How we grow apps with Apple Search Ads and ASO, how we build Dreamify, Viola and GERG, what we tell a brand thinking about an app, and how a small team ships. 63 posts, one new piece most days.
Voice cloning in a kids' app is a trust problem before it is a technical one. The rules we set for Dreamify: one purpose, user-owned, deletable in one tap, and never used for anything else.
How to build an Apple Search Ads keyword list when you have no history: category terms, competitor names, problem phrases, and the one source most people forget, the App Store search bar itself.
Twelve uncomfortable lessons from managing close to $100,000 a month in Apple Search Ads budget: on brand terms, Search Match, country targets, custom product pages and why installs are the wrong number.
A custom product page lets an ad land on a store page built for that exact search. How to decide which intents deserve a page, what changes between pages, and how to measure the difference.
Before we raise an Apple Search Ads budget, we check campaign structure, match types, brand defense, country-level CPA targets, custom product pages and the negative keyword list. The full checklist.
Installs are the number everyone shows and nobody can spend. The three that run a subscription business: cost per trial, trial-to-paid rate, and first renewal. How to get them, by keyword and by country.
Most apps localize for a handful of large markets and leave the rest in English. The storefronts in between often have real search volume, little competition and lower ad prices. How to pick them and what to localize first.
The AI app graveyard is full of products that were demos, not habits. What separates the apps that are still growing a year later: a ritual, a personal detail, and a reason to come back without a notification.
Ratings decide your position in search results and the conversion of your page. When to ask, how to ask, and the moments that produce five stars without a single nagging pop-up.
A video ad for an app has about three seconds before the thumb moves. What goes in those seconds, how many variants to test, and how to judge them on trial starts instead of views.
How Dreamify turns a short recording into nightly bedtime stories told in a parent's voice, how the story is personalized for the child, and what we did to keep it safe.
TikTok rewards ads that do not look like ads. What that means in practice for an app: native-feeling creative, sound on, a hook in the first second, and a product page that matches the video.
Most apps lose more users in the first ninety seconds than in the next ninety days. The seven onboarding mistakes we see most, and what to do instead.
Bidding on your own name feels like paying for installs you would have had anyway. Sometimes it is. The two situations where it is clearly worth it, and how to run it so the numbers stay honest.
A practical App Store Optimization checklist: title, subtitle, keyword field, screenshots, ratings, localization and the product page tests to run. Twenty minutes, no tools required.
Every Apple Search Ads account leaks money on searches that will never convert. The negative keyword list is how you stop it. What to add, where to add it, and the weekly routine that keeps it growing.
The App Store subtitle is the most valuable copy most developers never think about. How to use the 30 characters: keywords, promise, and the one word that should never be in it.
App Store featuring is free, enormous, and not a lottery. What the editorial team is looking for, how to use the featuring request form, and why timing your pitch to a platform feature matters.
Most people decide on your app from the first screenshot alone, in about a second. A framework for the first three frames: one promise, one proof, one reason to keep going.
Ad prices and demand move with the calendar: new year, back to school, holidays, summer. How to plan budgets around the peaks, what to do in the troughs, and the seasonal moments specific to each app category.
Most subscription apps spend months on features and an afternoon on the paywall. What the screen should say, when it should appear, and the dark patterns that cost more than they earn.
Since App Tracking Transparency, most iOS ad attribution is probabilistic or aggregated. What each channel can tell you, how to build a measurement setup that is honest about its gaps, and why first-party data matters more than ever.
Viola, GERG and Dreamify went live within three months of each other. The approach: one ritual per app, a shared technical backbone, and releases small enough to learn from.
Why a piece of cardboard with eight circles on it outperforms most loyalty programs, the psychology behind it, and what we kept when we turned it into an app for GERG.
A single price converted into every currency is wrong almost everywhere. How to set storefront prices from local purchasing power and conversion data, what to do with trials, and the markets where cheaper is not better.
Dreamify and Viola are built around listening, which makes them naturally useful to people who cannot see the screen well, as long as the controls work. What we learned testing with VoiceOver and Dynamic Type, and what changed.
A preview video autoplays in search results and on your page. Done well it shows the product doing its thing in seconds. Done badly it replaces your best screenshot with a logo animation. How to decide and what to show.
Two of our apps are used in the dark, by tired people, right before sleep. What that does to every design decision: brightness, contrast, audio pacing, timers, and the screens that should never appear at night.
Every Dreamify story gets its own picture. Generating an illustration for a freshly written story has a consistency problem, a safety problem and a speed problem. How we handle each, and what we gave up.
We do not run roadmap meetings. We read reviews, support messages and usage data every week and let patterns decide the next release. How that works in practice, what counts as a pattern, and what we ignore.
A sleep timer sounds trivial: fade the audio, stop. In practice it has to survive the screen locking, background audio rules, a fade that nobody notices and a child who wakes up if it ends abruptly. How we built ours.
Parents are tired of subscriptions and generated content costs money every night. The trade-offs we weighed for Dreamify, why it ended up a subscription, and the rules we follow to make that fair.
