What to Fix First in a Humanitarian Emergency: The 3 Most Common Mistakes

In the first 72 hours of a humanitarian emergency, the noise is deafening. Aftershocks. Screaming. Satellite phones ringing. Every agency on the ground wants to do something — anything — fast. But here is the thing: speed without direction kills. I have seen teams drop pallets of bottled water into a flood zone where the only thing people needed was oral rehydration salts and a latrine. The water sat. People died of diarrhea anyway.

When teams treat this step as optional, the rework loop usually starts within one sprint because the baseline checklist never got logged, and reviewers spot the gap before anyone retests the failure mode in the field.

The mistake is not a lack of heart. It is a lack of a fix-first mentality. You cannot fix everything. You have to pick the one leverage point that stops the worst from getting worse. And most responders pick wrong. This article walks through the three most common mistakes — and what to fix first instead.

Most readers skip this line — then wonder why the fix failed.

Why the First 72 Hours Create or Break a Response

The golden window — and why most teams miss it

The first seventy-two hours are not a suggestion. They are the only real chance you have to set the trajectory of an entire humanitarian response. I have watched teams arrive with world-class logistics plans, only to spend two of those three days arguing over who controls the water point. That is not a failure of will. It is a failure of priority. The window itself is brutally short: by hour forty-eight, malnourished children start slipping past the point of easy recovery. By hour sixty, the rumor mill has already decided which aid group is corrupt. You do not get a do-over on first impressions. What breaks first is trust — and trust, once cracked, takes weeks to repair.

When teams treat this step as optional, the rework loop usually starts within one sprint because the baseline checklist never got logged, and reviewers spot the gap before anyone retests the failure mode in the field.

Cost of a wrong first move

Wrong order. That is the hidden killer. Teams rush to distribute high-protein biscuits before they have secured a latrine site — and two days later, cholera blooms in the shelter area. I have seen this exact sequence play out. The biscuits sat in a warehouse while the medical team scrambled to treat the first diarrhea cases. The catch is that the biscuit distribution felt productive. It generated photos. It gave donors a warm moment. But it stole time from the one task that would have prevented the outbreak: separating human waste from drinking water sources. The trade-off is brutal: you spend political capital on a visible action, and the invisible infrastructure collapses. Most teams never recover the lost seventy-two hours because they refuse to admit the first move was cosmetic.

'We unloaded tents for three straight days. On day four, we discovered everyone was drinking from the same pond the latrines drained into.'

— field coordinator, anonymous debrief, 2019

Why experience is no shield

Veteran responders make this mistake more often than rookies. That sounds backwards. It is not. Experience creates scripts: 'In the last earthquake, we did X, so we do X here.' But the first seventy-two hours of a flood response differ fundamentally from a conflict displacement. Experienced teams overcommit to familiar solutions — they truck in bottled water when the local aquifer is intact, or they set up a field hospital before they have mapped the nearest functioning clinic. What usually breaks first is the assumption that context is a detail rather than the whole equation. A seasoned logistician I worked with once spent eighteen hours sourcing a generator for a cold chain that did not yet exist. Meanwhile, the actual cold chain — a set of solar fridges already on site — sat unplugged because nobody had checked the local voltage. That was day one. Day two was spent apologizing.

The Core Idea: One Leverage Point Changes Everything

Leverage vs. Volume

Most teams arrive with a truck full of stuff. Blankets, rice, water sachets. They measure success by tonnage discharged. I have watched a warehouse fill to the rafters while three blocks away children drank from a sewage ditch. Volume feels productive. It is often just expensive noise.

The trick is finding the single constraint that, when released, pulls ten other problems loose. One broken water pump in a camp of eight thousand people — fix that, and you slash diarrhoea rates, free up women who spent four hours queuing at the distant tap, reduce the fire risk from families boiling river water indoors. That is leverage. One repair, four cascading benefits. The alternative is throwing more jerrycans at the queue, which treats the symptom and buries the real fault.

Wrong order. First find the pump.

The water-sanitation-food hierarchy

It sounds obvious until the donor calls and says 'we need nutrition numbers by Friday'. Nutrition matters. But you cannot treat moderate acute malnutrition with plumpy'nut if every sip a child takes re-infects her gut. The hierarchy holds: safe water first, then toilets that people actually use — not the blue plastic latrines that fill in a week and nobody cleans — then hygiene behaviour, then food. That sequence is not a suggestion. It is a physics problem.

Evidence from recent crises keeps confirming the pattern. In one camp I visited, the medical team had a pristine cholera treatment unit and zero confidence because the water point sat downstream of the latrine field. They spent three weeks negotiating a new borehole location. That borehole, once drilled, cut acute watery diarrhoea by sixty percent within ten days. The treatment unit sat quiet. The leverage point was not more medicine. It was a hole in the ground, two hundred feet deep.

