Utilization-Focused Evaluation: designing for use from day one

Primary intended users, by name. Use happens through people, not organisations. UFE requires identifying the actual individuals with the authority, interest, and tenure to act on findings, and engaging them throughout — because involvement builds the understanding and ownership that use runs on. Patton calls this the personal factor: the presence of an identifiable person who cares about the evaluation is the best predictor of use.

Intended uses, negotiated up front. Instrumental decisions (continue, revise, scale, kill), conceptual learning, accountability, programme improvement, process uses (what participating in the evaluation itself changes) — the users choose, explicitly, and the choice governs everything downstream. An evaluation cannot serve all uses equally, and pretending otherwise is how reports end up serving none.

The framework’s standard reference is Patton’s Utilization-Focused Evaluation (SAGE, 5th edition); the empirical base on what drives use — the literature the personal factor emerged from — is synthesised in Johnson and colleagues’ Research on evaluation use, a review of 25 years of evaluation-use studies.

• Questions: derived from decisions users actually face, not from the logic model’s completeness. The test: “what would you do differently depending on the answer?”
• Methods: chosen for credibility to these users at the precision these decisions need — methodological gold-plating that arrives late is a UFE failure, not a virtue.
• Timing: findings sequenced against the users’ decision calendar; an interim memo before the budget meeting can outrank the final report.
• Reporting: formats users will absorb — working sessions, decision memos, dashboards — with the archival report as record rather than vehicle.
• Facilitation as core skill: the UFE evaluator negotiates, teaches, and keeps users engaged across the study — active-reactive-adaptive, in Patton’s phrase.

A foundation commissioned an evaluation of its flagship youth-mentoring portfolio. A conventional scope — outcomes across 40 grantees — was reframed at the readiness step: the real decision, eighteen months out, was whether to concentrate funding on intensive models or breadth of coverage, and the primary intended users were three named people — the programme director, the board’s education chair, and the finance director.

Negotiated intended use (one instrumental decision) collapsed the question set from nineteen to four, all comparative: relative outcomes, relative cost per sustained match, implementation preconditions of the intensive model, and grantee capacity to convert. Simulated-findings sessions killed one planned survey (the education chair, shown mock results, admitted no version would move her) and added cost-modelling nobody had scoped. Interim findings landed as working sessions timed to board cycles; the final session was a facilitated decision meeting, not a presentation.

The board chose concentration, citing the comparative cost-outcome evidence; the finance director’s early involvement meant the modelling was already trusted; the programme director used the preconditions findings to design the transition. A use review six months later — part of the evaluation contract — documented the decisions taken and the two findings that had not been used, with reasons. The report was forty pages; the use was the point.

• “Stakeholders” instead of users. Broad consultation without named decision-owners reproduces the diffuse-audience problem UFE exists to solve.
• Uses assumed, not negotiated. The evaluator infers what would be useful; the users disagree at delivery, politely.
• Engagement as kickoff meeting. Users consulted once, then re-encountered at the final presentation — ownership never forms.
• Question sprawl. Every interested party’s question admitted; prioritisation is the framework’s teeth, and pulling them is the commonest failure.
• Ignoring turnover risk. The champion leaves, the evaluation orphans; UFE practice hedges with user groups and documented agreements.
• Use unassessed. Ending at report delivery without the follow-through that checks whether intended uses happened — the framework judged by its own standard.

UFE’s user-centring is a values choice with costs: questions that matter to non-users — participants, the public, the field — can be deprioritised, and accountability audiences may distrust an evaluation so visibly shaped by its subjects’ decision-makers. Transparency about the negotiation, and honest boundaries on what the evaluation can claim, are the mitigations.

The personal factor cuts both ways: the framework’s dependence on engaged individuals makes it fragile to turnover and politics, and demands facilitation skills not every evaluator has.

And situational responsiveness resists standardisation — two UFE evaluations of similar programmes may legitimately look nothing alike, which complicates synthesis and replication. UFE optimises the single evaluation’s usefulness; field-level learning needs additional scaffolding.

Formative evaluation is a natural intended use inside the framework when improvement is the decision.

Developmental evaluation is Patton’s sibling framework for innovation contexts without a stable model to evaluate.

Participatory evaluation shares the engagement machinery with a different rationale — rights and empowerment rather than use.

A theory of change often anchors the negotiated questions when users’ decisions concern the programme’s causal claims.

UFE’s pace is set by decision calendars, so analysis speed is strategic: user working sessions need interview and open-text evidence coded, quantified, and quotable by the meeting date. Evidano turns weeks of interview analysis into days with quote-linked themes users can interrogate live — “show me who said that” is exactly the credibility interaction UFE sessions run on — and cross-segment views (by site, grantee, model) match how decision-makers actually compare options.

The framework’s core work — naming users, negotiating uses, keeping engagement alive — is facilitation, and no tool does it. What the tool removes is the classic excuse for unused evaluations: findings that arrived after the decision.