The wedge.
Pet food is genuinely opaque, the existing tools are either too shallow or too academic, and nobody has built a fast scanner that respects pet parents' intelligence. That's the gap.
What's in the market today
Dog Food Advisor and its sister site Cat Food Advisor are the gold standard for depth. Long-form text reviews, brand histories, recall tracking, expert analysis. Beloved by serious pet parents. Also: ad-supported, no barcode scanning, you have to know the product name to search, terrible on mobile, and "give me a score in 4 seconds" is not their model.
Whole Dog Journal is paywalled magazine-style content. Same depth issue, plus a paywall.
Yuka technically scans pet food but uses the same human-food algorithm — Nutri-Score on dog kibble is meaningless. They flag "additives" that are required for AAFCO completeness. The verdicts are misleading.
Petco/Chewy reviews are gameable, retailer-biased, useless for actual nutrition decisions.
Nobody owns fast scan + honest verdict + species-aware scoring + personal to your pet. That's the whole concept of MindYoPet in one line.
The audience.
"Pet parents" is too vague. The wedge audience is anxious-but-not-fanatic — people who want a quick honest check, not a 2,000-word breed-specific deep dive.
The target: thoughtful pet parents
They read the back of the bag but get confused by terms like "meat byproducts" or "ash content." They have a vague sense their current food might not be optimal but don't know how to evaluate. They've been spooked by recent recalls (copper toxicity in Hill's, salmonella in raw foods, the FDA DCM scare). They feel like Dog Food Advisor is too deep — they want a quick check, not an article.
They don't trust the marketing claims on bags ("with real chicken!" "grain-free!" "human grade!") but they also don't have time to research every product. They want a friend who knows what to look for, in their pocket.
The thoughtful, busy pet parent
Wants a fast honest check. Trusts vets. Reads labels but appreciates help interpreting them. Switches brands when there's a reason to.
The hardcore raw-feeding zealot
Already has a strong food philosophy. Distrusts vets. Will fight you in the comments about by-products. Doesn't want an app — wants a Facebook group war.
Picking your fight matters. The thoughtful audience is bigger, more reachable, more loyal, and more likely to share. The hardcore raw-feeding crowd will never be happy with any algorithm-based tool — engaging them is a trap.
The scoring philosophy.
Pet nutrition has unwinnable debates. Grain-free vs grain-inclusive. By-products good or bad. Raw vs kibble. Specific vet-trusted brands vs holistic-favorite brands. The honest move: score on measurable things, flag debated things, don't take a side.
What we score on (uncontroversial, measurable)
- Protein, fat, fiber, moisture content — straight off the guaranteed analysis panel
- First-ingredient analysis — is it a named meat (chicken), a meat meal (chicken meal), or vague (animal protein)?
- AAFCO statement — formulated to meet, or actual feeding trials passed?
- Life-stage match — does the labeling match your pet's actual life stage?
- Caloric density — kcal per cup, important for portioning
- Recall history — does the brand have a clean record?
- Country of manufacture — where is it actually made?
What we flag but don't penalize (debated)
- Grain content — we tell you whether it's grain-free or grain-inclusive, with one-line context on the FDA DCM debate. We don't dock points either way.
- By-product inclusion — we note it and explain what it is (organ meats, bone, sometimes feet — actually nutritionally dense, but the term sounds gross). Don't dock.
- Specific protein sources — list them, but penalize only if the user has flagged that protein as an allergy
What you configure in your pet's profile
- Species — dog or cat. Different scoring entirely.
- Life stage — puppy/kitten, adult, senior
- Known allergies — common ones: chicken, beef, fish, lamb, dairy, grain
- Health concerns — urinary, kidney support, weight management, digestive sensitivity
- Activity level — for caloric density assessment
The personalized score
Once a pet profile is set, the same product can score differently for different pets. A high-protein wet food might score 92 for an adult cat and 58 for a senior cat with kidney concerns (high protein is harder on compromised kidneys). The score isn't about the food alone — it's about this food, for this pet.
