What is OWNML and what does it offer?

OWNML is a planning resource for machine learning projects, built around a single artifact: the Machine Learning Canvas. The premise is stated directly on the site — "Every ML system should start on one page." That page lays out what to predict, what decision the prediction feeds, and what a mistake costs, before anyone builds anything. It's aimed at teams that need to choose which ML projects to fund and agree on what each one should do. The canvas has been downloaded by 15,000+ people since 2015.

The core resource: the Machine Learning Canvas

The canvas is one page with 10 boxes. According to OWNML, it covers everything that has to be agreed before anyone builds, from what's being predicted through to what happens after the system ships. The boxes are:

• Predictions
• Data Collection
• Task
• Sources
• Decisions
• Models
• Impact
• Features
• Value Proposition
• Monitoring

The intended use is collaborative: filled in together, in one place, before the first line of code.

What running the canvas is meant to get you

OWNML describes four outcomes from working through the canvas.

Deciding which projects are worth building

The canvas forces explicit answers on what a false alarm costs, what a miss costs, and the accuracy a model would have to reach before it's worth building at all. Because those answers are on one page, projects can be compared on the same terms — and the decision can be made before anyone writes code.

A spec engineers or agents can build from

The filled-in canvas is meant to be precise enough to work from directly, rather than reconstructing requirements later from a deck and a chat thread. It also answers what an AI coding assistant would otherwise have to guess: what's knowable at decision time, and what counts as success.

Compliance documentation, built in

The canvas holds the intent, data, decision, and monitoring documentation that regulatory model reviews ask for — written during scoping instead of reconstructed months later.

Projects that reach production

OWNML cites a consultancy building anomaly detection for a US telecom manufacturer that stopped modeling after several rounds because the problem wasn't defined well enough to build against. Fixing the definition first is what got the system live. The claimed result is fewer rounds.

The problem it addresses

The site frames the core issue as disagreement that can't be shipped past. Its examples:

• Three weeks in, the team still doesn't agree what counts as churn, who acts on the score, or what a false alarm costs.
• "AI-driven churn prediction" meant five different things to the five people who approved it. It shipped, and nobody agrees whether it worked.
• Twenty AI ideas on the list, enough people to build five, and no common way to compare them.

The stated constraint: the project is yours to land, but the definition isn't yours alone to fix — the people who own those answers have to be in the room.

Who it's for

Teams that need to plan, fund, or land ML projects, particularly when scoping is contested or when multiple projects compete for the same people. OWNML also offers a free 30-minute project review on a project you're scoping, and a separate conversation for planning ML across several teams.