Pawrent.
One calm home for the four things every dog owner Googles at midnight.
Pawrent is a thought experiment: if I were to build a companion app for dog owners, this is how I would actually run the project. There is no global home for dog people today. Breed knowledge lives on forums, training lives on YouTube, vet questions live in 11pm panic-Googling, and dog-friendly places live in scattered blog posts. The brief below is the UX process I would follow, the screens I sketched to pressure-test it, and the honest places I expect it would break.
How I would frame the project
- 01
Purpose and problem area
Dog ownership is a global, emotional, high-stakes job we hand to people with no central support. The cost is paid by the dogs in mismatched expectations, avoidable behaviour issues, and late vet visits. The hypothesis worth testing: one calm, breed-aware home would beat ten loud tabs, even if it covers less surface area. The first job of the research is to confirm or kill that hypothesis before a single hi-fi screen exists.
- 02
Methodological approach
Six to eight semi-structured interviews across three segments (first-time owner, experienced owner who moved abroad, rescue foster). A diary study with three owners for two weeks, one prompt a day, to surface the moments the existing tool stack actually fails (mostly evenings, mostly alone). A competitor audit of Rover, 11pets, Dogo, Woofz, and the breed-specific Facebook groups, mapped against a JTBD frame. An open card sort with around twelve owners to pressure-test the four-pillar IA (Know, Train, Pack, Map) before committing to a tab bar. Paper sketches, then a low-fi Figma flow, then five-task usability tests on a clickable prototype.
- 03
Potential shortcomings
Recruitment skews toward people who already use apps for their dog, so the diary study would undercount the offline-only owner who is arguably the target. JTBD framing can flatten emotional nuance (grief, guilt, identity as 'a dog person') that probably drives retention more than utility does. The four-pillar IA is the bet I am least sure about: card sorts often surface a fifth bucket (health records, food) that users insist on, and holding the line at four would frustrate them. Ask a Vet is the riskiest surface, regulated, regional, and easy to get dangerously wrong. Without a licensed vet in the loop from week one, that pillar cannot honestly be tested.
- 04
Intended outcome
A clickable concept prototype across nine screens (login, breed-aware home, two training surfaces, group sign-up, in-session player, community, map, vet chat), a service blueprint for Ask a Vet including escalation paths, a one-page positioning doc, and a written set of product principles. The success bar I would set with the team: a usability test where five out of seven participants can describe Pawrent's value in their own words after 30 seconds, and at least three of them name a pillar other than the one they expected to like.
- 05
What I would expect to learn
My prior is that the map is the hero. My suspicion is that testing would push back: the breed-aware home is what earns enough trust to use everything else. Knowledge as the front door, the map and meetups as the reason to stay. If I pushed it further, I would prototype vet triage with a real licensed vet, and stress the cold-start problem by manually seeding the map for one city (probably Copenhagen) before pretending it works globally.
How I would approach research
A handful of conversations, looking for one shape. The pattern I would expect to find: every owner has a 'dog Google folder' of screenshots, saved Reddit threads, half-read training PDFs, and the name of a vet a friend recommended. No one trusts a single source. The pain is not lack of information, it is the cognitive load of triaging it alone, usually at 11pm with a worried dog on the sofa.
The insight I would write on a post-it before designing anything: build the calm room, not another loud feed. Pawrent should feel less like a content app and more like a quiet, well-organised home where the answer is already waiting when you walk in. Every later design decision (no streaks that shame, hidden follower counts, an honest empty-state map) traces back to that sentence.
"Build the calm room, not another loud feed.
Three concept directions I would explore
The Companion
Four equal pillars, one calm home.
Know, Train, Pack, Map, plus Ask a Vet as a persistent surface. Equal weight in the tab bar, unified visual language, breed-aware content threading through all four. The direction I would pick. Hardest to scope, but the only one that solves the actual jobs to be done.
The Encyclopedia
Breed-first, content-led, reference-grade.
Wikipedia for dogs with a thin booking layer for vets. Cheaper to build. I would rule it out in a heartbeat: reference is not a habit, people read it once and forget the app. Retention would die in week two.
The Network
Social-first, meetups and map only.
Strava for dog people. Strong daily use, but it would feel fun without being useful. The breed and vet pillars are what make Pawrent serious. Ruled out.
Login, the promise on the door
Home, Louis at the centre
First-week onboarding
- Step 01
Tell us about your dog
Breed (or mix), age, neuter status, city. From this single input, every pillar personalises: training opens to the right program, Explore filters to your area, Ask a Vet routes to your timezone. This is the only data Pawrent insists on, which is itself a research finding: people give more when asked less.
- Step 02
Pick one thing to learn
The app surfaces the most-skipped chapter for that breed (recall for Shibas, leash pulling for Huskies, separation for Doodles). Learn one thing before being asked anything else. The single early win is what carries a new owner past day three, where most habit apps lose them.
- Step 03
Save your first spot
Three dog-friendly places within a kilometer, ranked by the local pack. Save one. Now the map has a reason to exist for that user, and we have a real signal against the cold-start problem.
The four pillars, one tap away
Solo training, the daily loop
Group training, the in-person loop
Product principles on the one-pager
- Four pillars, no fifth. If a feature does not serve Know, Train, Pack, Map, or Ask a Vet, it does not ship.
- Breed context is the unifier. Every screen knows what dog you have and adapts.
- Ask a Vet is non-emergency, loudly. Emergency routing lives one tap away on every chat surface.
- The map is honest. Empty cities show as empty, not faked. Trust over coverage.
- Copy is written like a calm friend who happens to know dogs. Never like a brand, never like a coach.
- No streaks that shame, no leaderboards, no public follower counts. The dog is the reward.
What I would do next
Prototype the Ask a Vet triage with a licensed vet in the loop, including the regulatory edge of what can and cannot be answered in writing per region. Without that, the pillar is theatre. Manually seed the map for one city (probably Copenhagen) and A/B test dense local data against thin global data on early retention. My bet is that local depth wins by a wide margin, but it is the kind of bet you only know by running it.
Run the diary study across three owners and three very different dogs (small reactive, large working breed, senior rescue), because a single persona would quietly break a product like this. And resist the feed. Every super-app falls into one eventually. Pawrent's whole personality depends on not doing that.
Tiny Rituals.
OpenA pocket app for building one tiny habit at a time.