Most teams skip this: they jump straight to the visible emergency — the sick child, the collapsed shelter — and miss the upstream cause that will produce another sick child tomorrow.

'We kept asking for more medical staff. What we needed was one plumber and a truck of PVC pipe.'

— Field coordinator, South Sudan, 2019

Why the hierarchy breaks

The catch is that water-first logic fails when the aquifer is saline or the government blocks drilling permits. Then volume becomes the only lever left. You truck water. You truck it from sixty kilometres away. Diesel costs eat the budget. The hierarchy still holds as a diagnostic tool — it tells you exactly where the system is broken — but it cannot tell you how to fix a broken system with no water source. That is a different problem, and it demands a different leverage point: negotiate a temporary pipeline agreement, or shift to rainwater catchment before the dry season locks in.

What usually breaks first is not the technical solution. It is the assumption that the obvious fix will work. The hierarchy gives you the right question, not the easy answer.

Under the Hood: How a Needs Assessment Really Works

Rapid assessment tools — faster than perfect

The window slams shut fast. You do not have two weeks for a census. What you need is a rapid needs assessment — a stripped-down snapshot that tells you who is dying and why. Most teams default to a long survey form they used last year. Wrong move. In the first 72 hours, paper forms waste time you do not have. I have watched teams spend six hours translating a 40-question tool while a cholera case walks past the tent. A better bet: a single-sided, 10-question form with pictures. Use icons for water source, shelter type, and malnutrition signs. Pair it with a mobile-phone-based collector like KoboCollect or ODK. That data uploads in real time. You spot the trend before lunch.

The trap here is tool fetish — believing the app itself fixes bias. It doesn't. An app just digitises garbage faster. What matters is the question order. Ask about what people need before you ask who they are. If you lead with 'Name and ID number,' survivors freeze. They fear registration equals deportation. Start with 'What is the one thing your child ate today?' That builds trust. That gets real data.

Data triage — throw away most of what you collect

Most field teams gather too much. They measure everything. Roof type, cooking fuel, education level, phone ownership. Stop. You have hours. Triage the data into three buckets: life-safety, critical gaps, nice-to-know. Life-safety means: are people drinking safe water right now? Are pregnant women bleeding out with no transport? Critical gaps: where are the latrines relative to where women sleep? Nice-to-know: how many children were in school before the crisis? That last bucket gets dropped. The catch is painful — you will miss context. You trade depth for speed. That is the deal.

One concrete example: during a rapid assessment I joined in the eastern Sahel, the team recorded 23 indicators per household. Two days later we had 1,100 data points and zero actionable insight. We fixed it by cutting to five indicators: water source distance, child diarrhoea in 24 hours, open defecation visible, unaccompanied minors, and cooking fuel availability. That one sheet told us where to dig boreholes and where to distribute soap. The other 18 indicators stayed in the backpack. They still sit there.

Common data traps — and how to dodge them

The biggest trap is convenience sampling. You park your assessment team near the main road. You interview the people who walk past. Those are the able-bodied men. Women, elderly, and disabled people stay behind. Your data says 'no one needs food.' But you are measuring the wrong crowd. Fix this: use a sector-sampling method. Divide the camp or village into four quadrants. Send one team to the far corner, not just the entrance. If you cannot walk that far, you are already misleading yourself.

‘We interviewed 200 families. Every single one said they had enough blankets. Then we walked 300 meters into the flood zone and found 80 families sleeping on plastic sheets.’

— logistics officer, South Sudan flood response, 2022

The second trap is translation bias. Your interpreter is a local student who speaks English. He also happens to be from the dominant ethnic group. He softens answers. He rephrases complaints. I have seen interpreters turn 'we are hungry' into 'they would appreciate some support.' Train your interpreter to repeat exact words. Use role-play. And if possible, run parallel interviews with a female assessor for women-only groups — because a man asking about menstrual hygiene products in a mixed tent yields silence, not truth.

That sounds like extra time. It is. But one bad assessment sends a million dollars to the wrong sector. A good one saves lives in hours. Which cost do you prefer?

Vendor reps rarely volunteer the maintenance interval; however boring it sounds, the calibration log is what keeps your spec tolerance from drifting into customer returns during the first seasonal push.

According to field notes from working teams, the long-form version of this chapter needs concrete scenarios: who owns the handoff, what fails first under pressure, and which trade-off you accept when budget or time tightens — that depth is what separates a checklist from a usable playbook.

