Blog
December 5, 2025
A PATH Toward Purpose, Feasibility, and Real Impact
Across higher education and the nonprofit sector, leaders are wrestling with the same tension: AI is moving faster than our organizations are built to adapt. In a recent discussion with university and nonprofit executives, we saw this challenge play out firsthand. When asked to physically choose between bold, organization-wide AI initiatives or small, incremental pilots, participants split almost evenly. Not a single scenario produced consensus.
This divide reflects a broader truth: mission-driven organizations are eager to leverage AI, but lack the frameworks needed to do so responsibly, quickly, and in direct alignment with mission. Traditional technology evaluation and procurement processes simply weren’t built for the pace of AI innovation. And without a reimagined approach, organizations risk investing in tools that are outdated before they’re even deployed.
The Procurement Problem: AI Moves Faster Than Your RFP Timeline
In many universities and nonprofits, evaluating and selecting major technology solutions can take 60–90 days. But in the world of AI, a 90-day selection cycle might as well be a decade. By the time an RFP is completed, the “best” tool may already be obsolete or surpassed by a more capable model or agent-based workflow.
This mismatch creates real operational and financial risk. Organizations are making long-term decisions using processes optimized for legacy technology, not AI, which evolves at near-exponential speed.
To address this challenge, we need a new method for assessing potential AI investments, one that prioritizes mission, feasibility, experimentation, and risk-aligned innovation.
Introducing the PATH Framework
To support leaders navigating this complexity, Attain Partners developed the PATH framework: a practical approach to evaluating AI opportunities across four dimensions, each with three actionable tactics.
P: Purpose Alignment
Begin every AI conversation with mission, not technology. Organizations frequently pursue AI because the market, vendors, or internal excitement suggest they “should.” But the first question must always be: Does this initiative advance our mission in a meaningful, measurable way?
This leads directly to a core concept we discussed at length at our fall Leadership Symposium in Nashville: Return on Mission (ROM). AI initiatives shouldn’t be justified by novelty or assumed efficiency. They should be judged by their contribution to outcomes like improved student success, increased donor engagement, or greater operational resiliency.
A: Assess Feasibility Before Impact
Unlike traditional design thinking (which balances feasibility and impact), AI requires a different approach. Many use cases that appear similar on paper vary dramatically in technical feasibility. Two solutions with only minor conceptual differences may sit on opposite ends of the “can AI actually do this?” spectrum.
As a result, organizations must evaluate feasibility first. Start with what AI can reliably do today, not what you hope it might eventually do.
T: Test Fast, Learn Quickly
Generative AI is a hands-on technology. The fastest way to understand its value is to pilot quickly, learn aggressively, and identify natural “stop lines.” These stop lines function as predetermined decision points: clear definitions of success and failure established before any investment begins.
If a pilot meets the success metric, then continue and scale. If not, stop. No sunk costs. No endless iteration.
H: Harmonize Risk with Use Case
AI maturity isn’t measured in years. Vendors claiming “a decade of AI experience” are often referencing outdated machine learning techniques, not today’s generative models. Counterintuitively, newer, less entrenched vendors may be more technically advanced than long-standing providers.
What matters most is not the age of the tool but the risk exposure of the use case. Many AI applications pose minimal data risk and can be safely piloted using modern, high-performing public models. In other cases, such as system operations or sensitive data workflows, a more guarded approach is required.
AI evaluation must shift from, “Which vendor is most mature?” to, “What level of risk does this use case truly pose, and what model performance does it require?”
A New Mindset for AI Experimentation
Much like the early days of Excel, ChatGPT, Copilot, and other AI agents represent a leap in capability, but only for those willing to build fluency. The organizations that learn, test, and adapt today will see compounding returns tomorrow. Small wins stack quickly. Over-engineering or over-governing, however, can stall progress entirely.
As one CIO wisely put it during our Symposium: “If you’re stuck in governance, you’re doing it wrong.”
Identifying low-risk, high-feasibility use cases, and piloting them fast, is the path forward. The future belongs to organizations that can navigate ambiguity with clarity of mission, clarity of success criteria, and a willingness to adapt.[MG1]
Attain Partners – AI Experts for Mission-Driven Organizations
Attain Partners helps higher education institutions and nonprofits strategically evaluate, adopt, and scale AI in ways that are both mission-aligned and measurable.
Interested in Attain Partners conducting a foundational discovery session to see how your organization can leverage AI? Fill out the form below to request more information.
2025-12 DIG: AI Interest Form
About the Author
Alexander Brown is a Managing Director at Attain Partners. He serves as Enterprise Advisory Capability Lead, overseeing a dynamic business unit that integrates the firm’s tech-agnostic strategy, ERP transformation, and app modernization services. In this role, Alexander leads a multidisciplinary team delivering strategic management consulting, enterprise transformation, application modernization, and custom technology solutions to higher education, nonprofit, state and local government, and commercial clients.
