Where can AI improve work now?
Identify recurring tasks where existing tools can improve speed, consistency, or capacity.
AI Productivity & Readiness Review
Recommended first stepFind the AI improvements worth making—before you buy more tools or build anything.
A practical assessment for owner-led teams that need to understand where AI can improve real recurring work, what information and decisions require human control, and what to do over the next 30 days.
Typically $2,000–$4,000 · Recommended standard duration: 7–10 business days
MIT education · Former Amazon product leader · Longtime CTO · You work directly with Jason
The decision this solves
You do not need a list of AI ideas. You need to know what deserves action.
What should stay under human control?
Clarify sensitive information, approvals, escalation, and the actions that should not be automated.
Which tools are already available?
Find useful capability inside current ChatGPT, Claude, Microsoft, Google, CRM, transcription, or workflow subscriptions.
What should happen next?
Separate immediate team methods from workflow configuration, automation, integration, or custom software.
The output is not a catalog of possible AI use cases. It is a prioritized recommendation tied to your people, systems, constraints, and recurring work.
Who it is for
A good first step when AI use is real but the operating method is not.
- Owner-led businesses with approximately 10–100 employees
- Leadership teams that feel pressure to move but want to move deliberately
- Employees already using AI informally
- Recurring knowledge work across sales, operations, administration, reporting, or client service
- No internal owner for AI productivity and governance
- Interest in using current tools before commissioning software
- A need for a practical plan that can begin within 30 days
Especially useful before you
- Buy more AI licenses
- Hire an automation vendor
- Commission a custom application
- Roll out an AI policy
- Ask the entire company to adopt AI
- Choose between ChatGPT, Claude, Copilot, voice AI, or another vendor
When not to buy it
The Review is probably not the right first purchase when:
Actively disqualifying poor fits protects your time and ours. If one of these describes you, an Executive Session, a Workflow Upgrade, or a different conversation is usually the better start.
- You only want a keynote or a general AI overview
- There is no executive sponsor willing to provide access to real workflows
- You want a guaranteed ROI figure before the work is examined
- The goal is to replace staff without redesigning and validating the work
- You are seeking unsupervised automation of high-risk or irreversible actions
- You already have a mature internal AI operating team and need only narrow engineering capacity
- You cannot share even sanitized examples of recurring work
- The immediate need is a clearly defined implementation project with known requirements and acceptance criteria
What you receive
A decision package leadership can use immediately.
Executive findings memo
A concise explanation of what leadership should do now, defer, avoid, and measure.
2–4 pages
AI productivity maturity baseline
A scorecard covering current use, repeatability, leadership visibility, review practices, data boundaries, measurement, and ownership.
Supports discussion, not scientific certification
Existing-tool and license inventory
A practical map of AI capability already available in your current software and subscriptions—by tool, users, features, workflows, data concerns, and keep/expand/test/stop.
Workflow opportunity map
A prioritized matrix of recurring work by role, current burden, candidate method, expected benefit, human-review point, data sensitivity, and recommended next step.
Data-boundary and human-review guidance
Plain-language rules for what information may be used, what requires additional controls, what outputs require verification, and what actions remain human-approved.
Engagement guidance, not a compliance certification
Competitive-pressure assessment
A practical view of where a faster or lower-cost competitor could create pressure through better research, drafting, response, reporting, or coordination.
Prioritized 30-day action plan
A sequenced plan with a small number of useful experiments, named owners, success measures, and decision points.
Normally no more than three primary experiments at once
Recommended engagement path
A clear recommendation—proceed independently, run an Executive Session, launch a 30-Day Program, upgrade one workflow, investigate a custom build, establish fractional support, or do nothing yet.
“Do nothing yet” remains a legitimate outcome
Sample outputs
Show the work before asking you to imagine it.
An illustrative, operationally realistic sample based on a company archetype. Clearly labeled as an example—not a real client report.
