360° AI Blueprint
The 360° AI Blueprint is a diagnostic roadmap across your entire organization. It pinpoints where performance leaks, where AI creates the biggest impact, and what each initiative returns — quantified and sequenced from quick wins to long-term moves. It's where we start when the starting point isn't clear; if you already have a specific project in mind, we jump straight in.
The 360° AI Blueprint is a company-wide AI opportunity assessment: the engagement to choose when you have more candidate AI ideas than certainty about which to fund. It assesses the operation function by function, ranks AI and automation candidates by ROI, cost, and time-to-value, and sequences them into a roadmap. It does not build systems, and a good Blueprint will tell you where AI is the wrong answer.
On this page
At a glance
- For executives with more candidate AI ideas than certainty about which to fund.
- Assesses the operation function by function and ranks AI and automation candidates by ROI, cost, and time-to-value.
- Matches each candidate to the simplest tool that works, which is often not AI.
- Ends in a sequenced roadmap; building anything is a separate, scoped implementation.
Who the Blueprint is for
Many organizations weighing AI are not short of ideas. They have a list: a vendor demo someone liked, a pilot a department started, a board member's question about what competitors are doing.
What they lack is a defensible order. Which of these ideas would actually move revenue or cost? Which would stall on data, access, or change management? And which should be solved with ordinary software, or not at all?
The Blueprint exists for that situation. It is built for an executive sponsor who has to decide where the next AI budget goes, and for the functional leaders who will live with the result. It usually earns its place when at least one of these is true:
- Several departments are proposing AI projects and there is no common way to compare them.
- A first pilot worked technically but moved no number anyone reports on.
- Leadership wants an AI plan for the whole organization, not a single use case.
- The cost of manual work is felt everywhere but has never been measured in one place.
Scope and exclusions
The Blueprint covers the functions agreed at the start, which can be the whole organization or a defined set of departments. Its output is a decision document: what to build, in what order, with what expected return, and what not to build.
- It does not build or deploy anything. Implementation is a separate engagement, scoped from the roadmap.
- It is not a legal, regulatory, or security audit. Where a candidate raises compliance questions, the roadmap flags them for your own counsel or compliance team.
- Its estimates are not measured results. A modeled return is a planning figure. It becomes a result only after a system runs and is measured against the baseline.
The scope and exclusions above describe how we typically run a Blueprint. We don't publish engagement prices; scope and fees are set out in a written proposal before work begins. Final scope, deliverables, and terms are set in the written engagement agreement.
What you receive
The Blueprint produces four deliverables. Each is described below in the words we use for the offer itself.
- Precise identification Pinpoint the inefficiencies draining profit — and convert them into growth opportunities.
- Prioritized AI portfolio The initiatives worth building: AI and automation candidates ranked by ROI, cost, and time-to-value.
- Sequenced ROI roadmap The order to build them in: quick wins first, compounding into measurable EBITDA gains.
- Department-level recommendations Specific moves for each function, with expected impact on top-line and bottom-line performance.
We write recommendations so that someone who wasn't in the room can check them. Each one names the workflow it concerns and the baseline it starts from. It also states the assumptions behind its estimated return, and what would have to be true for it to succeed. Where the evidence is thin, the roadmap says so rather than rounding the uncertainty away.
How the assessment works
The core of the Blueprint is an operational assessment. On our services page we describe it as a rigorous assessment of your operation: where hours, errors, and dollars concentrate — and which processes justify a production agent system versus a point tool.
In practice that means looking at workflows one at a time, with the people who run them. Each candidate is scored on the same criteria so they can be compared across departments.
| Criterion | What we look at | What pushes a candidate down the list |
|---|---|---|
| Volume and frequency | How often the work happens and how long each unit takes today | Low or sporadic volume, where automation never pays back its upkeep |
| Cost of errors | What a mistake costs downstream: rework, write-offs, compliance exposure, customer impact | Errors that are cheap and easy to catch by eye |
| Rule clarity | Whether the decision rules are written down, stable, and agreed | Rules that live in one person's head or change every month |
| Input quality and access | Where the inputs come from, their format, and whether we can read them | Data that can't be exported or shared, or that is mostly missing |
| Decision authority | Which steps need a human to approve, and who that human is | High-stakes, irreversible decisions with no one available to review them |
| Integration surface | Which systems the workflow touches and how they can be read and written | Systems with no export, API, or file interface, where access is the whole project |
| Measurability | Whether a baseline exists, or can be measured, before anything is built | No way to tell afterwards whether the change worked |
Matching each candidate to the right tool
The same scoring decides the tool, not just the priority:
- Deterministic automation This, or a configuration change in software you already own, is usually the better fit for a workflow with clear, stable rules and structured inputs.
- An agent system with human review This may be justified for a workflow that needs judgment on messy documents.
- An upstream fix This is the right call when the problem is really a form that collects the wrong field, or a policy nobody enforces. Those get a recommendation too, even though it isn't AI.
How returns are estimated
Each candidate's return is modeled from the operation's own numbers, not from industry averages. The model weighs:
- the volume of the work, and the time and loaded cost per unit;
- the cost and frequency of errors;
- any revenue effect, such as faster billing or fewer lost opportunities;
- against all of that, the cost of building, running, and maintaining the system, including the human review time it will still need.
Where a figure can't be measured, it's recorded as an assumption, with the range we used and who supplied it.
