Paid engagement

AI Opportunity Assessment

A structured engagement that examines where AI could genuinely help your organization, compares the options by value, complexity, and risk, and recommends a first workflow to build—or explains why nothing should be built yet.

It is not the free AI Opportunity Call: the call decides fit, while the Assessment is the paid work that produces the decisions, the recommended pilot, and the written roadmap. It is the paid offer inside the Find the Right Opportunity stage, and the services page compares the two conversations.

Who the Assessment is for

The Assessment fits organizations that have real work to improve and the willingness to change how it is done. It is a poor fit for anyone looking for a demonstration or a guarantee.

A good fit when

  • Work is spread across Microsoft 365 and one or more business systems, and people move information between them by hand.
  • You can name a workflow that consumes real time or attention, even if you cannot yet say what it costs.
  • Someone with authority over that workflow will take part in the assessment.
  • You treat permissions, privacy, and human approval as requirements rather than obstacles.
  • You want a recommendation you can act on, and you are prepared to hear that the answer is no.

Not the right step when

  • You want a product demonstration or a vendor comparison rather than a decision about your own work.
  • The goal is to apply AI broadly, with no defined workflow and no one accountable for the result.
  • Guaranteed savings or autonomous replacement of employees is the expectation.
  • Formal legal, regulatory, or compliance certification is required. The Assessment reviews security considerations; it does not certify anything.
  • Unlimited strategy work is expected from the initial call instead of a paid engagement.

Problems it resolves

These are the situations that make an assessment worth paying for rather than guessing.

No obvious first project

Several ideas look promising, and no one can rank them without arguing about tools.

Unclear feasibility

Nobody knows whether the information a workflow needs is accessible, correctly permissioned, and current enough to be useful.

Risk discovered late

Access, retention, and approval questions surface during implementation, when changing direction is expensive.

A pilot that proves nothing

Work starts without success criteria, so afterwards nobody can say whether it worked or what should happen next.

What is examined

Every assessment covers the same six areas. How deeply each one is examined depends on the scope agreed before work starts.

Stakeholder interviews

Short conversations with the people who do the work and the people accountable for its result, covering what happens today, where it breaks, and what they would consider an improvement.

Workflow review

How each candidate workflow actually runs, including the manual steps, the rework, and the workarounds that never appear in a process document.

System and data inventory

Which systems hold the information, how it is structured, who can already reach it, and where a workflow would have to cross a boundary.

Access and security review

What each candidate would need to read or change, how that maps to least privilege, what would be logged, and where a person must authorize the outcome.

Value and feasibility analysis

Time returned, error reduction, and service quality weighed against integration complexity, data readiness, and the effort required to maintain the result.

Ownership and adoption reality

Who will operate the result, what they will need to do differently, and whether the organization is prepared for that change.

How opportunities are prioritized

Prioritization is a method, not an opinion. Each candidate is put through the same four steps so the ranking can be reviewed by people who were not in the interviews.

  1. Describe the work as it happens today

    Each candidate is written down as a sequence of real steps and decisions rather than as an idea or a tool preference.

  2. Weigh value against effort

    Expected benefit is compared with integration complexity, data readiness, and the ongoing cost of keeping the result working.

  3. Test the constraints

    Candidates that cannot meet least privilege, approved data handling, or the required human approval are marked as such, or reshaped until they can.

  4. Rank and recommend

    You receive a ranked list with the reasoning attached, a recommended first pilot, and the conditions that would change the ranking.

What the ranking weighs

Value
Time returned to the team, fewer errors in handoffs, and better service to the people the work exists for—described as a mechanism rather than as a promised number.
Complexity
How many systems are involved, how ready the information is, and how much of the result must be maintained after launch.
Risk
How sensitive the information is, what a workflow would be allowed to change, whether a person stays accountable, and how reversible an error would be.

What you receive

The Assessment ends in written work you can act on without EverythingAI in the room.

A prioritized opportunity list

Each opportunity carries its value, complexity, and risk reasoning, so the ranking can be defended internally after the engagement ends.

A recommended first pilot

One workflow, with success criteria that can be observed in normal work rather than asserted after the fact.

A written roadmap

Dependencies, approval requirements, sequencing, and the constraints each step has to respect.

