Developmental evaluation is an approach to program evaluation designed for innovation, complexity, and adaptation. Unlike formative evaluation, which improves a defined program, or summative evaluation, which judges results after implementation, developmental evaluation supports initiatives that are still emerging and changing. I have used it with public health pilots, workforce partnerships, and community-based systems change efforts where goals, activities, and outcomes could not be fully specified at the start. In those settings, conventional evaluation designs were too static. Developmental evaluation provided a disciplined way to learn in real time while a program evolved.
At its core, developmental evaluation embeds evaluative thinking into strategy, decision-making, and implementation. The evaluator works alongside leaders, staff, and partners to collect rapid feedback, test assumptions, interpret patterns, and inform course corrections. Key terms matter here. A complex environment is one where cause and effect are not fully predictable, multiple actors influence results, and solutions must emerge through iteration. A theory of change is still useful, but in developmental evaluation it is treated as provisional and updateable rather than fixed. Learning questions replace narrow performance questions, and data collection is calibrated to usefulness, timing, and context.
This matters because many of today’s hardest social challenges do not behave like stable programs. Climate resilience, behavioral health access, economic mobility, and cross-sector collaboration all involve shifting conditions, political constraints, and diverse stakeholder goals. Organizations need evaluation models and frameworks that do more than prove success after the fact. They need methods that help them navigate uncertainty without sacrificing rigor. As a hub within program evaluation, this article explains where developmental evaluation fits among major evaluation models and frameworks, when to use it, how it works in practice, what methods support it, and what leaders should watch for when building an effective learning system.
Where developmental evaluation fits within evaluation models and frameworks
Developmental evaluation sits within the broader family of evaluation models and frameworks, but it serves a distinct purpose. Traditional experimental and quasi-experimental designs ask whether a defined intervention caused measurable outcomes. Logic models map inputs, activities, outputs, and expected results. Utilization-focused evaluation emphasizes intended use by intended users. Theory-driven evaluation examines how and why change occurs. Developmental evaluation borrows from all of these, yet it is specifically built for adaptive initiatives where the intervention itself is changing as learning unfolds.
In practice, I position developmental evaluation as the right fit when three conditions are present. First, the program is innovative, meaning no validated implementation blueprint exists. Second, the environment is dynamic, so assumptions may shift quickly because of policy, market, community, or operational changes. Third, decision-makers are willing to act on evidence in near real time. If any of those conditions are missing, another evaluation framework may be more suitable. For example, a mature afterschool program with stable operations may benefit more from outcomes evaluation or cost-effectiveness analysis than from developmental evaluation.
One useful way to understand the landscape is to compare common models by purpose, timing, and level of uncertainty.
| Framework or model | Primary purpose | Best use case | Typical timing |
|---|---|---|---|
| Formative evaluation | Improve implementation | Defined program needing refinement | Early to mid implementation |
| Summative evaluation | Judge overall merit or impact | Mature program with stable design | End of cycle |
| Process evaluation | Assess delivery, reach, fidelity | Programs requiring operational insight | During implementation |
| Theory-based evaluation | Test causal assumptions | Programs with explicit logic and mechanisms | Throughout program lifecycle |
| Utilization-focused evaluation | Support intended decision use | Stakeholder-driven learning needs | Designed around decision points |
| Developmental evaluation | Adapt strategy in complex settings | Innovative, evolving initiatives | Continuous, real time |
This comparison shows why developmental evaluation functions as a hub topic under evaluation models and frameworks. It does not replace other approaches; it complements them. Many organizations begin with developmental evaluation during design and early scaling, then transition into process, outcome, or impact evaluation once the model stabilizes. That sequencing is often more credible than forcing experimental rigor too early, when the intervention is still being invented.
When developmental evaluation is the right choice
Developmental evaluation is appropriate when uncertainty is substantive, not incidental. If a team simply lacks a dashboard or baseline data, that is a management problem, not necessarily a reason to choose this approach. The stronger signal is that the initiative has to adapt because relationships, incentives, implementation pathways, or target populations are changing. In one regional workforce project I supported, employers altered skill requirements every quarter, community colleges revised training modules in response, and transportation barriers varied by neighborhood. A fixed evaluation plan would have documented instability; developmental evaluation helped leaders respond to it.
Several scenarios routinely justify this model. Early-stage social innovation is one. Systems change is another, especially when multiple organizations share partial control over outcomes. Policy implementation in contested environments also fits, because frontline interpretation of rules often determines what a program becomes. Collective impact initiatives, equity-centered redesign efforts, and pilots using human-centered design are common examples. In each case, learning has to keep pace with action.