Real-World Walkthrough: The 2017 Rohingya Influx

The First Week: Missteps That Cost Lives

When the Rohingya exodus hit its peak in August 2017, most agencies rushed in with food. Obvious, right? People were starving. But I watched a field team burn three days distributing high-energy biscuits to a crowd that had already walked through paddy fields for a week. The real crisis was hiding in plain sight—dehydration and open defecation. One coordinator told me: “We handed out rice while children were drinking from ditches.” That sounds brutal, but it's exactly how the first 72 hours usually go. The pull of visible hunger overrides the quieter, deadlier threat of waterborne disease.

— field coordinator, Médecins Sans Frontières, 2017

The Pivot to Water and Sanitation

The fix-first approach meant killing the food pipeline and rerouting logistics. Hard sell? Absolutely. Donors wanted photos of food distribution, not latrine slabs. But the needs assessment—which hadn't happened yet—finally showed that under-5 mortality was spiking from acute watery diarrhea, not malnutrition. We shifted every available truck to hauling chlorine tablets, jerrycans, and digging tools. Wrong order if you're thinking calories. Right order if you're thinking survival. Within ten days, the cholera outbreak was contained. That one leverage point—clean water—did what a thousand food parcels couldn't.

Measurable Outcomes Nobody Talks About

Most teams skip this: you measure success not by what you delivered, but by what stopped killing people. That sounds obvious. In practice, it means telling a donor “we didn't need your rice shipment” and watching their face fall. Harder than the emergency itself.

Edge Cases: When the Obvious Fix Isn't Right

Political interference

The textbook says: assess need, then act. In practice, that neat line gets shredded by the first phone call from a district commissioner who wants his own village listed as priority zone. I have watched a perfectly rational water-distribution plan collapse because a local power broker insisted the tanker stop at his cousin’s compound first. The cost? Two hours of daylight, and a cholera-risk neighborhood waited until dusk. The obvious fix—put the tanker where the diarrhea rates are highest—was technically right but operationally impossible without a back-channel deal: we gave the commissioner visible credit in the daily sit-rep, and he stopped blocking access. Political interference rarely goes away. You route around it, you trade something small, or you accept that the textbook fix will arrive too late.

“The cleanest needs assessment in the world means nothing if the gatekeeper demands a photo op before the truck moves.”

— logistics coordinator, Somalia drought response, 2022

Access constraints

Most teams assume the fastest route to the biggest gap is a straight line. Then the rainy season turns a dirt road into a mud river, or the only bridge was blown two weeks ago. The obvious fix—send the heavy truck with the most supplies—becomes the wrong fix because that truck will be stuck for three days. What actually works is splitting the load: lighter vehicles on the secondary track, even if they carry only half the water capacity. One trip gets through. Two trips get through. The single heavy truck? Still bogged down. The catch is that split logistics feels inefficient on paper. It is inefficient on paper. But paper doesn't have mud up to its axles. When access breaks, the first fix to abandon is the perfect one.

That hurts. I have made that mistake myself—insisted on the single high-volume convoy because the numbers looked clean. Three vehicles lost to a washed-out culvert. We fixed it by pre-positioning smaller caches at waypoints, accepting a 30% overhead in fuel burn, and getting water to the clinic 48 hours earlier than the straight-line plan would have managed.

Cultural taboos

You identify the need: malnutrition in children under five. The obvious fix is a high-protein supplementary feeding program. Except the community believes that porridge made from the provided grain is spiritually unclean for infants during a mourning period. Nobody says this in the first meeting—they just don't show up. The feeding center stays full of sacks, not children. The real problem isn't the supply chain; it's the failure to ask who eats what and why.

We fixed this once by swapping the grain for a locally accepted variety—same nutritional profile, different name, different ritual status. Took three days of elder consultations to learn the difference. Three days that felt wasted until attendance hit 90%. The obvious fix was wrong not because the data was bad, but because the data stopped at calories. Edge cases like these punish speed. They reward the slow, awkward work of listening before acting.

Limits of This Approach — and How to Compensate

Data Scarcity: Making Decisions in the Fog

The fix-first strategy assumes you know what to fix. Reality disagrees. In the first hours after a disaster, numbers are guesses — bandied about by officials who want aid and media who want headlines. You might have population figures from last year’s census, but half the roads are gone. I once watched a team spend six hours building a water distribution plan around a satellite image that turned out to be from 2019. That hurts. The workaround? Accept that your first assessment is a sketch, not a blueprint. Run a rapid reality check — send three people on foot to three different zones, not one person to the obvious site. Triangulate fast. Then adjust.

