Recipe Fox

Problem

I love cooking and finding new recipes online. But discovering a recipe and having time to cook it rarely happen at the same moment - I'd get excited, then forget it existed in a browser tab or notes app.

Online recipes are their own problem: optimized for Google, the actual recipe buried under a life story, banner ads, autoplaying video, and pop-ups.

So the problem was two-fold:

1. I don't have anywhere to save the recipes I find online and pull them out whenever I actually have the time to cook.
2. Recipes online are filled with content that is making it hard to actually follow along the recipe.

My Role

Recipe Fox is a solo project - I owned product strategy, UX and visual design, the App Store and Google Play listings, and launch. Claude Code carried the engineering and the same-day OTA fixes that keep it maintained. No handoff - I work the way I describe in How I build.

Decisions

The most useful exercise was mapping the full journey - from spotting a recipe to standing at the stove. The two ends are opposite contexts: browsing is casual (time, both hands, full attention); cooking is none of those. No app I tested reflected that - all built for browsing, leaving you to manage the kitchen yourself.

01

Radical scope restriction

Before setting scope, I mapped and tested 12 recipe apps, read 200+ one-to-three-star reviews of the top competitors, and searched cooking subreddits for frustrations. The field had clustered in one place: social-first, feature-heavy, built around sharing - not cooking.

Three complaints recurred across 200+ reviews: too many features, too complicated, "why am I paying $40 a year just to store my own recipes." The considered-and-rejected path was adding adjacent features (meal planning, shopping lists - proven, and most competitors have them). I went the opposite way: strip it to one thing and make that genuinely good. No shopping list, no social feed, no meal planning.

Same data shaped pricing: a $9.99 one-time unlock, not a $30-50/yr subscription. Pay-once doubles as a demand signal - if people pay up front for a stripped-down app, the scope bet is validated. I also didn't user-test before building - deliberate. The category is proven, so demand wasn't the question; the research answered whether a real gap remained. Building before testing isn't reckless if you've done the homework.

Cutting scope was a design decision, not a resource constraint.

02

Share extension over copy-paste

V1 shipped with URL paste. It worked, but barely - Sentry reports and my own daily use kept surfacing failed imports (TikTok and Instagram especially), and the seven-step flow made every save a chore. I replaced it with a native share extension.

Before - URL paste (7 steps)

1. 1 Open recipe in browser
2. 2 Tap share
3. 3 Copy link
4. 4 Switch to Recipe Fox
5. 5 Tap add
6. 6 Tap input field
7. 7 Paste & save

After - share extension (2 steps)

1. 1 Tap share
2. 2 Tap Recipe Fox

Import friction determines whether users build a recipe library or abandon the habit after a few saves.

Share extension in the shipped product.

03

Cook Mode - one step at a time

Most apps had a basic cook mode, but none designed for the cooking context. I built a dedicated Cook Mode showing one step at a time, not a full scrollable page. A scroll is fine with both hands free - terrible when they're wet or holding a spatula. One step at a time removes navigation entirely: the cook sees exactly what they need now, nothing else.

Cook Mode in the shipped product.

Outcome

The number that matters most: two of every three recipes saved come in from TikTok or Instagram through the share extension. That's the product's core bet - people cook from social video now, not just websites - showing up in production data, and it's why the share extension got the design investment it did. Beyond that: 500+ sign-ups with $0 in paid marketing, 233 of them in the last 30 days - the strongest month yet, driven almost entirely by TikTok, which users name as how they found the app. People who save recipes keep 4+ each.

Soft launch - no press, no paid marketing. I found people who already had the problem: cooking subreddits, parenting forums, and food groups full of "what do you use to save recipes?" threads, and recommended Recipe Fox where genuinely relevant.

Longer term, I started building Instagram and TikTok content - clips of the share extension and Cook Mode - to grow an audience before the next major update, not after. 200+ sign-ups came in the first month, now past 500, still $0 paid - and growing faster than at launch, a better signal than a spike that fades in a week.

Lessons learned

• Slow down on shipping. Early on I pushed releases fast, thrilled by how quickly an idea became a feature. That energy was better spent on clarity than velocity - not every idea needed to ship in week two.
• Start with distribution, not the product. The hard problem isn't building something that works - it's finding the people who need it. Next time: build an audience before the app exists - waitlist, content, community.
• A well-built app is invisible without a growth strategy. The product doing exactly what you imagined only matters when other people find it.