This post first appeared on BetterEvaluation on 28 October 2026.
By Arnaldo Pellini, Men Yon, Juairia Sidabutar, Rizky Firtiani, Mark Fiorello, Paul Wright
In this blog, we want to share some of our reflections on the Outcome Harvesting journey with the Kemitraan Indonesia Australia untuk Infrastruktur (KIAT, ‘Indonesia Australia Partnership for Infrastructure programme’) between 2022 and 2025. Unlike traditional evaluation approaches that measure progress against predetermined targets, Outcome Harvesting works backward to identify both expected and unexpected changes in the behaviours, policies, or practices of social actors, or, in KIAT’s case, government partners.
The Outcome Harvesting (OH) approach explicitly helps monitoring teams describe the specific change that occurred, analyse its significance, and verify how a specific programme contributed to it. In a government partnership such as KIAT, the goal is to contribute to policy decisions (in a broad sense). Because many factors can contribute to these decisions, no single actor can, or should, claim 100% credit for a new regulation or institutional shift. By using OH as part of the MEL toolbox for KIAT, we were able to focus on KIAT’s plausible contributions to policy and practice changes. This provided KIAT’s Monitoring, Evaluation, Learning, and Performance (MEL-P) team with a more nuanced account of how institutional change actually happens within a complex sector such as infrastructure. This was then applied in KIAT’s reporting to DFAT and GoI.
The main takeaway from this experience is that OH doesn’t have to be a retrospective post-mortem reserved for a programme’s end. It can be adapted and serve as an insightful compass for real-time adaptive management. OH can be a useful approach for monitoring and learning during a programme implementation. Below, we dive straight into our main insights—including the benefit of a combined internal-external lens, the necessity of tracking “emergent” outcomes, and why monitoring teams may look beyond signed decrees to capture actual changes in implementation and practice. We then describe how we adapted the method and close with practical considerations for monitoring teams.
Our main takeaways and insights
Tracking emergent outcomes
A facility such as KIAT is fundamentally different from a traditional development programme. Facilities are defined as “a delivery mechanism that provides flexible (adaptive and responsive) services managed in an integrated way. Objectives (or end-of-facility outcomes) are specified, but the pathways to deliver them are left unspecified” (Pieper, L., 2018).
Because KIAT’s portfolio regularly shifted to match the Government of Indonesia’s evolving infrastructure agenda, a single OH evaluation at the end would have missed the story of how change actually unfolded across the various policy issues where KIAT was involved. To capture this reality, we integrated the OH exercises into the ongoing monitoring and reflection work of the MEL-P team around the mid-point of KIAT’s implementation. We tagged the outcomes we were harvesting as Established (verified actions or decisions), Emergent (promising but pending progress), or Potential (early signs of change).
By running three systematic rounds (a pilot in 2022, mid-2024, and late 2025), we moved away from a binary success-or-failure perspective. Instead, we were able to actively map how early Potential outcomes matured into Emergent progress and finally became fully Established policy shifts or decisions, allowing us to track change over time and recognise early signs of progress before they met the formal, traditional definition of an ‘outcome’ in OH.
By the final round, 13 outcomes first logged as Potential or Emergent had become Established. Wastewater management in Palembang, a city of roughly 1.7 million people in Indonesia’s South Sumatra province, shows what this looks like. The 2024 harvest captured a cluster of linked changes: city regulations establishing Perumda Tirta Musi as a combined municipal water and wastewater utility, with tariffs set the following year; a utility that had only ever run water supply taking over the new treatment plant, creating a wastewater division, hiring staff and writing operating procedures; and the province and city jointly funding household connections, which we had first logged as emergent. The 2025 harvest then picked up something we had not anticipated: a scheduled, tariff-funded desludging service, launched in September 2024 and expanded by a staff-owned cooperative buying trucks.
Combining internal and external lenses
One of the strengths of this exercise, we believe, was the synergy between the insider knowledge of the KIAT team and the outsider perspective of the external consultants. We found that the KIAT MEL-P team and technical teams brought a depth of knowledge which comprises the tacit knowledge held by the teams, and the hard evidence collected through KIAT’s internal monitoring systems. We paired this deep internal knowledge with the critical perspective of external consultants through a clear division of responsibilities: the internal team took the lead on unearthing data and drafting initial outcome statements, while the external consultants provided the objective challenge, verification, and methodological rigour. This experience taught us that such a partnership allows internal data to meet high harvesting standards, ensuring the final product is both locally grounded and externally credible without the need for external validation.
Focusing on both policy and behaviour change
In public sector governance, it is tempting to view a signed decree or an enacted regulation as the ultimate victory. However, our OH journey with KIAT taught us that while change in policy content is necessary, what really matters is policy coherence and implementation. This insight showed us a possible limitation of OH as a monitoring tool: because the methodology requires outcomes to be discrete and verifiable, it can easily bias a harvest toward tangible policy documents rather than the sometimes-messy shifts in institutional mindsets that bring those policy documents to life.