Most teams skip this: they treat data gaps as a problem to solve later. Wrong order. You compensate by building slack into the first 24 hours. Assume your water point count is off by 40%. Assume the number of displaced families is double what the local authority reports. Over-order consumables by a third. It feels wasteful. It isn’t — because the cost of running out mid-distribution is always higher than the cost of surplus. One concrete anecdote: during a cyclone response in Madagascar, the official count said 12,000 people needed shelter kits. Two days later, it was 18,000. The team that had ordered for 16,000 wasn’t scrambling. The team that ordered for exactly 12,000 was.

“You don’t need perfect data. You need a directional arrow that you’re willing to correct every six hours.”

— logistics coordinator, after the 2017 Bangladesh floods

Donor Pressure: The Clock That Breaks the Logic

Donors want photos. They want press releases. They want to show their taxpayers that money moved fast. That sounds fine until the pressure forces you to deliver water pumps before anyone has dug a latrine — because pumps photograph better. The fix-first approach collapses when your funder’s priority list doesn’t match the ground truth. I have seen a team deliver 2,000 tarps to a site where the real need was cholera kits. They knew it was wrong. They did it anyway because the grant agreement specified tarps. The compensation is brutal but necessary: pre-negotiate flexibility clauses into every emergency grant. Frame it as a safety valve, not an excuse. Say: “If my assessment contradicts the proposal, I will shift 20% of funds without new approval.” Most donors accept this if you explain why. The ones who don’t? Worth flagging—you may need to supplement their funding with unrestricted cash from a separate pool. Keep a small emergency buffer, even if it’s just $5,000, for exactly these moments.

Coordination Failures: Too Many Cooks, One Broken Stove

What usually breaks first is not the supply chain — it’s the meeting schedule. Clusters, working groups, sub-sector coordination cells — they multiply like roaches after rain. Every actor wants to fix what they see first, and nobody sees the whole picture. The fix-first method assumes a single decision-maker. In a real emergency, you have twenty. The catch is that coordination takes time you don’t have. How do you compensate? Designate one person as the “bad cop” of coordination — someone whose job is to decline meetings, redirect duplicate efforts, and kill irrelevant working groups before they form. Not popular. Effective. I worked with a WASH coordinator who did nothing else for the first week; she sat in a tent with a radio and a whiteboard, and she told every arriving NGO: “You are doing latrines in Zone C. Don’t argue. That’s where the gap is.” It felt authoritarian. It saved three days. The trade-off is bruised egos. The alternative is paralysis. Pick your poison.

Frequently Asked Questions

How to start with no money?

You do not wait for a budget line. I have seen teams freeze for three days because procurement hadn't approved water trucking. Wrong order. Start with barter, with borrowed satellite phones, with the local pharmacist who will front you oral rehydration salts against a signed receipt. The first cash you find should go to one thing only: transport for a three-person assessment team. Everything else can wait 48 hours. Without eyes on the ground, you are guessing — and guessing kills.

The real trap is thinking small money equals small action. It doesn't. A single prepaid SIM card and a motorbike driver can map five water points in an afternoon. That beats a $50,000 logistics plan that won't unload for a week. Start ugly. Start fast.

How to avoid donor distortion?

Donors love visible stuff — tarps, tents, branded clinics. The catch is that visible stuff isn't always what kills people fastest. Diarrhea kills faster than exposure in most hot-climate emergencies. But nobody photographs a bucket of chlorine. So you face pressure: spend on what funders can show on Instagram, or spend on what actually drops mortality. That hurts.

“The worst humanitarian decisions I have made were the ones that looked good in the proposal but wrong in the graveyard.”

— emergency coordinator, Chad, 2021

Push back by showing donors the math. One latrine slab costs X. One case of untreated cholera costs 12X in medical evacuations, funeral costs, and lost productivity. Frame it as efficiency, not ethics. Most program officers can defend a pipe budget if you hand them the numbers. If they still refuse, take the tarps — but spend the remaining 10% of your energy negotiating for the chlorine tablets nobody wants to fund.

When does relief become recovery?

Not when the funding line changes. That is a bureaucratic fiction. Relief becomes recovery the moment people stop dying from preventable causes at a crisis-driven rate. For the Rohingya response in 2017, that tipping point hit around week six — when diarrhea mortality dropped below the emergency threshold but malnutrition rates still climbed. The mistake is flipping the switch too early. I have seen agencies declare "early recovery" on day 20 because the cluster meeting agenda said so. The result: water systems half-built, hygiene kits diverted to "livelihoods training," and a cholera spike that erased two weeks of gain.

You know the transition is real when your logistics team starts complaining about concrete instead of tarps. That is a good sign. But keep one emergency response team in reserve for at least three months after you announce the shift. Because the first post-recovery rainstorm always floods someone. Always.