If you want to see the method before engaging, our guide on how to calculate AI ROI before you build walks through the same structure.
What the engagement looks like
The Blueprint is the first stage of our delivery process, Diagnose: A focused operational assessment. We quantify the cost of the status quo and rank automation candidates by return. Within that stage, the work typically follows this sequence; the exact steps for your organization are agreed at the start.
- Agree the scope
Confirm which functions are in scope, who sponsors the work, who owns each workflow, and what data can be shared and how. Access is agreed with your IT and security team before any data moves.
- Map the operation
Interview the people who run each workflow and look at the actual artifacts: exports, reports, queues, spreadsheets. The goal is where time, errors, and money concentrate, not an org chart.
- Quantify candidates
Build a baseline for each candidate workflow and estimate what changing it would return and cost, with every assumption recorded.
- Rank and sequence
Score candidates on the criteria above, decide the right kind of solution for each, and order them so early work builds the data, access, and confidence later work depends on.
- Review with leadership
Walk the sponsor and functional leaders through the roadmap, challenge the assumptions together, and leave with a plan leadership can fund or reject item by item.
What we typically need from you
- An executive sponsor who can act on the findings. Without one, the Blueprint becomes a report.
- Functional leaders and process owners for each department in scope, available for interviews and for reviewing the numbers about their own area.
- Access to operational data, usually as exports or read-only access agreed with IT. We don't need production write access to do a Blueprint.
- A contact in IT or security who can answer questions about systems, access, and data-handling requirements.
Ownership and what happens next
The roadmap is written so your team can act on it whether or not we build any of it. When a prioritized item is ready to build, it becomes a scoped implementation with its own acceptance criteria.
Our use-case pages show what that looks like for specific workflows, for example academic scheduling optimization, insurance invoice analysis, and financial reconciliation automation. You can browse all of them on the use cases page.
What you can check before you engage
This page doesn't cite client results or outcome figures. You can judge the Blueprint on things you can check before paying for it: the criteria it uses, the way it estimates returns, what it excludes, and the terms below.
Terms
For the reasoning behind the guarantee, see why we put a money-back guarantee on the Blueprint.
Before you book
Three resources help you judge fit before you book:
- The AI readiness assessment: twelve questions worth answering first.
- Our guide to the 360° AI Blueprint: the approach in more depth.
- Questions to ask an AI consulting firm: for comparing firms, and they apply to us too.
Common questions
Do we need a Blueprint if we already know which project we want to build?
Usually not. The Blueprint is where we start when the starting point isn't clear; if you already have a specific project in mind, we jump straight in. It's most useful when you have several candidates and no agreed way to rank them.
Is there a money-back guarantee on the Blueprint?
Yes. If you don't walk away from the 360° AI Blueprint with quantified opportunities to increase revenue and reduce costs — plus a clear AI roadmap across the organization — we refund your investment.
How long does it take, and what does it cost?
Both depend on which functions the work covers and on what data your teams can share, and in what form. They're set out in a written proposal before any work starts. We don't publish prices.
Will the Blueprint recommend AI for everything?
No. Each candidate is matched to the simplest thing that works. That is often deterministic automation, a change to software you already have, or a process fix, and the roadmap says so.
Who should be involved from our side?
An executive sponsor, the leaders of the functions in scope, the people who actually run the workflows, and someone from IT or security who can speak to systems and data access.
How is the Blueprint different from a free AI readiness assessment or quiz?
A quiz or checklist gives you a maturity score. The Blueprint is a company-wide AI opportunity assessment: it interviews the people who run each workflow, scores real candidates against the criteria above, and returns a sequenced, fundable roadmap — not a category label.
What's the difference between an AI readiness assessment, an AI strategy and an AI roadmap?
As we use the terms: a readiness assessment checks whether you can start at all — sponsorship, data, and systems access. A strategy sets direction. The Blueprint is the roadmap itself: the ranked, sequenced list of candidates worth building, with the return and cost behind each one.
How do you prioritize AI use cases?
Every candidate is scored on the same criteria — volume and frequency, cost of errors, rule clarity, input quality and access, decision authority, integration surface, and measurability — then ranked by modeled ROI, cost, and time-to-value, and sequenced so early work builds the data and confidence later work depends on.
What does the final deliverable look like?
Four things, in the words we use for the offer: pinpointing the inefficiencies draining profit and converting them into growth opportunities; a prioritized AI portfolio ranked by ROI, cost, and time-to-value; a sequenced roadmap starting with quick wins; and department-level recommendations with expected impact.
Do we have to use Foresight to build what the Blueprint recommends?
No. The roadmap is written so your team can act on it whether or not we build any of it. If you want us to build a prioritized item, that becomes a separate, scoped implementation with its own acceptance criteria — never an obligation created by the Blueprint itself.
Capabilities
- AI agent opportunity audits
- C-suite AI agent strategy
- Agent-powered workflow design
- Dashboards, audit trails, and monitoring
How we work
- 01AI Agent Opportunity Audit
- 02Workflow and systems diagnosis
- 03Agent architecture and implementation roadmap
- 04Prototype or MVP
- 05Production deployment
- 06Monitoring, improvement, and expansion
Who this is for
Best fit
Companies where the C-suite knows AI matters, but needs a practical path from strategy to deployed agent systems.
Not a fit
- Generic chatbot requests
- Isolated AI experiments
- Companies looking for AI theater without operational ownership