AI Decision Sprint / AI readiness assessment
Know exactly where AI is worth it, in two to six weeks and in writing.
Knowing what's worth building, and what isn't, is the hard part. The model never was. The AI Decision Sprint gives you that call in two to six weeks: a clear, prioritized decision you own, and the freedom to act on it with us, another team, or on your own.
The AI Decision Sprint is a two- to six-week engagement from ReasonPeak that ends in one clear, written decision on where AI is worth it in your business: what to build first, what to buy, and what to leave alone. The decision comes with the reasoning, the full cost to run it, and the order to do it in. You own the document and can act on it with us, another team, or on your own. Fixed fee from $7K.
2–6 weeksDecision document
The first call is to understand fit and scope. You are not committing to a paid Sprint by booking the call.
Why do AI projects stall?
The hard part is deciding what's worth building. Many AI projects start because a tool looks good in a demo, and the cost shows up later in a system nobody uses. The AI Decision Sprint makes that decision first. Before any build, you know where AI pays off in your operations, what to skip and what it will cost to run.
What can the Sprint look at?
One question leads; two more sharpen it.
Where AI is worth it
The core read: which problems in your operation AI fits, which it doesn't, and what to build, buy, or leave alone. This is the call the Sprint exists to get right.
The economics of running it
Costing AI the way it will run: the right mix of larger and smaller models, token costs kept in view, and an architecture you can change as the market moves. Not a sticker price; the full cost to operate.
Whether you're set up to make it succeed
A sound build still stalls if the operation around it isn't ready. We read that too (executive backing, the process and people change a rollout implies, the discipline to run AI day to day) and factor it into the call: build now, sequence it, or get ready first.
How does it run?
Two to six weeks of structured working sessions, in three moves:
Map where AI could earn its place
We work through your operation with you and surface the places AI could help: candidates worth a serious look.
Assess the candidates
Each candidate gets assessed on effort, impact, the cost to run it, and whether it holds up past a demo. An assessment, not a pitch.
Make the calls
Build, buy, or skip on each, sequenced into what to do first, what to defer, and what to leave alone, with the reasoning written down.
What we need from you: 2–5 people from your team for the working sessions, fully remote.
What's the deliverable?
One decision document, yours to keep:
- A prioritized list of where AI fits, ranked by effort and impact
- A build / buy / skip call on each opportunity, with the reasoning
- What to do first, what to defer, and what to leave alone
- The cost to run it, the key risks, and an honest read on whether you're set up to make it succeed
- Yours to act on, with us, another team, or in-house
Why trust the call?
Real economics, not a wish list
The recommendation is costed the way AI will run: the right mix of larger and smaller models for the job, token costs kept in view, an architecture you can change as the market moves. You get a number you can take to a budget, not a vision.
Engineering discipline where AI meets delivery
The call accounts for what it takes to build with AI and keep it trustworthy (review, attribution, and governance), so "worth building" means worth building and maintaining, not just worth a demo.
Model-agnostic by design
We are not tied to any vendor's partner program, so the read on what fits is based on your problem and your data, across the leading commercial AI models and platforms (Claude, ChatGPT, Gemini), open-weight models (Mistral, Qwen, Muse Glimmer), and small language models or specialized machine learning where they fit the task, not on a stack we are committed to selling.
Discovery before delivery, yours to keep
This is a paid, bounded engagement that ends in a decision you own. If the answer is to skip or wait, you get that too, before you spend on a build.
How much, how long, what form?
Fixed fee · 2–6 weeks · Decision document
The fee is fixed and agreed in writing before we start. It depends on the scope: how many processes and teams are involved, how many systems we need to look at, and which governance and regulatory requirements apply. No retainer, no open-ended scope.
All three include an AI readiness assessment, a business case, and a written recommendation for the agreed scope.
All prices in USD
Focus
One decision or one process. For example, a vendor proposal, a stalled pilot or a single workflow.
From $7,000
2 weeks
You get
A written recommendation on what to build, buy, improve, or stop, with the business case, expected costs, and next steps.
Up to 3 interviews
Business unit
One department or business line, with up to 5 priority processes.
From $12,000
2–4 weeks
You get
A prioritized AI roadmap for the team, with business cases, delivery and running costs, and an adoption plan.
Up to 6 interviews
Company-wide
Several departments, with shared priorities, governance requirements, and cross-team dependencies.
From $20,000
4–6 weeks
You get
A company-wide AI investment roadmap, with ownership, funding priorities, and a phased rollout plan.
Management interviews and workshops
If a build makes sense afterwards, we can carry it straight into delivery, since by then we already know your operation. It is scoped and priced separately.
Questions before you book
What exactly do you deliver?
One decision document you own. It lays out where AI is worth it in your operation: a prioritized list of opportunities by effort and impact, a build/buy/skip call on each with the reasoning, what to do first versus defer or leave alone, the cost, the key risks, and a read on whether you're set up to make it succeed. You can act on it with us, another team, or in-house.
How long, and how much?
Two to six weeks, depending on scope. A focused Sprint on one decision starts from $7,000, a Sprint for one business unit starts from $12,000, and a company-wide Sprint starts from $20,000. The fee is fixed and agreed before we start. If a build follows, it is scoped and priced separately.
What can the Sprint look at?
It leads with one question (where AI is worth it) and sharpens it with two more: the economics of running it (the right mix of larger and smaller models, token costs, an architecture you can change) and whether you're set up to succeed with it (executive backing, the process and people change, the discipline to run it). The result is a call you can act on.
How is this different from a strategy deck or a free consultation?
It is paid, bounded, and it ends in a decision you own, not a pitch and not a slideshow. A free consultation is built to sell you the next thing; the Sprint is built to give you an answer, including "don't build this," and the reasoning to defend it. You leave with a document, not a follow-up call.
Who is it for?
Teams facing a real AI decision with money or credibility on it: a stalled pilot, board pressure to "do something with AI," a vendor proposal that needs a second read, or several ideas and no clear order. If you need a defensible call on where AI is worth it before you spend on a build, it fits.
What if the Sprint concludes we shouldn't build?
Then that is the deliverable, and it is a good outcome. A clear "skip this" or "wait until X" saves you a build that wouldn't have paid off, and you keep the reasoning to act on later. We would rather tell you that than sell you a project. If we don't see a fit before we start, we say so.
Schedule your AI Decision Sprint
Not sure which service fits? Start with a short scoping conversation. We scope before we sell.
Email us at team@reasonpeak.com to scope an AI Decision Sprint.
Ready to find out where AI is actually worth it?
The first call is to understand fit and scope. You are not committing to a paid Sprint by booking the call.