CLARITY · MAP OUT YOUR AI PLAN
I find your Big AI Idea.
Three weeks, run on your business. I find the highest-leverage thing AI should build for you, read the architecture standing in the way, and hand back a sequenced roadmap — so you know exactly what to build, and what it'll take, before you spend a dollar building it.
The Idea phase, run for you.
Clarity is a focused three-week engagement for teams who know AI belongs in their product but not what to build first. I work directly on your business — your product, your data, your systems, and the places AI actually has leverage — and find the Big AI Idea: the single highest-value, most-feasible thing to build, named and scored, with the reasoning a technical leader can pull apart.
It is not a brainstorm, and it is not a strategy deck. It's a worked diagnosis — seven artifacts that model your business as it runs today and as it would run with the idea live, then name the idea, the gap, and the sequence — done by someone who builds this software, not someone who only advises on it. Whatever you decide to do next, you do it knowing what you're building and why.
And it's run for you. Expect a few hours a week from you and one or two key people — the depth comes from my work between sessions, not from occupying your team's calendar.
The most expensive decision is what to build.
Clarity models the decision; Launch builds it. That line is the entire scope — three weeks aimed at what to build and whether your foundation can hold it, not at building it.
95% of AI pilots never deliver measurable ROI. Most don't fail in the build — they fail before it, because the team aimed at the wrong thing or stood it up on architecture that couldn't hold it. By the time that shows up, you've spent six figures and most of a quarter to learn it.
Clarity moves that decision to the front, where it's cheap. Seventy-five hundred dollars to know exactly what to build, why it beats the alternatives, and what it will take — against a build that can run $200K and a quarter aimed at the wrong target. It surfaces the architecture gaps now, while they cost a conversation instead of a mid-build rewrite.
It also closes the gap nothing else covers. A strategy firm can name an idea but can't read your architecture or build it. A dev shop will build whatever you hand them, right or wrong. Clarity does both halves — the aim and the foundation — because the person doing it has shipped production software, not just slide decks.
of AI pilots never deliver measurable ROI (MIT)
How It Works
Three weeks, to an answer.
Run for you, on a weekly rhythm. You and one or two key people give a few hours a week; I do the work between sessions and bring it back. Here's each week.
Week 1 — Discovery
We go deep on your business in working sessions — your product and where it's headed, the data you already hold, the systems AI would have to live inside, and the jobs your customers are actually trying to get done. This is where the whiteboard comes out: I map the territory with your team in the room, and the candidate ideas surface from what's really there, not from a list of AI trends.
End of week one: a mapped view of your product, data, and systems, and a shortlist of candidate AI ideas grounded in your business.
Week 2 — Scoring & architecture read
I score each candidate on two axes — the value it creates for the business, and the feasibility of building it on your stack — and the Big AI Idea is the one that wins on both. Then I read the architecture standing between you and building it, and bind the two models together: for every vertex where AI would sit, what the system reads for context and what it writes back as a business record. That binding is the part the platform vendors skip, and it's where most integrations quietly fail. The runners-up get documented too, with the reasoning, so the choice is defensible.
End of week two: the Big AI Idea named and scored, the runners-up documented, and the architecture gaps identified.
Week 3 — Roadmap & handback
I sequence the build into a roadmap — phases in order, dependencies, what to build first and what each phase needs — scoped to your team and your stack. Then we sit down for a readout: I walk you and your leadership through the idea, the field it beat, the gaps, and the plan, and answer the hard questions live. You leave with a document built to be forwarded.
End of week three: a sequenced build roadmap and a readout you can take straight to your board.
What you leave with is yours. Build it in-house, with me, or with anyone else — the roadmap stands on its own.
What You Get
The Read — seven worked artifacts.
The Business Process Map
You can't place AI deliberately at a vertex you can't see.
Your operation mapped as it runs today, and again as it runs with the Big AI Idea live — handoffs, approvals, exception paths, and every point where AI could sit, marked. Two maps side by side are what make the placement arguable instead of asserted.
The Software Architecture — C4
Nobody has drawn your system, so nobody can say what AI would have to plug into.
Your systems-of-record landscape at Context and Container level: what talks to what, where data lives, where the AI pieces slot in. Drawn current-state and required-state — and the difference between the two, in red, is your architecture gap. It stops at Container. Component detail is a build concern and belongs to Launch.
The Data Model — Conceptual and Logical
AI is only ever as good as the data it can actually reach.
Your entities, their relationships, the system of record for each, and quality flags on the messy ones — modelled honestly, mess included, because the mess is the readiness gap. Then the required-state model the idea depends on. Physical schema is Launch's job.
The Big AI Idea Report
A winner named on its own is an assertion. Named next to the field it beat, it's a decision.
The highest-leverage thing AI should build in your business — named, placed on a vertex, with its value thesis and its feasibility basis — followed by the runners-up, each scored on the same two axes. The criteria are set before the scoring, so your most skeptical engineer can pull the ranking apart and still land where you landed.
The Readiness and Architecture Gap
Knowing what to build doesn't tell you what has to exist before you can build it.
Required state minus current state, read straight off the models above: what's missing in the architecture, what's missing in the data, and which of it is load-bearing for the idea. Stating the gap is Clarity's job. Closing it is Launch's.
The Roadmap
A list of phases isn't a plan until the dependencies are in it.
The build sequenced by leverage and readiness — what to build first, and what has to be standing before each phase starts, with the dependency map inside the roadmap rather than beside it. Scoped to your team and your stack, and detailed enough to hand to your engineers or to any builder.
The Executive Summary
The person who approves the budget will never open the other six.
The decision written for the CEO, the CFO, the board member you have to convince: the idea in one breath, why it's right and why now, what it will take, and the sequence. Written last — its only credibility is that every claim in it is already proven a layer down, with the artifact behind it when someone asks.
You stop guessing. You know.
You walked in with a sense that AI mattered and a list of maybes. You walk out with one named, scored idea, the gaps between you and shipping it, and a roadmap you can act on — no more betting a quarter and six figures on a hunch.
You can take it to your board with evidence instead of a pitch. You can hand it to your engineers and start, bring me in to build it, or take it anywhere else — it's yours, and it stands on its own. And if it turns out AI isn't the right move right now, you'll know that too — before you spend on finding out the hard way.
Find out what to build — before you build it.
Clarity is three weeks to the Big AI Idea, the gaps, and the roadmap. It starts with a 30-minute Discovery Call — no pitch, just an honest read on whether it's a fit and where you'd start.