How I check an app idea in an hour

On October 1, 2026 I ran every Northstave candidate through the App Store and the web before it entered the Brand & Shipping Plan. Most ideas failed. In 2026, nearly every simple local-first utility already has several polished indie versions.

That finding is why the studio builds brand through craft and cadence rather than through hunting empty niches.

What survived

New-construction punch list for homeowners (working name Closeout). Existing punch-list apps serve contractors and builders. I did not find a homeowner-side, deadline-aware app aimed at 1-2-10 warranty windows and the 11-month walkthrough.

What got cut or shelved

Idea	Why
Renter move-in/out inspection	Saturated (TenantDesk and peers)
Standalone warranty vault ("Covered")	At least seven local-first trackers in 2026; features salvaged into Upkept
Mac step-by-step guide recorder	StepGrab already does it on-device
Mac SwiftData / Core Data viewer	DataScout, Core Data Lab, and others
Mac calendar time analytics	Calendar Insights, CalStats, Timeview
Mac Foundation Models eval tool	LLM Eval Suite + Apple Evaluations
Standalone web tools	Saturated; weak no-backend monetization — web stays a funnel

Technical checks that changed the specs

• SwiftData still has no CloudKit sharing between Apple accounts → Closeout uses Core Data with NSPersistentCloudKitContainer.
• Both working names "Closeout" and "Covered" are taken on the store — rename work is a gate before marketing.
• Public recall APIs (NHTSA, CPSC) need no key, so recall checks need no backend.

The full launch order and dates live in the shipping roadmap on this site’s repo. The punch-list app is the flagship; Upkept and MenuPlan carry the home story with it.