◆ Executive Summary
Most boards begin AI conversations by asking "what should we buy?" — the wrong starting point. This framework outlines the 12 questions every leader should answer first, covering mission alignment, data readiness, governance, workforce preparation, ethics, and total cost of ownership.
Organizations that ask these questions before committing budget are measurably more likely to achieve sustainable outcomes — and far less likely to join the growing list of costly, abandoned AI initiatives.
Most boards begin their AI conversations in the same place: "What AI system should we buy?" It is the wrong question — and starting there is one of the most reliable predictors of a disappointing, expensive implementation.
The right question is: "What problem are we solving, and how will AI improve our organization's ability to fulfill its mission?" Whether the organization is a Fortune 500 corporation, a federal agency, a nonprofit, or a university, successful AI implementation depends far less on the technology itself than on strategic planning, governance, workforce preparation, and organizational alignment.
McKinsey's research confirms the scale of the problem:
1%of companies have fully mature AI capabilities
23%report meaningful cost reductions from AI
>33%see zero revenue impact from AI
The technology is not failing. The approach is.
“AI implementation is not fundamentally a technology project. It is an organizational transformation project. The institutions that benefit most are not those with the largest budgets — they are those that ask the right questions before they commit.”
— Thomas Davis, Contributing Advisor, Harshwal & Company LLP
The Complete 12-Question Framework
Each question below has surfaced — in our advisory work — as a genuine inflection point between AI implementations that succeed and those that quietly get shelved.
How the Framework Flows
Mission & Outcomes
→
Data & Infrastructure
→
Risk & Governance
→
Workforce & Ethics
→
Finance & Success Metrics
01
What organizational problems are we addressing?
AI should not be implemented because it's fashionable. Identify specific challenges where AI can genuinely improve performance or create strategic advantage.
Can you describe the problem in one sentence — clearly enough to measure if AI solved it?
02
What measurable outcomes do we expect?
Define measurable objectives before deployment. Not "improve efficiency" — a specific number, timeline, and named owner.
"Reduce close from 12 to 5 days by Q3" is measurable. "Improve reporting" is not.
03
Is our data usable, accurate, and accessible?
AI systems are only as effective as the data supporting them. Fragmented data infrastructure is the most common failure point.
Ask three departments how they define "revenue." Three answers means a governance problem.
04
Do we have the infrastructure to support AI?
Evaluate computing capacity, network bandwidth, cybersecurity protections, and integration with existing software honestly.
Post-deployment infrastructure gaps cost 3–5× more to fix than gaps caught in planning.
05
What risks does AI introduce — and who manages them?
Every deployment creates operational, ethical, legal, and reputational risk. Leadership must understand the exposure.
Risk isn't a reason to avoid AI — it's a reason to govern it like financial reporting.
06
Who will govern and oversee the AI system?
AI should never exist without human accountability — defined approvers, monitors, auditors, and escalation paths.
Governance before an incident is a safeguard. Governance after is a crisis strategy.
07
How will employees be trained — technically and ethically?
Workforce preparation is the most underinvested element of AI implementation across every sector we advise.
The goal isn't button-clicking — it's building professionals who can challenge and override AI.
08
How will AI change existing jobs and workflows?
AI rarely eliminates work entirely — it changes how work is performed. Plan for workflow redesign deliberately.
Organizations that benefit most use AI to elevate people, not reduce headcount first.
09
What ethical standards should guide AI use?
How transparent should decisions be? How will bias be monitored? What uses should be prohibited entirely?
For public-serving institutions, ethical AI isn't optional — it's foundational to trust.
10
What are the long-term financial implications?
Organizations consistently underestimate licensing, infrastructure, training, maintenance, and scaling costs.
Build a 5-year TCO model before the vendor presentation — not after.
11
Should we build, buy, or partner?
The answer depends on budget, internal expertise, security requirements, and strategic goals.
Vachi.ai exists because no commercial platform solved our clients' specific finance workflow problem.
12
How will we measure success after implementation?
Include ongoing evaluation through performance metrics, audit systems, and operational benchmarking.
If you can't answer "is AI performing as well as 6 months ago?" — you have hope, not governance.
The Bottom Line
The institutions that benefit most from AI will not necessarily be those with the largest budgets or the most advanced software. They will be those that carefully align AI systems with mission objectives, governance structures, workforce preparation, ethical standards, and long-term strategic planning.
Organizations that ask the right questions before implementation are far more likely to achieve sustainable operational improvements — and far less likely to join the growing list of costly AI failures.
❖ Key Takeaways for Executive Teams
AI implementation is an organizational transformation project — not a technology purchase.
Data quality determines AI output quality. Fix data governance before deployment, not after.
Governance, accountability, and workforce preparation are as important as the technology itself.
The hidden costs of AI — training, maintenance, upgrades, security — often exceed the license fee.
Organizations that ask the right questions before committing consistently outperform those that don't.
How Harshwal & Company LLP Helps
🏛 Advisory Perspective
Through AI Assurance frameworks, Audit in Motion, and Vachi.ai, Harshwal & Company LLP helps nonprofits, governmental organizations, educational institutions, and businesses evaluate, plan, implement, and govern AI technologies responsibly — bringing CPA-level accountability to a discipline most firms treat as a technology problem alone.
Frequently Asked Questions
The most common reason is starting with the technology rather than the problem. Organizations purchase AI platforms before clearly defining what specific, measurable outcome they are trying to achieve. Without a clear problem and success criteria, even the most advanced AI system will fail to deliver value.
Traditional IT governance was built for deterministic systems where the same input always produces the same output. AI systems are probabilistic — they generate predictions that can vary, drift over time, and introduce bias. AI Assurance requires continuous, real-time monitoring rather than point-in-time compliance checks.
Vachi.ai is Harshwal & Company LLP's AI platform built specifically for finance and operational workflows. It automates accounts payable, receivable, reconciliation, and reporting while maintaining full auditability from day one.
A responsible AI implementation typically takes 6 to 18 months for a mid-sized organization, depending on data readiness, infrastructure, and the complexity of the use case. Organizations that skip governance and workforce preparation experience higher failure rates.
Harshwal & Company LLP brings CPA-level accountability and audit discipline to AI implementation. We operationalize AI Assurance through Vachi.ai and Audit in Motion, serving clients across the U.S. and India, including nonprofits, government agencies, and educational institutions.
Considering an AI Implementation?
Schedule a complimentary AI Assurance Readiness Assessment with our team. Bring your questions — we'll tell you exactly where to start.
✓ No-obligation conversation
✓ CPA-led advisory
✓ U.S. & India coverage
About the Authors
TD
Thomas Davis
Contributing Advisor, Harshwal & Company LLP
Author of Sustaining the Forest, the People, and the Press (State University of New York Press). Decades of experience in organizational strategy, environmental policy, and AI implementation advisory.
SH
Sanwar Harshwal, CPA
Founder & Managing Partner, Harshwal & Company LLP
Leading advisory firm specializing in AI Assurance, audit modernization, and financial advisory services across the United States and India.