There are also times not to use developmental evaluation. It is a poor fit when funders require definitive attribution on a short timeline, when the intervention must be delivered with high fidelity to meet regulatory standards, or when leadership says it wants learning but actually wants validation. It can also drift if no one owns decisions. Real-time feedback only has value when decision rights are clear and leaders are prepared to change course based on evidence.
A practical screening test is simple: Can the team state what it is trying to influence, what assumptions it is making, what uncertainties matter most, and what decisions will be revisited in the next ninety days? If yes, developmental evaluation can be highly productive. If no, the first step may be strategic clarification before formal evaluation begins.
How developmental evaluation works in practice
The core operating model is embedded learning. The evaluator is not a detached auditor who appears at the end of a grant period. Instead, the evaluator becomes part of the learning infrastructure, while still maintaining analytic independence. In my work, that usually means joining design meetings, reviewing implementation data weekly or monthly, facilitating reflection sessions, and translating mixed-method evidence into decision-ready insights. The evaluator helps the team distinguish weak signals from noise and avoid reacting impulsively to isolated anecdotes.
The process generally starts with framing decisions rather than metrics. Leaders identify critical uncertainties, strategic hypotheses, and adaptive choices. From there, the evaluation team creates a learning agenda: the set of questions that must be answered to improve the initiative. Examples include: Which participant segments benefit most from the current model? What contextual factors are blocking uptake? Which implementation adaptations improve engagement without increasing cost? Those questions guide data collection priorities.
Data collection is intentionally mixed-method and often rapid-cycle. Short pulse surveys, dashboard trends, participant interviews, observation notes, journey maps, implementation logs, network analysis, and document review can all play a role. Developmental evaluation often uses triangulation instead of relying on a single metric. For example, if a community health pilot sees increased attendance, the evaluator may examine referral source data, interview participants about barriers, and review staff notes to determine whether the increase reflects better outreach, easier access, or a temporary external event.
Sensemaking is the distinctive discipline. Teams review evidence at defined intervals and ask what patterns are emerging, what surprised them, what assumptions need revising, and what action should follow. Good developmental evaluation makes those conversations structured rather than informal. It documents decisions, tracks adaptations, and revises the theory of change as the model matures. That audit trail is important because it preserves organizational memory and later supports more formal evaluation designs.
Methods, tools, and indicators that strengthen rigor
Because developmental evaluation is adaptive, critics sometimes assume it is loose or subjective. In strong practice, the opposite is true. Rigor comes from explicit questions, transparent evidence standards, disciplined documentation, and repeated testing of assumptions. The evaluator should define decision criteria upfront where possible. For instance, a pilot might expand to a second site only if participant retention exceeds a threshold, implementation costs remain within range, and qualitative feedback shows the model is culturally acceptable.
Several tools are especially useful. A living theory of change captures how program logic evolves. Outcome harvesting identifies observable changes and works backward to determine how the initiative contributed. Contribution analysis helps assess plausible influence when randomized attribution is unrealistic. Developmental rubrics translate abstract capacities, such as partnership quality or community ownership, into observable levels of performance. PDSA cycles, borrowed from improvement science, support structured testing of small changes. In digital programs, cohort analysis and funnel metrics can reveal where users disengage, while qualitative interviews explain why.
Indicators in developmental evaluation should include both performance and adaptation signals. Performance indicators might track reach, engagement, timeliness, or early outcomes. Adaptation indicators track how the initiative is changing and whether those changes are promising. Examples include the number of tested variations, time from insight to decision, stakeholder alignment, implementation burden, and evidence of fit across settings. In one multisite initiative, we monitored not only youth enrollment and completion but also the frequency of curriculum adjustments, facilitator confidence, and employer feedback on relevance. Those adaptation measures showed which site-level changes improved outcomes before final impacts were visible.
Data governance matters as well. Teams need version control for learning memos, clear protocols for confidential information, and realistic cadences for reporting. Tools such as Airtable, Power BI, Dedoose, NVivo, and Miro can support workflow, but software does not create learning culture. The decisive factor is whether evidence is synthesized into actionable guidance at the right moment.
Leadership, roles, and common implementation mistakes
Developmental evaluation succeeds or fails largely because of leadership behavior. Executives must tolerate ambiguity without becoming vague. Program leaders must distinguish adaptation from drift. Evaluators must build trust while retaining enough distance to challenge assumptions. Funders, when involved, should reward transparent learning rather than punishing every course correction as failure. The strongest arrangements define governance clearly: who reviews evidence, who approves changes, which adaptations require stakeholder consultation, and when the initiative will be considered stable enough for another evaluation design.
