Luke Angel
An end-of-2023 illustration in warm orange: a new house outline wired with network drops and a litter closet, beside an aging dog whose activity sparkline bends downward into a flagged dip — the year the household moved and a decade of pet data first became load-bearing.

2023 in pet IoT — a new house, and the year the data started to matter

by
#pet-iot#year-in-review#forecast

End of 2023. Move-in year, and the year Atom turned ten. Both of those matter, but the second one is the one that changed what this notebook is about. For a decade I've been putting gear on healthy animals and collecting data because I could. This year the data did something — it noticed Atom slowing down before I was willing to.

Scoring the 2022 forecast

#PredictionConfidenceOutcomeVerdict
1Mars acquires another notable player80%Champion Petfoods (Orijen/Acana) closed in February, plus smaller deals
2A credible AI behavior product ships65%Furbo Dog Nanny + Companion both mostly theater
3Atom's baseline shows real drift (turns 10)70%Yes — a sustained activity drop and broken sleep, mid-year
4Move into the new build, redesign the layout95%Done, October
5Petivity changes its model (acquired or adds a fee)50%No — still no subscription
6A non-Mars/non-Purina health tracker launches45%Nothing credible
7More Halo-class GPS shock-fences ship60%A few small competitors, same welfare problem
8First credible Matter-compatible pet device50%Matter hubs (Aqara and others) handle some pet gear, but nothing pet-specific✗ (partial)

Four hits, one partial — call it ~56%, my worst year on this scorecard. But the one that mattered most, bet 3, landed — and I'd have traded all the others to be wrong about it. More below.

A scorecard of eight 2022 predictions for 2023, drawn as a vertical list with a verdict mark per row. Four rows carry a green check — Mars acquired Champion Petfoods, Atom's baseline drifted, we moved into the new build, and more Halo-class fences shipped. One row is a half-filled amber mark — Matter hubs handle some pet gear but nothing pet-specific. Three rows carry a red cross — no credible AI behavior product, no Petivity model change, and still no non-Mars health tracker. A note at the foot reads four of eight plus a partial, the worst score yet, with the one that mattered — the senior-dog drift — circled as the bet the author wishes had missed.

What entered the house this year

  • A new house (October) — and for the first time I designed the pet IoT from the studs: 42 wired network drops, conduit run before drywall, a ventilated litter closet with the litter boxes and their sensors in one place. A decade of retrofits, finally a chance to do it deliberately.
  • A Litter-Robot 4 (one unit; a second still on order for the upstairs). The 4 is a real step up from the III — quieter, better weight sensing, more reliable per-visit data.
  • A second SureFlap door at the new back entry; the original moved to the patio.
  • A vet visit for Atom that the collar effectively scheduled. Whistle Health doesn't measure heart rate — it watches behavior — and what it watched this summer was a steady drop in Atom's activity and more broken sleep over several weeks. That nudged me to book a check-up earlier than his annual. The vet, with a stethoscope — the actual instrument — heard a mitral murmur, the common senior-Lab heart thing. Lifestyle changes and monitoring for now; no medication yet.

A timeline of the household's 2023 across twelve months. The dog lane runs all year — Atom on Fi plus Whistle Health, Quark on Fi — and mid-year a marked dip appears on Atom's behavior trace (activity down, sleep broken) leading to a vet-visit flag in late summer where a heart murmur is found. October: a move-into-new-house marker spans the lanes, annotated with 42 wired drops and a litter closet. The cat lane gains a Litter-Robot 4 and a second SureFlap door after the move. A note marks 2023 as the year the data first prompted a real-world action.

What worked

Whistle Health's behavioral signals earned their keep for the first time. Not by doing anything fancy — by noticing, across weeks, that Atom was moving less and sleeping worse, and surfacing it as a trend instead of a vibe. I'd been half-seeing it and half-explaining it away ("he's just getting older"). The trace didn't let me. That's the whole value proposition of a long boring baseline: it catches the slow change you rationalize. The collar didn't diagnose anything — it can't, and it shouldn't pretend to — but it got us to the vet a couple months earlier, and for a heart issue that monitoring window matters.

The new house's pet infrastructure is the quality-of-life win I waited a decade for. The litter closet alone — both robots, ventilated, sensors wired, out of the living space — is worth the whole exercise.

Petivity's per-cat attribution keeps doing its job: it agrees with my own weight-script the large majority of the time, and the disagreements are still the near-identical-weight visits nothing can call.

What didn't

The Litter-Robot 4's cycle timing is more aggressive than the III's, and the cats were briefly confused by a box that tidied itself while they were still thinking about it. Settled down after a couple weeks.

The "AI behavior detection" pitch, again. I tested it properly this year and turned all of it off. The data underneath is real; the interpretation layer is theater.

Forecast for 2024 — bets, with how sure I am

#What I expect in 2024ConfidenceRead or wish?
1More pet trackers join Apple's Find My network70%Read — the program's open; pets are an obvious fit
2Atom's heart needs active monitoring, maybe medication65%Read I'd rather be wrong about
3A non-Mars/non-Purina health tracker finally launches35%Wish — fifth year running
4Mars Petcare buys another competitor80%Read — it hasn't slowed once
5A consumer product does multi-pet ID by RFID, not inference50%Coin-flip — the obvious fix nobody ships
6Halo restructures or pivots under the welfare pressure35%Wish, mostly
7Atom's eleventh year brings the hard conversations75%Read — and the one I most want to push off
8A DIY ESP32 feeder matures into a real homebrew project65%Read — the vendor-cloud lesson is sinking in community-wide

The two bets I care about aren't the product ones. They're 2 and 7 — Atom. After ten years of writing about gadgets, the forecast that matters is about a dog's heart, and I'm hoping the high-confidence reads on that one are the ones I get wrong.

A 2024 forecast chart plotting eight predictions as horizontal bars by confidence from zero to one hundred percent, each shaded to mark a grounded read versus a wish. The two highlighted bars are about Atom — his heart needing active monitoring at sixty-five percent and his eleventh year bringing hard conversations at seventy-five — drawn in a heavier outline as the ones the author is rooting against. High reads cluster right: another Mars acquisition at eighty, more Find My pet trackers at seventy, a maturing DIY ESP32 feeder at sixty-five. Mid: RFID multi-pet ID at a fifty coin-flip. The lowest, dashed wishes: an independent health tracker at thirty-five (fifth year running) and a Halo restructuring at thirty-five. A line separates grounded reads from wishful thinking.

What I'm buying in 2024

  • More attention to Atom's data, not more devices — there's still only Whistle on the behavioral side, so it's the same collar and a closer eye on the trend, plus whatever the vet wants to track.
  • A DIY ESP32 feeder I build myself — vendor-cloud-independent, because Petnet taught me what happens when a feeder's brain lives on someone else's servers and the someone else folds.
  • Whatever pet tracker joins Apple's Find My in a credible way, if one does.

What's next

2024 is Atom's year, and I already know it. A senior dog with a heart murmur turns a decade of decorative data into something load-bearing — the activity trend, the sleep trace, the "is today a worse day" question I used to ask casually and now ask carefully. The category I've documented for ten years is finally going to be tested for the thing it was always supposed to be for: noticing, early, when an animal you love starts to change. I'd rather it had stayed a hobby. It didn't. So I'll write that year honestly too.

Keep reading

shares tags: #pet-iot · #year-in-review
tools
2014 in pet IoT — the dog side grew up, the cat side never started
Dec 23
tools
2015 in pet IoT — the cat side finally shipped, the dog side split in two
Dec 22
tools
2016 pet IoT in review — Whistle goes cellular, Mars buys in, Furbo lands
Dec 21