This is something Yuka and similar scanners literally cannot do with their current architecture. It's the strongest moat MindYoPet has.
Hill's Science Diet Adult Chicken & Rice Dry Dog Food
The same bag, scored against two different dogs in the same household. Both scores are honest; both reflect what matters for that specific dog.
The data problem.
Open Pet Food Facts has only ~10,000 products globally, and only 44 of those have nutrition data populated. That's catastrophic for a scan-first app where the value prop is "scan anything and get a result." This is the single biggest risk to the project.
Three options for v1 data sourcing
Pick ~300 products and own them
Top sellers from Chewy, Petco, Walmart, Petsmart. Cat: Friskies, Fancy Feast, Purina One, Blue Buffalo, Hill's, Royal Canin, Wellness, Tiki Cat, Stella & Chewy's, Smalls, Open Farm. Dog: similar Top 30 brands, popular SKUs.
Pros: Hit rate is great. Scoring quality is high (you can manually verify each entry). Doesn't pretend to be comprehensive.
Cons: You're hand-curating in v1. Scales poorly. Niche brands aren't covered.
Use the open database, prompt users to add misses
Same approach as KindlyChecked. Scan misses get a "help us add this" flow. The database compounds over time.
Pros: Scales. Open data ethos.
Cons: First 6 months UX is terrible — most scans return "not found." Friends test will be frustrating. Pet food is a small enough niche that contribution may never reach critical mass.
Different interaction model entirely
No camera. Just a fast typeahead search. "Type your pet food brand and product name." Smaller database needed because you don't need every variant.
Pros: No data coverage embarrassment.
Cons: Loses the magic moment of "scan and know." Less viral. Different product entirely.
Same family. Slightly softer.
MindYoPet sits in the KindlyChecked design system — same lime, same serif headline language, same manifesto voice. But pets are emotional in a way human food isn't. The voice should be ~10% warmer.
Headline candidates
The parallel structure with the parent brand is doing real work. Visitors instantly read "this is the pet version of that food scanner I like." Some options:
- "Mind yo pet." — Cleanest parallel. Works for both species. Probably the right call.
- "What's in their bowl?" — More specific to category, less brand parallel.
- "Read the bag, before they do." — Cute, slightly cheeky.
- "Mind yo pup." + "Mind yo cat." — Only works if we go single-species apps.
Tagline (sub-headline)
Parent app uses: "Healthy choices shouldn't be behind paywalls."
For pets, something like: "Honest scoring for what's in their bowl. Personalized to your pet. No moralizing. No paywall. Ever."
What the home screen says
Imagine a pet parent landing on mindyopet.app for the first time. The hero should immediately do three things: (1) tell them what it is in five words, (2) show that it understands their specific pet (not generic "pets"), (3) demonstrate the personal scoring with a real example.
Scoring rubrics differ in real ways between dogs and cats — protein needs, taurine, dietary obligation. The app asks once at setup, then everything is tuned.
What MindYoPet is not.
The negative space is as important as the positive. Here's what we explicitly refuse to be — drawn from watching what trips up other pet food apps and content sites.
- Not a vet replacement. Explicit disclaimer everywhere. "Talk to your vet about diet changes" appears in the app and on the site, especially around health-concern flags.
- Not anti-vet. We cite veterinary nutritionists (real DVMs with board certification in nutrition), not just internet influencers. Vet-prescribed diets get a pass on most flags.
- Not raw-feeding evangelists. Or anti-raw. We score what's in the bag, not the cooking method. Raw foods that meet measurable criteria score the same as kibble that meets the same criteria.
- Not breed-specific. Whole Dog Journal does this brilliantly. We don't try. Our wedge is fast scanning, not breed encyclopedias.
- Not a recall newsfeed. But scanned products do show recall warnings if applicable, with date and severity.
- Not a marketplace. No affiliate links to Chewy. No "buy now" buttons. No retailer commerce ever.
- Not subscription-funded. Same free-forever stance as KindlyChecked. Donations welcome eventually; paywalls never.