Illustrative: 35-person professional-services firm
Executive summary
- Do now: standardize proposal drafting and client-briefing prep in Claude and ChatGPT.
- Defer: CRM automation until the drafting method is proven.
- Avoid: unsupervised client email; keep approval on all outbound.
- Measure: proposal turnaround, editing burden, and adoption across 3 roles.
Maturity baseline
- Current use: Level 1 — Experimenting
- Repeatability: Low — methods trapped with individuals
- Leadership visibility: Low
- Data boundaries: Undefined
- Target in 30 days: Level 2–3 on two roles
Workflow opportunity matrix
- Proposals · ~4 hrs/wk · Claude draft + review · High value · Med effort · Test first
- Client briefings · ~3 hrs/wk · GPT research brief · High value · Low effort · Test first
- Intake · ~2 hrs/wk · Templates + assistant · Med value · Low effort · Next
- Reporting · ~2 hrs/wk · Assisted summary · Med value · Med effort · Later
Human-review & data boundaries
- Client-identifying data: sanitized before model use
- Outbound communication: human approval required
- Legal/financial figures: verified against source
- Vendor terms: reviewed before any sensitive workflow
30-day plan
- Week 1: baseline + role selection (owner: managing partner)
- Week 2: proposal + briefing methods on real work
- Week 3: review rules, templates, and a shared project
- Week 4: measure turnaround and adoption; decide next step
Illustrative example. A downloadable sample report can be provided on request; opens are measured as a conversion-assist event, not gated behind a form.
How the Review works
A short assessment, not an open-ended consulting project.
Before
- Confirm goals, scope, roles, and data constraints
- Gather a short tool and license inventory
- Select representative workflows and documents
- Schedule leadership and role interviews
- Establish what success would make the Review worthwhile
During
- Leadership kickoff
- Interviews or working sessions with selected roles
- Review of recurring tasks and representative inputs and outputs
- Demonstrations or small tests using appropriate current tools
- Assessment of human-review points, data boundaries, and operational risk
- Prioritization by value, feasibility, frequency, and effort
After
- Findings and recommendations assembled
- Leadership readout
- Final deliverables provided
- Questions and revisions within an agreed review window
- Decision on the next 30 days
Recommended standard cadence
- Day 1: Kickoff
- Days 2–5: Interviews, workflow review, and tool analysis
- Days 5–7: Small demonstrations or tests where useful
- Days 7–9: Findings and plan
- Day 10: Leadership readout
This is a proposed operating model, confirmed against delivery capacity. Timelines are described as “typical” and scope is defined in writing.
What Jason needs from you
The Review works when leadership provides access to real work.
- One executive sponsor
- A 45–60 minute kickoff
- Access to two to five representative roles, depending on scope
- Examples of three to eight recurring workflows
- Existing policies or constraints relevant to data and tool use
- A list of major software and AI subscriptions
- Sanitized or approved examples of documents, reports, messages, or tasks
- Honest information about where work slows down, gets repeated, or gets dropped
- Participation in a final leadership readout
- Timely answers to scope questions
The Review does not require unrestricted access to company systems. Access and examples are minimized to what is necessary, and sensitive-data handling is agreed before work begins.
Scope and price
Typically $2,000–$4,000, based on the operating surface being reviewed.
The range is explained, not arbitrary. Scope is set in a short written document before work begins.
Price is affected by
- Number of roles
- Number and complexity of workflows
- Number of systems and licenses
- On-site versus remote work
- Sensitive-data requirements
- Number of interviews
- Need for demonstrations or small proof tests
- Depth of executive reporting
Focused review — near the lower end
- One leadership sponsor
- Two roles
- Three representative workflows
- Remote delivery
- Standard findings package
Standard review — middle of range
- Leadership team plus three to five roles
- Five to eight workflows
- Tool and license inventory
- Small demonstrations where useful
- Full findings package and readout
Expanded review — near the upper end
- Multiple teams or locations
- More complex data or system boundaries
- On-site work
- Broader competitive or workflow analysis
- Additional stakeholder readouts
These are proposed examples, not rigid packages. Scope follows the work.
