Weekend AI Project: $1,000 MRR Readiness Checklist

You have a weekend AI project that solves one screen-level task. Arc AI Screen Assistant's Android app had 50,000-plus downloads and $1,000-plus MRR by Sept. 9, and a macOS version was available.

The $1,000 monthly recurring revenue (MRR) line is a useful target because it is small enough to reach solo and large enough to cover basic tooling. The checklist below turns that target into a decision.

One screen is the whole product

Dashboards, teams, and long onboarding flows keep it a side project. Screen-level tasks are easier to support, easier to price, and easier to demonstrate. Arc AI Screen Assistant is an Android overlay that reads the current screen when triggered via accessibility services.

Good screen tasks have a before and after: the user sees a problem, triggers the app, and gets a result they can use immediately. Results that need another screen to make sense mean the task is probably too broad. The overlay pattern borrows the user's existing context. The app does not ask the user to retype a link, paste a document, or explain the situation. It works where the problem already appears.

Judge the product quickly: does it make the current screen more useful without taking over the user's day? A yes gives the project a business shape. Otherwise, spend the weekend narrowing the job.

A limited free tier proves the habit

The free tier should show the core job. It needs to be limited enough that power users feel the ceiling, but generous enough that a stranger can try the product. Arc's free tier allows seven basic-feature requests each week, with a paid upgrade option.

Tie the limit to the core action, not a vague time window. Request caps are easier to understand than trials that expire while the user is mid-task. The upgrade should remove a real ceiling: more requests, faster responses, or a harder job. A paid plan that only removes a nag will not hold.

The free tier gives users a reason to return and gives the builder a clear upgrade path. Too broad leaves revenue flat. Too thin stops the habit from forming.

Public numbers and cheap billing make the call

Solo growth is easier to judge when the numbers are public. Arc's paid plans had $1,150 in total revenue on June 13, and the build was solo. By July 5, MRR was $206. Arc earned $629 during July. On Aug. 20, Arc had 50,000 installs.

Public updates turn a private hunch into a pattern. They also create accountability. When a developer posts install counts and revenue, the next update has to be better, flatter, or an honest explanation of why it stalled. That pressure is useful. Visible traction makes the next decision easier, even without viral fame.

Costs should stay small while the project is proving itself. Chargebee, a subscription billing platform, provides a free Starter plan through the first $250,000 of cumulative billing and then charges 0.75% of billing volume. AppFollow, an app analytics tool, provides a free tier and paid plans starting at $29 per month.

Can the tool pay for itself by changing a decision? Billing dashboards that show churn by segment are worth it when you are about to change pricing. Analytics subscriptions are worth it when you are about to spend on ads. At the start, the cheapest toolset is often the best toolset. You need to know when a subscription fails, when a user cancels, and whether a new feature is being used. You do not need a data warehouse.

Five quick tests decide the next weekend

Use the list when you are deciding whether to keep the project personal or shape it into a paid product. Each item is quick to test.

It solves a screen-level task.
It can run as a low-cost overlay or workflow.
It gives away a limited free tier.
It has a premium upgrade.
It shows public install or revenue updates.

Pass every item and ship the minimal version that can become a small paid business. Fail one, and fix that piece before adding features. The next weekend should be spent on the weakest link, not on a new platform.