Message a number without saving it, on WhatsApp, Messages or a call. Our own iOS app, from day one to now.
Status
In review at Apple
Started
20 Jul 2026 · day 1 of the log
Scope
Positioning · naming · listing · launch
Built
Alongside the code, from day 1
The brief
We message one-off numbers all day: a delivery driver, a one-time client, a number on a flyer. Contacts wasn't built for that, and the only other way was a web form. We wanted the native version, with a widget, and to launch it in public, every decision logged.
20 JulDay 1Started
A personal habit becomes a product
Every send was a trip to a browser. The bet: a native app plus a Home Screen widget, where a repeat send is one tap.
Week 1Before codeDecided
One claim that has to stay literally true
No account, no tracking, nothing saved to a server. Every product decision since gets tested against that line: no analytics SDK, no bulk send in v1.
Weeks 2–5Rejected ×2
Two names, killed
Both for legal exposure, not taste. The one that survived reads as a phrase and echoes neither the competitor nor WhatsApp. Lesson logged: lock the name before the first commit.
AugShipped
The listing, to the character
The subtitle and keywords do the search work, so the name can stay readable.
App name15/30
Subtitle24/30
Keywords100/100
27 AugDay 39Shipped
First TestFlight, first submission
From a habit to Apple's desk in 38 days, with positioning, naming and the listing built alongside the code rather than after it.
28 AugDay 40Rejected by Apple
Guideline 4.2.3: a dead end
The reviewer had no WhatsApp installed, so our fallback led nowhere. Fair. That's exactly the kind of moment a launch log is for.
A rejection became a positioning improvement.From the fix, two days later
30 AugDay 42Fixed
No WhatsApp? Share the chat link
Without WhatsApp, the app now opens the share sheet with the chat link. That makes it useful on its own and makes the privacy claim stronger.
OctShipped
Hebrew, right to left
The whole app in Hebrew, mirrored properly. Country names stay in English, because that's how people search for them.
21 OctDay 94Submitted
Build 3, back at Apple
Resubmitted with the share-sheet fix, Hebrew and a polished listing.
22 OctShipped
A guide for the first minute
A first tester could not tell what the app was for. The intro grew to four pages with a drawn number-to-chat demo and a use-case page, and a ? button on the main screen opens the same guide any time. Built while build 3 waited: one build in review at a time.
28 OctDay 101Rejected again
Guideline 4.2.3 again: WhatsApp still required
Same rule as August. The share sheet had fixed the dead end, but the app still needed WhatsApp installed to do its job. Our read: the fix was a patch on the symptom. The dependency was the problem.
Fixing the dead end was not enough. The dependency had to go.From the build 3 rejection
28–31 OctFixed
WhatsApp becomes one channel of three
Sending is now channel-neutral. WhatsApp, Messages and Call sit side by side as “Open with” chips, and only the channels the phone can actually open are offered. The share sheet appears only when nothing else opens. Recents and the widget remember the channel each number was used with.
31 OctIn review
Build 4, at Apple
Submitted with Messages and Call as first-class channels, the in-app guide and Hebrew. Waiting for review.
NextPlanned
Launch, then the numbers
Product Hunt, Reddit communities and a launch video. Results get logged here after they're measured: downloads, widget use, launch rank. Nothing is claimed before it's measured.
Number, pick a channel, one button. Nothing goes to a server.
The country picker: every dial code, searchable.
The widget on a Home Screen. A repeat send is one tap.
No WhatsApp on the phone: the chat link goes to the share sheet, not a dead end.
Made
Product positioning
Naming and trademark screen
App Store listing: name, subtitle, keywords, promotional text, description