Every app eventually faces the temptation to add the adjacent feature. Stories plus lullabies plus a sleep tracker plus a parenting forum. Why we say no, what it costs us, and the one condition under which we add.
What we would do if we were starting an iOS app from zero today: pick a daily moment, make the first output personal, ship in weeks, and buy attention only after the store page earns it.
Most parental controls are toggles nobody touches. Dreamify has one text field: things to never mention. Why it works, how the story engine treats it, and what parents put there.
"We miss you" is a notification that admits the app has nothing to say. The rule we use for every push: it must carry something the person did not have before. Examples that pass, and the ones that get an app muted.
GERG was a client project in a studio that makes consumer apps. It changed how we think about the counter, the second user, the boring features and the difference between a download and a regular.
Some features are off the table in Dreamify no matter what they would earn: ads, links out, chat, photos of the child, manipulative paywalls. The list, the reasoning, and why a short list of refusals is a product strategy.
Personalization for a child needs a name in the story. Privacy for a child says collect as little as possible. The nickname is where those two meet, and it shaped how we ask for everything else about the child.
Three apps, all iOS. Not out of loyalty: because a small team, a subscription business and a design-led product all point the same way. The reasoning, honestly, and the conditions under which Android comes.
Not every good idea should become an app. The five questions we put every concept through, why most fail at the second one, and what a passing answer looks like.
Most new apps spend their first budget on installs. If week-two retention is weak, that money buys users who leave. A simple rule for deciding whether to spend on traffic or on the product, and what number to check first.
The affirmation app graveyard is full of pretty sentences nobody reads twice. What we changed with Viola: scenes instead of slogans, your own details, and audio built for the last minutes before sleep.
The two channels most iOS apps start with answer different questions. Apple Search Ads catches people who already want an app; Meta creates the want. How to decide, and how to run them together.
We built the loyalty app for GERG Stroop & Coffee. Here is what mattered: one QR code at the till, instant points, a stamp card people already understand, and a menu that pulls its weight.
More voices is an easy feature to add and a hard one to make good. Why Viola ships with six narrators, how we picked them, and what pacing and warmth matter more than accent.
Most App Store rejections come down to a handful of guidelines: crashes, incomplete metadata, missing purchase restore, unclear subscription terms, data use, minimum functionality, spam and moderation. What each one means and how to pass.
Tiers give regulars something a stamp card cannot: recognition. But a tier that unlocks nothing is a badge. How we structured GERG’s six levels, what moves people between them, and what each level should feel like.
A loyalty app has two users at the counter, and most teams design for one. How we built GERG’s scan flow for the person behind the till: one code, one scan, one confirmation, no exceptions to explain.
GERG’s app shows every item with its photo, price and allergen information. It is unglamorous work, and it is the feature regulars mention most. Why accurate, visible allergen data matters and how to keep it true.
Generic affirmations slide off. Viola writes short, specific scenes from inside the life you described, with your city, your work and your own details, and reads them to you.
Most order-ahead features try to let customers skip the queue, and quietly break the shop. GERG lets you prepare your cart before you arrive. Why that small difference matters at a busy counter.
Viola delivers a fresh set of affirmations every morning as cards you swipe. The format is not decoration: it forces a decision, builds a library, and keeps the morning to two minutes. The reasoning behind it.
A Wallet pass is a loyalty card that lives in Apple Wallet with no download. For many businesses it is all they need. For some it is not. The honest comparison, and the point at which a full app earns its cost.
A loyalty app is only as good as what it is connected to. The five connections that make one work: the till, payments, the menu, push notifications and analytics. And the integration brands ask for that rarely earns its cost.
The launch of a brand app does not happen on the App Store. It happens at the counter, one regular at a time. How GERG’s first month went, what staff need, what to say at the till, and the numbers worth watching.
Most small businesses do not need an app. Some do. The difference is whether there is a repeated moment at the counter that an app makes faster, and whether regulars would keep it on their phone. A short test to decide.
Brands ask what an app costs before they ask what it should do. The honest answer is that the price follows a handful of decisions: integrations, platforms, the moment at the counter, and who maintains it after launch. The drivers, without a price list.
The most common thing developers say about app success is that it is luck. It is a comforting story and a harmful one. What is actually within reach: the store page, the first session, the channel, the second week. And what luck still touches.
Every app we ship has one person who knows it end to end: design, code, store page, reviews. Why that beats specialised teams at our size, what it demands of the person, and where it stops scaling.
Frequent releases are how a small studio learns fast, and how it ships bugs to everyone at once. The habits that let us release often: small diffs, feature flags, staged rollouts, and a rule about Fridays.
Our apps ship in two languages from the first build. It is not free: every string, every screenshot, every store page, twice. What it costs in practice, what it teaches about the product, and why the second market is the one that keeps you honest.
Every story, scene, picture and narration costs money to generate, and a subscriber pays a fixed price. How we keep generation costs predictable: caching, right-sized models, limits that feel natural, and watching cost per active user weekly.
