Lessons Learned from a Successful AI Project – The Complete Interview with Tim Walter

Lessons Learned from a Successful AI Project – With Tim Walter

A Leadership Briefing with Tim Walter, CIO & VP, Edward Don & Company

Business Problems, Change Management, Tribal Knowledge, and AI Project Success

Guest: Tim Walter, CIO & VP, Edward Don & Company
Host: Alex Jarett, Founder, Technology Executives Club®

This Leadership Briefing combines Tim Walter's previously released Part One and Part Two discussions into a single presentation.

Many Technology Leaders are under increasing pressure from boards, executive teams, and business stakeholders to identify and implement AI initiatives. The challenge is that while interest in AI continues to accelerate, many organizations still struggle to identify the right projects, set realistic expectations, and achieve meaningful business outcomes.

In this Leadership Briefing, Tim Walter shares the lessons learned from Edward Don & Company's successful AI-driven RFP initiative, a project that dramatically increased throughput, improved customer responsiveness, and created a path toward significant new revenue growth.

Rather than focusing on AI hype, Tim explains the practical leadership, operational, and organizational principles that helped make the project successful.

Start with the Business Problem

One of Tim's most important lessons is that successful AI projects begin with a business problem, not with the technology itself.

At Edward Don, the challenge was clear. The company was only able to respond to a small percentage of incoming RFP opportunities because the process was highly manual, time-consuming, and resource intensive.

The team focused first on the business constraint and desired outcome. Only after defining the problem did they determine that AI could help solve it.

This approach helped ensure the project was tied directly to measurable business value rather than experimentation for its own sake.

Focus on ROI and Measurable Outcomes

Prior to the project, some large RFPs required five to six weeks of manual effort to complete.

By applying AI to product comparison and recommendation activities, Edward Don dramatically reduced turnaround times while increasing the number of opportunities the company could pursue.

Rather than hiring large numbers of additional resources, the organization used technology to improve throughput and better serve customers.

The lesson is simple: identify high-friction processes with clear business impact and focus AI efforts where measurable outcomes are possible.

Setting Realistic Expectations

Tim emphasizes that AI is often surrounded by unrealistic expectations.

Organizations frequently assume that AI will solve every problem immediately or operate perfectly from day one. That expectation can quickly undermine adoption and confidence.

Instead, the team focused on clearly defining what the AI would do, what people would continue to do, and what success would realistically look like.

Executive alignment, stakeholder engagement, and realistic expectations helped create the foundation for long-term success.

Change Management Matters

Technology alone did not determine the outcome of the project.

Tim explains that one of the most important success factors was helping people understand what they would gain rather than what they might lose.

Rather than positioning AI as a replacement for employees, the organization focused on how the technology would eliminate repetitive tasks while allowing employees to spend more time on customer engagement, validation, and decision-making.

This approach reduced resistance, increased engagement, and helped create buy-in across the organization.

Reinforcement Learning from Humans

A key insight from the project is that AI systems improve when they learn from the people closest to the work.

Instead of relying solely on technical teams to maintain and improve the system, Edward Don designed the platform to learn from employee interaction.

Each recommendation accepted, modified, or rejected helped improve future performance.

The result was a system that continuously evolved through real-world usage and feedback.

Capturing Tribal Knowledge

Many organizations have critical knowledge stored in the minds of experienced employees.

Customer preferences, operational exceptions, and decision-making processes often exist outside of formal documentation.

Tim discusses how the project helped capture and distribute that knowledge more effectively across the organization.

Rather than allowing information to remain isolated within individuals or spreadsheets, the system became a mechanism for sharing and scaling institutional knowledge.

This not only improved performance but also helped future-proof the organization.

Data Does Not Need to Be Perfect

One of the most practical lessons from the project is that organizations should not wait for perfect data before taking action.

While data quality matters, Tim argues that many AI projects stall because leaders believe every data issue must be solved before implementation can begin.

The Edward Don team accepted that their data was imperfect and built a system designed to adapt, learn, and improve over time.

The goal was progress, not perfection.

By focusing on one business process and continuously improving the system through usage, they were able to generate value without becoming trapped in a never-ending data cleanup effort.

Continuous Improvement Creates Long-Term Value

Throughout the discussion, Tim reinforces a central theme: AI success is not a one-time deployment.

It is an ongoing process of learning, refinement, and improvement.

Business objectives evolve. Teams learn. Data changes. Customer needs shift.

Organizations that approach AI as a continuous improvement initiative are far more likely to achieve sustainable results than those looking for a one-time implementation.

Why This Matters

Edward Don's project demonstrates that successful AI adoption requires far more than selecting the right technology.

The project succeeded because the organization focused on a real business problem, aligned stakeholders around realistic expectations, engaged employees throughout the process, and created a system capable of learning and improving over time.

The result was improved throughput, better customer responsiveness, greater organizational learning, and meaningful business impact.

Final Lessons for Technology Leaders

• Begin with a business issue, not with AI
  • Focus on measurable outcomes and ROI
  • Set realistic expectations from the start
  • Address change management head-on
  • Design systems that learn from human interaction
  • Capture and scale tribal knowledge
  • Do not wait for perfect data
  • Build for continuous improvement

This Leadership Briefing provides a practical framework for Technology Leaders seeking to improve the likelihood of success on AI initiatives while avoiding many of the common pitfalls that prevent projects from delivering value.

Learn more about Tim Walter and watch his Leadership Briefings on his Board of Advisors page:

https://www.technologyexecutivesclub.com/advisory-board-Tim-Walter

Board of Advisors Speaker: Tim Walter, CIO & VP, Edward Don & Company

Technology Executives Club®
Featuring real-world AI case studies from Technology Leaders in the Enterprise