSpecs compares pet tech from public sources: brand product pages and manuals, app store listings, owner reviews, independent reviewers who publish their measurements, and the CPSC recall database. Every number on the site carries its source and the date we checked it. We do not buy the products or use them with our own pets, so we never describe how a device felt or smelled; we show what the brand claims next to what others measured and what owners report.
Where the data comes from
Brand pages and manuals
Price, shipping, warranty, return window, subscription plans, consumables and specs come from the brand's US product page, its support pages and the PDF manual. We quote the manual with its page number when the product page and the manual disagree.
Apps and subscriptions
App ratings, rating counts and the privacy labels ("Data Linked to You" on the App Store, "Data safety" on Google Play) come from the store listings on the date shown. Plan prices come from the brand's plan page or its app store listing; when a price appears only after sign-up, we say it was not measured.
Independent measurements
Noise, days between emptyings, battery life and similar readings come from written and video reviews that name who measured, with what, under which conditions and when. A unit supplied by the brand is marked as such. Sponsored videos are not used for measurements.
Owner reviews
We read owner reviews on retailer sites, forums, app stores and review platforms, count complaints by topic, and quote owners word for word with the platform and date. Reviews that mention a free product, and quotes that make medical claims, are left out. When a site blocks automated reading, we say so instead of guessing.
Recalls
For every model we search the CPSC recall database and SaferProducts.gov by brand, model and product type. "No recall found" means our search on the date shown returned nothing, not that a product is free of defects.
How we verify
Each number on a page links back to one record in our database, and each record stores the URL and the date it was read. When two sources disagree, both numbers are shown with their sources; we do not average them. Prices are the brand's list and sale prices on the checked date (a named retailer's price when the brand lists none), before sales tax, which varies by state. A figure we could not find reads "not listed by the brand" or "not measured", never a blank or an estimate.
How we score
Every model is scored on five criteria. Each criterion gives 2, 1.75 or 1.5 points, and the score is out of 10. There are no weights. The same rules apply to every model in a category; the median is taken within the subcategory, or the whole category when a subcategory has fewer than three models.
The five checks
| Check | 2 points | 1.75 points | 1.5 points |
|---|---|---|---|
| Published specs | The key specs for the category are all published (for litter boxes: days between emptyings, minimum cat weight and litter type) | Some of them | None |
| First-year cost | Real Cost Index at or below the subcategory median | Up to 1.5 times the median, or the index cannot be calculated | Higher |
| Warranty and returns | Warranty at least as long as the median, and a return window of 30 days or more | One of the two | Neither (a final-sale item has no return window) |
| Owner reviews | Mostly positive | Mixed, or no reviews | Repeated complaints about breakdowns |
| App and subscription | Core features work without a paid plan, and the app rates 4.0 or higher with 500+ ratings in one store | One of the two, or no app and no plan | Core features need a plan, and the app rates below 3.5 in every store |
The key specs by category: for feeders, hopper capacity, meals per day and the smallest portion (or portion step); for fountains, water capacity, filter interval and material; for GPS trackers, claimed battery life, network and minimum dog weight (or the tracker's weight); for fences, correction levels, maximum area or wire length, and minimum dog weight; for cameras, resolution, field of view and the list of features that work without a plan; for DNA tests, the number of breeds, claimed turnaround and sample type.
"Mostly positive" means a retailer other than the brand's own store shows at least 25 ratings averaging 4.3 or higher, and no retailer listing averages below 4.0; app stores, forums and seller-review sites don't count here. "Repeated complaints about breakdowns" means that, out of at least 50 owner reviews we read, 3 or more and at least 5% report a mechanical or sensor failure (a jam for feeders, a pump failure or leak for fountains). That breakdown count is the only part of the owner data that feeds the score; independent measurements and the published Owner Complaint Index sit next to the score and do not change it.
The total sets the label: 9.5 and up is Outstanding, 9 to 9.25 Excellent, 8.5 to 8.75 Very good, 8 to 8.25 Good, and below 8 Decent.
Real Cost Index
The Real Cost Index is what a device costs in its first year: the price, plus shipping to the contiguous US, plus a year of the brand's plan, plus a year of consumables (liners, filters, cartridges) at the brand's own replacement interval and price. When the plan is optional, the total includes its cheapest tier and a second line shows the cost without it. Litter is left out, except the disposable crystal-litter trays that some boxes need, and so are treats and optional accessories. Sales tax is not included because it varies by state. If any part is missing, we publish the parts we have and write "Not enough data for a first-year total" instead of a partial sum. We also show "Each year after": the plan and consumables only.
Owner Complaint Index
The Owner Complaint Index is the share of the owner reviews we read that mention a problem, by topic: for litter boxes, sensor errors, app and Wi-Fi, odor control, a cat that won't use it, motor or mechanical failure, noise and customer support. A review counts once per topic, and reviews the brand picks to show on its own site are not counted. We publish the index only after reading at least 50 reviews of a model and show the top three topics at 5% or more, with the number read and the platforms. Below 50 reviews, the page says how many we read and where.
What affects rankings and what doesn't
Within a list, models follow the score; a tie goes to the lower price. A commission, or a brand's affiliate program before it pays one, affects three things: which models get a "Check price" button, which ones we pick, and the order, because models with a store link are listed before models without one (recalled and discontinued models come last). It never affects the score: a model without a store link keeps the same score. Brands cannot pay for a place in a ranking.
What we don't claim
We do not claim a device keeps a pet safe, healthy or calm. Safety features are described in the brand's own words with the source. We do not describe how a product performs with our own pets, and we do not run lab tests. For health questions, see the medical disclaimer.
Updates and corrections
Prices, plans and recall checks are refreshed from the same sources, and each page shows the date of the last check. If you find a wrong number, contact us with the page and the source; corrections are made in the database, so every page that uses the number changes together.
How to cite this data
Please cite the page and the date shown under its title, for example: "TailSpecs, Litter-Robot 4 review, reviewed 10/09/2026". The original source of each number is linked next to it on the product's spec page.