Scattered experiments
Useful one-off prompts never become a dependable way of working.
Practical AI workflow enablement
Make AI useful in the work already on your calendar.
Bring a real report, meeting, project update, or decision. Leave with a repeatable workflow, reusable prompts, and guardrails you can use the next day.
For managers, project leaders, and operations teams using Copilot, ChatGPT, Claude, or Gemini, without hype, jargon, or a new platform.
The application gap
You have the tools. The work still feels the same.
Many leaders are experimenting with AI at the edges of their role while reports, meetings, analysis, and follow-up still run through the same manual routines. The missing piece is not another feature tour. It is a practical way to redesign one recurring workflow while keeping judgment and accountability in the right places.
Inconsistent output
Results vary because the inputs, review steps, and quality bar are not defined.
No operating habit
The tool is available, but it has not been connected to a real management cadence.
What changes
Better work, not more AI activity.
The goal is to improve how information becomes a decision, an update, a plan, or a clear next action. AI supports the workflow; it does not replace the leader responsible for the outcome.
Clearer analysis
Structure information, compare options, surface assumptions, and identify what still needs human judgment.
Stronger follow-up
Turn meetings and rough notes into decisions, owners, risks, and next actions people can use.
Reusable workflows
Document the inputs, AI steps, checks, and outputs so the process works again next week.
Relevant experience
Grounded in the work leaders are responsible for.
My coaching is informed by operational work, project delivery, and current AI enablement inside a global logistics environment. Career results below are operational track record, not claims about coaching outcomes.
Recent executive enablement
Recently delivered role-specific AI training for a senior executive, focused on practical workflows for real business work.
13+ years of operating context
Experience across purchasing, supply chain, operations, KPI work, systems rollouts, and IT project delivery.
A public working model
See how a messy project update becomes a repeatable workflow with clear human review points.
Ways to work together
Choose the scope that matches the decision.
Start with one workflow, build a manager cadence, create shared team practices, or continue with adoption support. Each option is built around work you already need to do.
AI Workflow Working Session
$395 launch price
Turn one real report, meeting, update, or decision into a repeatable AI-assisted workflow.
Start with one workflowManager AI Sprint
$1,250
Build a practical set of workflows across meetings, communication, analysis, planning, and follow-up.
Explore a manager sprintTeam Workflow Workshop
Custom quote
Give a team shared language, role-specific examples, and safe operating guardrails for its actual work.
Discuss a team workshopAI Adoption Partner
Custom quote
Turn early wins into documented practices and decide what to scale, revise, or stop.
Discuss ongoing supportResearch context
Access is common. Effective application is not.
These findings do not promise a result from coaching. They explain why access alone is an incomplete adoption strategy and why workflow design, manager support, and human judgment matter.
15%
Average productivity gain in one customer-support study
In a peer-reviewed study of 5,172 customer-support agents at one firm, access to a generative AI assistant increased issues resolved per hour by 15% on average. Results varied sharply by experience and skill.
Brynjolfsson, Li and Raymond, "Generative AI at Work," Quarterly Journal of Economics, 2025.19%
Of surveyed AI users were in the Frontier zone
Microsoft surveyed 20,000 knowledge workers who use AI at work across 10 markets. Nineteen percent were classified in the Frontier zone, where individual capability and organizational readiness were both high. The readiness measures were self-reported.
Microsoft, 2026 Work Trend Index Annual Report.How it works
Start with the work. Build the system together.
The process makes the workflow visible before deciding where AI belongs. That keeps the engagement practical and makes the result easier to reuse without ongoing dependence.
Choose one recurring task
Bring a report, meeting, update, plan, or decision that repeats.
Map the work
Separate the inputs, judgment points, output, handoffs, and risks.
Build and test
Use the AI tools you already have to create a workable first version.
Practice and verify
Test the workflow, check the facts, refine the quality bar, and keep accountability human.
Leave with a playbook
Document the prompts, steps, review habits, and guardrails so you can improve it yourself.
Worked example
From scattered project inputs to a clear leadership update.
Illustrative workflow using fictional data. This demonstrates the method, not a client outcome.
- 1Collect
Notes, milestones, decisions, risks, and open actions. - 2Structure
Separate status, evidence, implications, and what needs attention. - 3Draft
Use AI to create a concise update for the intended audience. - 4Verify
Check numbers, owners, dates, risk language, and omitted context.
Continuous improvement is the through-line. AI is the most flexible tool I have worked with for improving how people think, communicate, and execute.
Why Harold
Business context before tool enthusiasm.
I came to AI through purchasing, supply chain, operations, and project delivery, not through demos. I understand both the work inside the process and the expectations leaders have for clarity, speed, accountability, and better decisions.
I now bring that operating mindset to practical AI enablement: finding the right place for AI, defining the review points, and leaving people more capable than when the engagement started.
Read Harold's backgroundFit and readiness
Useful work needs the right starting conditions.
A short fit check protects your time. The best starting point is one recurring workflow, an approved AI tool, clear data boundaries, and a person who will review the result.
Good fit
- You lead recurring knowledge work or business workflows.
- You have an approved AI tool available.
- You can bring one specific task or deliverable.
- You want a reusable method, not a list of generic prompts.
Not yet
- Your organization has not approved an AI environment.
- Data boundaries or review ownership are unclear.
- You want AI to make regulated or high-stakes decisions for you.
- You need a generic keynote rather than hands-on workflow work.
Bring one recurring task. Decide whether it is a useful place to start.
The fit call is a focused 20-minute conversation about your role, your approved tools, and the workflow creating the most friction.