To overcome this, we used the harvests to capture the dynamic, two-way relationship between formal policy and institutional practice. We learned that bottom-up actions and interactions between everyday departments create the transformative mechanisms that build institutional readiness and shape policy content change in the first place. Once enacted, that top-down policy becomes a structural mechanism that forms the new rules, constraints, and incentives governing how those actors behave moving forward.
The Integrated Road Management System version 3 (IRMS-V3) road-planning tool illustrates this. A 2021 circular letter making it the official modelling system is the kind of outcome any results framework would pick up. The harvest also captured what followed: all 34 regional road offices using it to set annual work plans, cutting scenario analysis from up to a month to a few days, and the Directorate General of Highways then using it to set its first bridge-condition targets in the 2025-2029 Strategic Plan. By documenting these behavioural shifts alongside formal policy milestones, the harvest showed that policy and practice are not separate endpoints, but mutually reinforcing steps toward sustained impact.
Outcome Harvesting in KIAT
Outcome Harvesting has been a core part of the portfolio of tools managed by KIAT’s MEL-P team. As noted earlier, KIAT engaged a small team of external consultants to design and adapt the OH process to the needs of the KIAT facility.
Our collaboration began in January 2022 with an underlying question: Can Outcome Harvesting work for a facility as complex as KIAT? We started with a pilot focused on just two workstreams (out of nine total). The results were clear: the methodology not only worked but provided useful insights and validation for some of the data and information being captured by KIAT’s internal MEL systems.
Based on the experience and learning from that pilot, we expanded to two full-portfolio harvests, conducted in mid-2024 and late 2025. By the final round, we were tracking progress across all five of KIAT’s End-of-Facility Outcomes.
This reflection comes at a pivotal moment: KIAT concluded in June 2026, and its successor KINETIK is now under way. The 63 established outcomes we have documented serve as an evidence base for what worked, and why, in the world of infrastructure policy and governance.
We decided to use OH as a continuous MEL tool to record progress about changes the programme was contributing to. To track the real-world influence of the programme, we applied the following definition of an outcome:
“A significant, observable, and verifiable change in the behaviour or actions of a social actor—such as a government agency—influenced by KIAT’s work.” (See the Outcome Harvesting page on BetterEvaluation)
Developing this explicit definition from the outset allowed the MEL-P team to establish a common interpretation of outcomes with the KIAT technical teams, ensuring everyone had a unified understanding of exactly what types of results the OH exercise was focused on capturing.
To ensure that OH methodology would accompany programme implementation, we designed and populated an OH Log as a spreadsheet featuring the three critical OH elements: a brief description of the change, its significance, and KIAT’s contribution.
We then added a number of tags to provide deeper context for every outcome, such as links to the End-of-Facility Outcomes, specific workstreams, national vs. local level changes, and alignment with the G20 Principles for Quality Infrastructure Investment (here). The idea was that these tags could reveal patterns in how the outcomes contribute to different aspects of the programme’s policy-influencing process.
Institutional and policy change tends to emerge over time through a series of incremental changes and contributing factors. By tracking our progress through these distinct stages in our OH Log, we avoided the trap of only counting “final” results. This allowed us to acknowledge the real time required for institutional change and explicitly map where future tracking would be most useful.
The “So What” for others thinking about using OH for monitoring
Based on our experience with OH in KIAT we want to offer four quick, practical takeaways for a programme that is thinking about adopting and adapting OH for monitoring purposes:
Integrate, don’t duplicate: While OH is traditionally used in data-poor settings, it can also work well in data-rich environments to sift, synthesise, and validate existing data. The key is clearly defining how it sits within the MEL systems.
Balance internal and external roles: Establish a clear division of labour. Let your internal team leverage their deep context to draft initial outcomes and use external consultants as an objective “quality-control loop” to maintain methodological rigour.
Engage technical teams as co-owners: OH cannot be a siloed MEL exercise. Technical leads hold both the tacit knowledge and the relationships with government partners that are vital to accurately explaining contribution and significance.
Match verification to the outcome type: Tailor your verification strategy. Tangible policy content (like a signed decree) can be verified by public documents, while behavioural shifts (like improved department collaboration) should lean on internal meeting logs and interviews.
Acknowledgments
The Kemitraan Indonesia Australia untuk Infrastruktur (KIAT) [Translated: Indonesia Australia Partnership for Infrastructure] was a facility implemented between June 2017 and June 2026. A partnership between the Government of Australia and the Government of Indonesia, KIAT provided technical assistance to improve infrastructure policy, planning, and delivery across: Water and Sanitation; Transport; Infrastructure Funding and Financing; Gender Equality, Disability, and Social Inclusion (GEDSI); and climate change. Working with central and sub-national governments, multilateral development banks, and civil society, the facility focused on driving sustainable economic growth, preparing high-quality projects for financing, and fostering climate-resilient, inclusive infrastructure systems. KIAT was funded by Australia’s Department of Foreign Affairs and Trade (DFAT).
Sources: Pieper, L. (2018). Review of Selected DFAT Facilities. Independent Consultant Report. Canberra: DFAT. Available at: https://www.dfat.gov.au/sites/default/files/independent-facilities-review.pdf