Risks, boundaries, and what is not included
Clear boundaries make the recommendations more useful.
- Not a legal, regulatory, security, or compliance certification
- Does not guarantee a specific return on investment
- Does not include a production custom application unless separately scoped
- Does not include unrestricted access to client systems
- Does not authorize autonomous external communications or irreversible actions
- Recommendations depend on current vendor capabilities, permissions, data rules, and platform terms
- AI outputs require review appropriate to the task
- Sensitive workflows may require approved vendors, contractual protections, or additional controls
- Implementation beyond light demonstrations is separately scoped
- Client leadership remains responsible for policy, personnel, legal, and operational decisions
The purpose is to make the next decision smaller, clearer, and better supported—not to pretend that every uncertainty disappears.
Why The Tomorrow Machine
Senior judgment and hands-on implementation in one person.
You work directly with Jason Merkoski from kickoff through the leadership readout. Jason combines an MIT education, early Amazon product leadership, long experience as a CTO and builder, and current hands-on work across ChatGPT, Claude, voice AI, supervised browser automation, integrations, and production systems.
That breadth matters in a Readiness Review because the recommendation does not have to force every problem into a workshop, one software vendor, or a custom build.
- Product and operating judgment, not only tool demonstrations
- Hands-on ability to test what is feasible
- Enough engineering depth to know when not to build
FAQ
Questions buyers ask first.
No. The Review may include demonstrations, but the product is a practical assessment and action plan tied to real recurring work.
No. Tool selection follows the workflow, data, controls, cost, and fit.
Yes, where those products are relevant to your current licenses and work.
Potentially, with explicit decisions about access, vendors, storage, logging, agreements, and human review. Sensitive examples can often be minimized or sanitized during assessment.
Yes. One purpose of the Review is to separate needs that can be handled through better methods or configuration from those that may justify integration or custom development.
You may execute the plan independently, run a 30-Day Program, upgrade one workflow, commission a qualified build, or use fractional support. No follow-on engagement is required.
A useful initial estimate is one kickoff, selected interviews or working sessions, light material gathering, and a leadership readout.
The Executive Session is better for a focused leadership question or first experiment. The Review is broader and produces a company-specific baseline, opportunity map, boundaries, and 30-day plan.
The public range is typically $2,000–$4,000. Final scope depends on roles, workflows, systems, data constraints, and delivery format.
Where this can lead
The Review points to the smallest useful next step.
Executive AI Productivity Session
A smaller, lower-friction alternative when you have one focused leadership question.
See the Executive Session30-Day AI Productivity Program
The likely next step once roles and workflows are chosen and you want measurable adoption.
See the 30-Day ProgramWorkflow Upgrade
When one recurring workflow is already understood and worth improving now.
See the Workflow UpgradeDiscuss a Readiness Review
Leave with a smaller number of better AI decisions.
Bring the recurring work, tools, or competitive pressure on your mind. We will determine whether a Readiness Review is the right first step and define the smallest useful scope.
Free 20–30 minute AI Fit Call · No sales team · You speak directly with Jason
Prefer email? Reach Jason directly at hello@tomorrowmachine.ai.
Based in Santa Fe, New Mexico. U.S. Mountain and Pacific working hours.
Book an AI productivity call
Pick a time for a free 20–30 minute AI Fit Call. We will look at how your team works today, where existing AI tools could help, and the smallest sensible next step.
What happens on the call
- 1.You describe the recurring work, pain point, or competitive pressure on your mind.
- 2.We decide together where existing AI tools could help—or whether a simpler fix comes first.
- 3.If there is a fit, we identify the smallest sensible next step: an Executive Session, Readiness Review, or 30-Day Program.
- 4.No sales team, no pitch deck. You speak directly with Jason.
Useful to bring
- The recurring work or decision you want to improve
- The AI tools and software your team already uses
- What would make a first step worth doing