Common mistakes are predictable. The first is treating developmental evaluation as constant data collection without prioritization. That creates fatigue and little insight. The second is confusing activity with learning; many teams generate memos and workshops but never make consequential decisions. The third is failing to document adaptations, which makes later interpretation impossible. Another mistake is relying only on internal voices. Complex initiatives need participant feedback, partner perspectives, and contextual intelligence from the field.
I have also seen teams overcorrect toward flexibility and lose comparability across sites. Some variation is necessary, but core components still need definition. A practical solution is to identify fixed elements, adaptable elements, and prohibited deviations. That simple architecture protects program identity while allowing local tailoring. Finally, developmental evaluation should not become permanent by default. Once an intervention demonstrates a stable model, repeatable implementation, and clearer causal pathways, organizations should consider shifting toward process, outcome, or impact evaluation to answer a different class of questions.
Developmental evaluation gives organizations a credible way to learn while acting in uncertain environments. It is most valuable when a program is innovative, context is shifting, and leaders need evidence to shape the next decision, not just assess the last one. As part of the broader field of evaluation models and frameworks, it fills the gap between planning and proof by turning strategy, implementation, and measurement into one disciplined learning process.
The main benefit is practical clarity. Teams stop asking for false certainty and start asking better questions: What is emerging, for whom, under what conditions, and what should we do next? When done well, developmental evaluation builds stronger theories of change, better adaptation records, and more mature programs ready for later outcome or impact testing. It is not a shortcut around rigor. It is the right rigor for complexity.
If you are building a new initiative, leading systems change, or managing a pilot that keeps evolving, use developmental evaluation to create a real learning system. Map your uncertainties, define decision points, choose a small set of meaningful indicators, and review evidence on a fixed cadence. Then connect this hub to your next step: deeper work on logic models, theory-based evaluation, process evaluation, and utilization-focused practice across your broader program evaluation strategy.
Frequently Asked Questions
What is developmental evaluation, and when is it most useful?
Developmental evaluation is an approach to evaluation built for situations where a program, initiative, or strategy is still evolving. Instead of assuming that goals, activities, and outcomes are already stable and clearly defined, it recognizes that many real-world efforts are shaped by uncertainty, experimentation, and changing conditions. In practice, this makes developmental evaluation especially useful for innovation-focused work, complex partnerships, early-stage pilots, systems change efforts, and community-based initiatives where people are learning what works as they go.
It is most valuable when leaders and stakeholders are not just asking, “Did it work?” but also, “What are we learning?”, “How is the context changing?”, and “What should we adapt next?” In public health pilots, workforce collaborations, and cross-sector partnerships, for example, the intervention itself may shift over time in response to participant feedback, policy changes, local conditions, or emerging opportunities. Developmental evaluation helps teams make sense of that movement. Rather than waiting until the end to judge outcomes, it provides ongoing evidence and structured reflection that support better decisions in the moment.
At its core, developmental evaluation is not about enforcing a fixed plan. It is about helping an initiative develop intelligently. That means tracking what is emerging, identifying patterns, testing assumptions, clarifying strategy, and feeding useful insights back into decision-making. When the path forward cannot be fully specified from the start, developmental evaluation offers a practical and disciplined way to learn while building.
How is developmental evaluation different from formative and summative evaluation?
The main difference is the level of stability in the program being evaluated. Formative evaluation is typically used to improve a program that already has a defined model, clear activities, and relatively established objectives. It asks how the program can be strengthened before or during implementation. Summative evaluation usually comes later and focuses on results, effectiveness, and impact after the program has been delivered. It is often used to determine whether an initiative met its goals and whether it should be continued, expanded, or replicated.
Developmental evaluation serves a different purpose. It is designed for initiatives that are still taking shape. In these settings, the evaluator is not simply measuring whether a predefined model was implemented correctly or whether expected outcomes were achieved. Instead, the evaluator helps the team examine what is emerging, what assumptions are being challenged, how the context is shifting, and what adaptations are needed. The emphasis is on strategic learning in real time.
Another important distinction is the role of the evaluator. In formative and summative work, evaluators are often more external to day-to-day decision-making. In developmental evaluation, the evaluator is frequently more embedded in the initiative, working closely with program leaders, staff, and partners to gather evidence and inform ongoing adjustments. That does not mean abandoning rigor. It means applying rigor in a way that fits complexity and supports adaptation. When a program is stable, formative or summative approaches may be the better fit. When the program itself is still emerging, developmental evaluation is often the more appropriate choice.