- Not a content site. No 2,000-word essays on grain-free debates. We're a tool, not a magazine. If users want depth, we link out to vetted sources.
v1 scope.
The discipline learned from KindlyChecked: ship the smallest version that delivers the wedge, then learn from real users. Here's what that looks like for MindYoPet.
In v1
- Curated database of ~300 most-common US dog and cat foods (kibble, wet, semi-moist)
- Barcode scan + search-by-name fallback
- Pet profile: species, life stage, allergies, health concerns, activity level
- Personalized 0–100 score with measurable / flagged / configurable breakdown
- Better picks engine — same category, higher score for this pet, with one-line "why" on each
- Recall warnings if the scanned product has been recalled
- Same offline cache, save, share patterns as KindlyChecked
- Multi-pet support (one household can have multiple profiles)
Explicitly out of v1
- Other species (rabbits, birds, fish, reptiles) — too niche, data is sparser
- Treats — separate category, would dilute focus, save for v1.5
- Supplements & vitamins — different scoring framework entirely
- DIY/homemade meal calculators — that's a different product
- Vet appointment reminders, weight tracking, photo logs — not what this is
- Community features (forums, comments, reviews) — adds moderation burden
- Multi-pet diet rotation tracking — possibly v2
The risks.
Five things that could turn this into a doomed project. Each has a mitigation. None are fatal if you see them coming.
Risk 1: Veterinary nutritionists trash you publicly
Mitigation: Cite them by name. Defer to AAFCO standards. Don't take sides on contested issues. Have an explicit "consult your vet" disclaimer in-app and on the site. Get one or two actual vet nutritionists to review the scoring rubric before launch — even informally. Their endorsement (or just "I reviewed it") is worth a thousand defensive blog posts.
Risk 2: Raw-feeding crowd brigades your reviews
Mitigation: Don't engage. They're not the audience. Set the precedent early that the app is for thoughtful pet parents who trust vets. Brand voice matters here — confident but not combative.
Risk 3: Pet brands send legal threats
Mitigation: Only show factual measurables. Never use words like "bad" or "harmful" — same as the KindlyChecked vocabulary repositioning. Cite sources. Allow brands to submit a "brand statement" that appears alongside the score (like Yuka does). If a brand is upset, point them to the scoring rubric (which is public).
Risk 4: Data quality dies if curation stops
Mitigation: v1 keeps the curated list small (~300) and accurate. Build a community contribution flow eventually, but only after launch teaches you what's actually missing. Set a quarterly "data refresh" rhythm — every 3 months, re-verify the top 50 products.
Risk 5: Personalized scoring is genuinely hard
Mitigation: Start with simple rules. A senior cat with kidney concerns docks high-protein foods. A puppy needs higher caloric density. An overweight dog needs lower. Build it as if/then rules first, not ML. You can refine later. The scoring being personalized at all is the wedge — it doesn't need to be sophisticated to be miles better than what exists.
What happens next.
This doc is on ice until KindlyChecked ships and proves out the friends test. The discipline is the whole point: build sequentially, learn fast, never have two unshipped projects at once.
Step 1 · Ship KindlyChecked
Real Vercel URL. Friends test active. 50+ contacted, 15+ active users. Zero critical bugs. v1 done as defined in the launch kit.
Step 2 · Learn from the food test
Spend 4 weeks watching what users do, ask for, complain about. Pay specific attention to: do food users mention pets unprompted? Do they ask for more categories? What's the most-requested feature?
Step 3 · Reserve the brand assets now
While the food test runs, claim the obvious URLs and handles: mindyopet.app, @mindyopet on socials. Cheap insurance against someone else owning the namespace.
Step 4 · Decide: pets next, or something else?
If users mention pets a lot, build this. If they mention supplements or baby food more, build that instead. Don't pre-commit to pets just because this doc exists. The doc is here so when you need it, you can move fast.
Step 5 · If pets is the call, build v1
Sister app, same brand family, separate URL. ~3-5 weekends. Curated 300-product database to start. Friends test the same way.