Why adaptive capacity should be treated as a measure of humanitarian performance

November 18, 2026

Rama Nassif

A flooded road with orange traffic barrels and a warning sign reading “Road Closed Ahead” due to water overflow, with a bridge and trees in the background.

Recovery programmes are designed amid uncertainty, yet many become harder to change precisely when new information arises. The problem is not always a lack of feedback. It is often an institutional architecture that rewards delivery against the original plan more reliably than learning from what implementation reveals.

The plan is often most flexible before it meets reality

Humanitarian and recovery programmes need plans. Budgets must be approved, procurement initiated, responsibilities assigned and results made measurable. The difficulty begins when these necessary instruments become commitments to assumptions that were provisional at the time they were made.

Conditions change after implementation starts. Prices move. Population patterns shift. Local organisations identify barriers that were invisible during assessment. A service that looked accessible on paper proves difficult to use. A technically sound intervention interacts with institutions, informal systems or social norms in ways that were not anticipated. At that point, a programme has learned something. But learning only matters if the programme can act on it.

Four ways learning becomes institutionally expensive

The first constraint is procurement rigidity. Once equipment, services or construction packages have been specified and contracted, changing course may require amendments, new approvals or delays. Even when teams recognise that a different intervention would work better, the transaction costs of adaptation can make continuation easier than correction.

The second is fixed indicators. Indicators are essential for accountability, but they can become substitutes for the objective they were meant to represent. A programme may continue reporting outputs successfully while the relationship between those outputs and the intended outcome weakens. If success is defined as delivering what was promised rather than testing whether the promise still makes sense, evidence that challenges the intervention can become administratively inconvenient.

The third is reporting architecture. Donors and organisations understandably need predictable reporting. Yet reporting systems often ask whether activities were completed, funds spent and targets reached more consistently than they ask what assumptions proved wrong and what changed because of that learning. This can create an incentive to narrate implementation as progress rather than treat implementation as a source of new evidence.

The fourth is contractual lock-in. Partnerships, timelines, deliverables and responsibilities are frequently formalised before local conditions are fully understood. The longer implementation continues, the more organisational commitments accumulate around the original design. Changing a programme then becomes a potentially long-winded renegotiation of budgets, relationships and institutional accountability.

Feedback is not the same as adaptive capacity

Humanitarian organisations have invested substantially in community engagement, monitoring, evaluations and feedback mechanisms. These are important. But collecting feedback does not establish that a programme can change in response to it.

A project can have consultation meetings, complaint channels, monitoring dashboards and mid-term reviews while remaining structurally unable to alter its core decisions. In that situation, information travels upward but authority does not travel back down. The feedback loop is technically present and operationally incomplete.

This is partly a question of decision rights. Teams closest to delivery should be able to make limited operational adjustments in response to feedback. Changes to eligibility, budgets or procurement commitments need a clear route to authorised programme managers or donors, with decisions made promptly and recorded. Otherwise, feedback may be heard without influencing delivery.

This distinction matters because participation is often measured at the point where information enters the system: who was consulted, how many people responded, whether feedback was recorded. Adaptive capacity should instead ask what happened after the information arrived. Did it change a budget? A procurement specification? An eligibility rule? A delivery mechanism? A timeline? If not, why not?

Recent humanitarian practice points in the same direction

A recent Humanitarian Practice Network reflection on learning and adaptation in humanitarian research makes the practical case clearly: realities rarely align neatly with initial plans, and structured reflection is valuable because it can surface the adaptations required for programme relevance and meaningful participation. The broader operational lesson is that adaptation should not be treated as a departure from good design. In uncertain environments, it is part of good design.

This is particularly important in recovery, where interventions extend beyond immediate relief and begin interacting with local institutions, markets, infrastructure systems and community practices over time. The longer the programme horizon, the less plausible it becomes that every important assumption made at the beginning will remain valid.

A different test of performance

This suggests a simple addition to how recovery performance is judged. Alongside expenditure, outputs, timeliness and coverage, programmes should be able to demonstrate adaptive capacity.

That does not mean constant redesign. Predictability matters, and changing direction carries costs. It means distinguishing disciplined adaptation from implementation drift. A programme should be able to identify which assumptions are critical, specify what evidence would justify revisiting them, and specify who can approve changes and what budget or contractual flexibility is available to respond when that evidence appears. For example, a field team might adjust delivery times within agreed limits, while changes to eligibility or procurement terms would require documented approval from programme managers or donors.

Three practical questions can help:

  • What did the programme learn during implementation that was not known at design stage?
  • Which decisions changed because of that learning?
  • Which decisions could not change, and what institutional constraint prevented adaptation?

These questions would make visible a form of failure that conventional performance reporting can miss: a project can deliver its planned activities while becoming progressively less appropriate to the conditions it is supposed to address.

Recovery should include a commitment to learning and adaptation

Accountability is usually associated with keeping commitments. In uncertain environments, however, there is another responsibility: recognising when keeping the original commitment literally would undermine the purpose for which it was made.

This is why adaptive capacity should not be treated as an optional feature added by unusually flexible teams. It should be designed into budgets, contracts, monitoring systems and decision rights from the beginning. Contingency is not evidence of weak planning when the environment itself is unstable. The inability to respond to evidence may be the greater planning failure.

Recovery is often tested against a straightforward question: did the programme deliver what it said it would deliver? But the more revealing question is: when reality changed, was the project able to change?


Rama Nassif is an environmental engineer and independent researcher based in Aleppo, Syria. Her work examines governance, institutional capacity, sustainability and recovery systems, with particular attention to the gap between technical interventions and the institutional conditions that determine whether they can adapt and endure.

Comments

Thanks for choosing to leave a comment. Please keep in mind that all comments are moderated according to our comment policy.

Let’s have a personal and meaningful conversation.

Can you help translate this article?

We want to reach as many people as possible. If you can help translate this article, get in touch.
Contact us

Did you find everything you were looking for?

Your valuable input helps us shape the future of HPN.

Would you like to write for us?

We welcome submissions from our readers on relevant topics. If you would like to have your work published on HPN, we encourage you to sign up as an HPN member where you will find further instructions on how to submit content to our editorial team.
Our Guidance