What does a developmental evaluator actually do during a project?
A developmental evaluator helps an initiative learn systematically while it is being designed, tested, and refined. This usually involves a mix of activities rather than a single evaluation report at the end. The evaluator may work with the team to clarify the current theory of change, identify key assumptions, map stakeholders, define learning questions, and decide what signals or indicators will show whether the effort is moving in the right direction. Because the work is evolving, these elements are often revisited and revised over time.
On an ongoing basis, the evaluator collects and interprets information that can guide adaptation. That may include interviews, observation, document review, rapid feedback loops, meeting analysis, participant input, implementation tracking, outcome monitoring, or sense-making sessions with staff and partners. In a public health pilot, for example, the evaluator might surface early evidence about who is being reached, where delivery barriers are emerging, and which design changes appear to improve engagement. In a workforce partnership, the evaluator might help the group understand how employer needs, partner roles, or participant pathways are changing and what those shifts mean for strategy.
Just as important, the evaluator helps translate information into action. Developmental evaluation is not only about collecting data; it is about supporting reflection, interpretation, and decision-making. A good developmental evaluator helps teams distinguish between noise and meaningful patterns, recognize when an adaptation is needed, and document what is being learned along the way. This role requires strong evaluation skills, but also facilitation, systems thinking, communication, and the judgment to work effectively in environments where there may not be simple answers.
What kinds of programs or initiatives are a good fit for developmental evaluation?
Developmental evaluation is a strong fit for initiatives operating in complex, uncertain, or rapidly changing environments. These are often efforts where cause and effect are not straightforward, multiple partners influence outcomes, and local context matters a great deal. Common examples include public health innovations, cross-sector workforce partnerships, community-based prevention strategies, equity-focused initiatives, policy experiments, and large-scale systems change efforts. In these contexts, it is often unrealistic to fully specify the program model in advance because the work must adapt as conditions change and learning accumulates.
It is especially useful when an organization is testing a new idea, entering unfamiliar territory, or trying to solve a problem that does not respond well to standard program designs. If stakeholders know the destination in broad terms but are still discovering the best path to get there, developmental evaluation can provide the structure needed to learn without forcing false certainty. It can also be valuable when a program is intentionally tailored across sites or communities, making it more important to understand how adaptation is happening than to enforce uniform implementation too early.
That said, developmental evaluation is not the best fit for every situation. If a program is already well defined, operating consistently, and mainly needs quality improvement or outcome measurement, formative or summative approaches may be more efficient. The key question is whether the initiative is stable enough to evaluate against fixed expectations, or whether it is still evolving and needs an evaluation approach that can move with it. When innovation and adaptation are central to success, developmental evaluation is often the right tool.
What are the main benefits and challenges of using developmental evaluation?
The biggest benefit of developmental evaluation is that it makes evaluation immediately useful in complex, changing environments. Instead of producing findings only after major decisions have already been made, it gives teams timely evidence they can use to refine strategy, adjust implementation, and respond to shifting conditions. This can reduce wasted effort, improve alignment among partners, and strengthen an initiative’s ability to learn and evolve. It also helps organizations document how ideas change over time, which is especially important in innovation work where the final model may look quite different from the original concept.
Another major advantage is that developmental evaluation supports more realistic learning. In many community and systems change efforts, progress is not linear. Stakeholders may have different goals, external events may disrupt plans, and outcomes may emerge gradually. Developmental evaluation is well suited to that reality because it does not force a rigid framework onto dynamic work. Instead, it helps teams understand relationships, feedback loops, unintended consequences, and emerging opportunities. Over time, this can lead to better strategy and stronger readiness for future formative or summative evaluation once the initiative becomes more defined.
The challenges are equally important to understand. Developmental evaluation requires comfort with ambiguity, strong collaboration, and a willingness to adapt based on evidence. Some organizations find this difficult because they are used to fixed plans, predefined metrics, or traditional reporting structures. It can also be challenging to explain to funders or stakeholders who expect clear outcomes from the start. In addition, because the evaluator is often closely involved with the initiative, the work demands careful attention to role clarity, credibility, and methodological discipline. When done well, developmental evaluation is highly valuable, but it works best when leaders genuinely want to learn, are open to change, and understand that in complex settings, the most important question is often not just whether something worked, but how it is developing and what should happen next.
