The challenge
Scattered pilots without alignment, no consistent way to prioritize, success left undefined or unmeasured, and ethical blind spots that go unchecked.
The AI Strategic Compass is a shared method for campus teams weighing an AI proposal. It scores the proposal on five dimensions, sets a review process that matches the stakes, and defines what success looks like before money, data, or trust is committed.
AI enters campus from every direction: faculty pilots, vendor features, departmental tools, enterprise platforms. Without a shared way to evaluate them, institutions end up with duplication, unchecked risk, and lost trust.
Scattered pilots without alignment, no consistent way to prioritize, success left undefined or unmeasured, and ethical blind spots that go unchecked.
Nearly every institution now has an AI strategy, but the guidance often doesn't reach the people doing the work. In EDUCAUSE's 2026 study, 46% of respondents weren't aware of policies or guidelines for using AI in their work, including 38% of executive leaders. Few campuses have a shared way to decide which AI initiatives to pursue, how to review them, and how to tell whether they worked.
Source: EDUCAUSE, The Impact of AI on Work in Higher Education (2026)
A clear rubric to evaluate fit, risk, and impact; a way to define success through KPIs, OKRs, and efficacy measures; and a shared language for collaborative decisions.
| Without a shared process… | The Compass asks |
|---|---|
| Which initiatives move forward depends on who asks loudest or who has budget. | Is it worth doing? Strategic alignment and value creation are scored first, against defined institutional goals. |
| Decisions are made by one office, and the people affected are absent. | Who gets to decide? A cross-functional team scores together: proposers, governance and review groups, specialist reviewers, and leaders, with stakeholder voices in the room. See who decides → |
| Tools are judged on features or cost alone; risks surface late. | Are all aspects considered? Five dimensions are weighed: strategy, ethics and compliance, finances, operations, and stakeholder impact. They sit side by side, with GRC maturity and an explicit equity & accessibility check alongside. |
| Every idea gets the same review, heavy or light, whatever the stakes. | Is the process proportionate? The review band sets how formal review is: a light check for low-risk pilots, full scrutiny for high-stakes systems. See proportionality → |
| No one can say whether an initiative worked, or why. | How will we know? KPIs, OKRs, and efficacy measures are drafted at intake, tracked through the pilot, and reviewed before scaling, with exit criteria and re-review triggers set up front. |
Proposers draft intended success metrics at intake. A rough draft is fine. Rubric scoring then informs the decision to proceed, pause, or revise, and formal success planning follows selection.
The five dimensions with level descriptors and examples, the four review bands, and how governance, risk, and compliance mature alongside.
Read the framework 02An interactive worksheet: describe, score, plan, and reflect. Then compare initiatives side by side in a portfolio and export the results.
Open the scorer 03How KPIs track what a project delivers, OKRs what the institution gains, and efficacy whether the AI itself works. A worked early-warning-system example.
Define successAnd what would be missing from that process? Score your candidates and find out.