Notebook · 29 May 2026 · 5–6 min read · Alpesh Patel

What we won’t build.

The list of things Keepwright has decided against is shorter than the list of things we’d like to build. Both lists matter equally. Saying no to the wrong work is how a small studio gets to ship the right work.

Why we publish the no-list

Most software companies don’t tell you what they won’t build. They don’t want to commit. The roadmap is a moving target, and “we never said we wouldn’t” is a useful escape hatch for the year you decide to pivot.

We’d rather close the escape hatch. The point of a pay-once studio is that the deal you bought from is the deal you keep. If we let ourselves drift into ad-funded apps, account-required logins, “premium” tiers added after launch, or a subscription pivot dressed up as an upgrade — we’d be a different studio. You’d have bought from one company and ended up using another.

So here are the categories we’ve committed to never enter. Each one is a no because of something specific.

The list

• A subscription business. The whole reason we exist is because we believe small mobile utilities should be one-time purchases. If we couldn’t make an app work as pay-once, we wouldn’t ship it — we’d quietly drop the idea and pick the next one. We will never “move to a subscription model” for an app you already bought. The price you paid is the price for as long as the app runs.
• Free with ads. Ad-funded apps are surveillance with a friendly UI. Each “free” install loads a third-party ad SDK that profiles the user’s behaviour across other apps to target what gets shown next. There is no version of this that’s compatible with a private-by-design promise. Even if it would make us money. Especially because it would make us money.
• A cross-app account system. No “Sign in with Keepwright”. Each app is independent. Your Mileage Tracker doesn’t know about a hypothetical future logbook, doesn’t share state with it, doesn’t federate any identity. Building an account system is how studios start collecting the data they swore they wouldn’t. We’re removing the temptation by removing the mechanism.
• A two-sided marketplace. Marketplaces are platforms. Platforms work at scale, on retention, with the network effect doing the heavy lifting. We build tools. The two business models are fundamentally different shapes; trying to be both is how a small studio dies of indecision.
• A behaviour-analytics shop. No Mixpanel. No Amplitude. No Firebase Analytics. No PostHog. No Heap. The only third-party SDK in a Keepwright app is Sentry, and Sentry is configured to strip every personally identifiable field at the SDK initialisation layer so crash reports don’t carry your data. We will not be reading event logs of what features you used last Tuesday. That data isn’t collected, so it can’t be leaked, sold, subpoenaed, or used to build the “engagement” loops we’re trying to avoid.
• A pivot to “the cloud”. The current generation of pay-once apps that got abandoned by their original developers usually died via this exact move — the “new version” required an account, then a cloud back-end, then a subscription to fund it. We’ve seen the playbook. If we ever introduce optional cloud features, the trade-offs are made plain at the toggle, the offline path stays first-class, and the existing purchase still works without any account.
• Selling or sharing your data. Not now, not ever, not in any partnership, not under any acquisition that asks for it. The studio brand depends on this being true forever; the day it isn’t true is the day the brand is over.

The deeper principle

You can read this list as a series of moral positions if you like. We’d rather you read it as a set of operational consequences of being small.

A small studio cannot out-compete a Series-B-funded startup on engagement loops, on personalisation algorithms, on multi-cohort analytics dashboards, on customer success programs, on conference booths or paid celebrity endorsements. We don’t have the people. We don’t have the capital. We don’t have the headcount.

What we have is a small group of people who’d rather pay once for a tool that does its job, and a willingness to charge them once for a tool that does its job. That’s a coherent business if — and only if — we refuse to play the games that need a bigger studio.

What this commits us to

The list above commits Keepwright to a small handful of design decisions that don’t change. If we ever found ourselves discussing how to relax one of them in a meeting, that’d be the moment to ask whether we’re still the company we said we were.

• The apps will keep working offline, with no account required.
• The price will stay one-time. Updates within an app stay free for the original buyer.
• The data will stay on the device by default.
• The third-party SDK list will stay one-item-long (Sentry, PII-scrubbed).
• The free tools on the website — the Mileage Calculator, the future ones — stay free, without an email field, forever.

If those commitments become impossible to keep, the right answer isn’t to weaken them. It’s to say so plainly and let the people who bought from us decide what to do with the news.

Previous: Why pay-once works at our scale → · Back to Notebook