Documented findings and unknowns

What was examined, what was assumed, and what could not be confirmed with the access available.

A recommendation on whether to proceed

Proceeding is never assumed. When the evidence does not support a pilot, the recommendation says so and explains why.

Security and privacy considerations

The Assessment is where access and data-handling requirements become concrete for your organization, before anything is built. Each candidate workflow is reviewed against the access it would need, the data it would touch, what would be recorded, and where a person must authorize the outcome.

Nothing sensitive is required to run the Assessment. Interviews and documentation review come first, read-only or non-production access is used only where the agreed scope requires it, and credentials are never sent by email. Findings describe data flows, permissions, and approval points rather than reproducing confidential content.

The Security Constraint, documentation, and human-authorization conditions that apply to every engagement—including this one—are stated once on the services page.

What the client provides

The output is only as good as the access and the information behind it. These are the responsibilities that sit with the client during the engagement.

Access to the right people

Short conversations with the people who do the work, the people accountable for the outcome, and whoever administers the systems involved.

Accurate information

Honest detail about systems, data, licensing constraints, and the parts of the process that are undocumented.

A decision-maker in the loop

Someone who can confirm priority and scope, so the roadmap reflects the organization rather than a wish list.

A safe review path

Synthetic records or a non-production path whenever a workflow touches real records. Production access is not a prerequisite for the Assessment.

Timely feedback

Prompt review of interview summaries and draft findings so the analysis stays based on current information.

If access to stakeholders or accurate system information is not available, the Assessment says so in the findings and limits its conclusions accordingly. It does not fill the gaps with assumptions presented as fact.

What an Assessment can recommend

Four recommendations are possible. All four are legitimate outcomes, and all four arrive with the reasoning written down.

Proceed to a pilot

The common outcome: one workflow is worth building first, with clear success criteria and known constraints.

Fix a prerequisite first

Sometimes identity hygiene, information quality, or an ownership gap has to be addressed before AI work would hold up.

Narrow the scope

The idea is sound but too broad for a first step. The roadmap proposes a smaller version that can still prove the value.

Do not proceed

Sometimes no candidate justifies the work yet. That is a valid, documented outcome of a paid engagement, and it includes what would have to change.

What happens after the Assessment

  1. You receive the written output

    The prioritized list, the recommended pilot, the roadmap, and the documented unknowns, in a form you can circulate internally.

  2. You decide

    Nothing proceeds automatically. The recommendation is yours to accept, defer, or decline.

  3. If a pilot goes ahead

    A fixed-scope AI Workflow Pilot implements one workflow against the agreed success criteria before anything broader is attempted.

  4. If it does not

    The written findings stay useful on their own: they record what was examined and what would have to change for AI work to make sense later.

Questions about the Assessment

Is the Assessment free?

No. It is a paid engagement, and scope, schedule, and commercial terms are agreed in writing before work starts. The free conversation is the AI Opportunity Call, which decides fit.

How long does it take?

It depends on how quickly the right people are available and how well the systems are documented. The schedule is agreed before work starts, and progress depends on the access described above.

What if there is no worthwhile use case?

Then the Assessment says so. You still receive the ranked comparison, the analysis of what was examined, and a written statement of what would have to change. A recommendation not to proceed is as valid an outcome as a pilot recommendation.

Do we have to provide access to production systems?

Not for the Assessment. Interviews and documentation review come first. Where a workflow has to be examined against real configuration, read-only or non-production access is agreed in scope, and credentials are never sent by email.

Is this a security audit or a compliance certification?

No. The Assessment reviews the access, data handling, logging, and approval requirements that the candidate workflows depend on. It does not certify compliance, and it does not replace legal or regulatory advice.

How is this different from a product demonstration?

It starts from your workflows and outcomes rather than from a tool. If a product is recommended later, it is because the workflow requires it—not the other way around.

Does the Assessment include implementation?

No. Implementation is a separate, fixed-scope engagement. The Assessment ends with a ranked recommendation, a recommended pilot, and a written roadmap.

What do we keep at the end?

The written roadmap and findings are the deliverable of the engagement, written to be usable whether or not you continue with EverythingAI.

Start with the free conversation

Bring the workflow that costs you the most time. The call decides whether an Assessment is the right next step for your organization.