When access closes, the conversation has to stay open
A water truck didn't show up. A pump broke down in a camp. Several families on the same street are running a fever and have diarrhoea. Each of those pieces of information, if it arrives the same day, lets a team act the same day. If it surfaces a week later, during a routine field visit, all it can do is confirm the damage.
For years, humanitarian organisations have lived with that gap. Teams knew information moved too slowly and cost too much. What was missing was a realistic way to do things differently. That has changed in just a few years, under three pressures arriving at once: crises that drag on, access that keeps shrinking, and funding that no longer keeps pace with need.
In January 2026, the French journal Défis Humanitaires gave us the chance to look back at how SOLIDARITÉS INTERNATIONAL set out to close that gap, and what the experience taught us. This article draws out those lessons, well beyond the tool itself.
The real problem: information is expensive, and it arrives late
In humanitarian work, knowing costs money. Every needs assessment, every post-distribution check, every site visit means vehicles, fuel, staff time and lead time. Those costs are shared by every organisation in every sector, because they rest on the same constraints: travel, meet, write it down, enter it, analyse it.
Traditional channels remain essential, but each has its limits:
- Field visits are rich but slow and expensive. They become rare, or impossible, once insecurity, a disaster or an outbreak cuts off access.
- Hotlines fill up fast and only run during office hours. They tie up dedicated staff on largely repetitive questions.
- Paper surveys take weeks between design, collection, data entry and analysis. By the time the results land, the situation may have moved on.
Anyone who has worked in the field knows the outcome: questions that never get asked because the answer would cost too much, monitoring rounds that get spaced out for lack of resources, and decisions made on data that is already stale.
Why speed now matters so much
None of this is new. What has changed is how much it weighs.
Crises are lasting longer. The Syria crisis has run for more than a decade. In Lebanon, Haiti and the Sahel, teams work in places where emergency has become the norm. Needs shift constantly, and what was true three months ago may no longer hold.
Climate shocks are intensifying. Floods, droughts and waterborne disease outbreaks are more frequent and more violent. Early warning systems have shown that critical information, sent and received early enough, sharply reduces loss of life and livelihoods.
Operational decisions live or die by timing. Knowing in good time that a distribution missed a group of families, or that the first cholera cases are suspected, lets a team act immediately and set priorities. The same signal, ten days late, arrives after the harm is done.
Speed is not the whole story, though. Availability matters as much as velocity. Useful information has to be there when people need it, not just when a team happens to visit.
What the field taught us
It started with one simple fact, in Lebanon
The SOLIS Bot was born in Lebanon, against the backdrop of a long-term Syrian refugee presence and the COVID-19 pandemic. Movement was restricted, lockdowns came and went, and the need for information about services and how to access aid never let up.
One number changed everything: more than 80% of Syrian refugees in Lebanon used WhatsApp. The channel was already there, in their pockets. Humanitarian organisations just needed a way to use it in a structured, secure way, at scale.
The goal was twofold: give people a simple way to be informed and heard, and give teams a direct line to strengthen accountability, understand unmet needs and organise their work more effectively.
One tool, three very different realities
As deployments grew, the same idea adapted to contexts with almost nothing in common:
- In Lebanon, it is the entry point to the feedback and complaints mechanism across more than 300 sites, a way to collect data during needs assessments, and a channel for emergency messages.
- In Ukraine, it simplifies collecting donor indicators in war-affected, front-line areas that are hard to reach.
- In Haiti, it provides a permanent, discreet line of contact with residents of gang-controlled neighbourhoods where no team can go.
In 2025, running in three country offices and with five partner NGOs, it enabled 50,000 people to exchange more than a million messages with the organisations supporting them.
What the people using it say
The response on the ground has been largely positive, with satisfaction above 80% in our surveys. Part of the reason is simple: writing from your own phone, whenever it suits you, without queuing at a desk, is often easier than calling a hotline during office hours.
Inclusion doesn't happen by itself, though. Voice messages let people who cannot read or write use the service. SMS fallback reaches those without internet access. These are not nice-to-haves: they are what stops the channel from only serving the best-connected.
The numbers that change the equation
Cost is the most persuasive argument. During a food security needs assessment in Lebanon, our team sent a survey to 90 people and asked them to pass it on. Within 48 hours, 700 responses had come in.
| Traditional survey | Via the bot | |
|---|---|---|
| Direct cost | Nearly €2,000 | €8 |
| Time | Several weeks | 48 hours |
| Reach | Capped by available staff | 700 responses |
In the same window, we would barely have had time to organise the logistics to begin a conventional survey.
These savings come from very low unit costs: around €0.02 per WhatsApp broadcast message, and €0.20 per response to a ten-question survey. Multiplied across a project, the difference frees up thousands of euros for other work.
Other gains are less visible but matter just as much:
- our monitoring and accountability staff save an hour a day on average;
- hotline workload dropped by 40%;
- average response time to information requests fell 80-fold once simple questions were automated.
What it means, sector by sector
In water, sanitation and hygiene, fast alerts about a missed water delivery, a broken pump or a latrine that needs emptying mean a technician goes to the right place at the right time, instead of running blanket rounds. On a ten-site project, savings on fuel, labour and logistics are estimated at €2,000–4,000 a month, with fewer service interruptions for residents.
In food security and livelihoods, short recurring polls on local prices, input needs or requests for agricultural support let teams adjust in-kind distributions or cash transfers within days, where a conventional exercise would take weeks. For a project covering 1,000 households, the savings reach €1,000–4,000 per monitoring round.
In targeting, responsiveness changes the kind of decisions a team can make. At our partner CARE in Lebanon, community feedback led to redirecting 40% of food distributions to households that had been underserved. A decision of that scale depends on knowing, quickly and at scale, who is getting what.
The biggest win may not be the savings themselves, but what they make possible. Systematic post-distribution monitoring, regular satisfaction checks, frequent water service follow-up: practices everyone agreed were worth doing, and few teams could afford.
The blind spots we can't afford to ignore
A digital channel is not a universal answer. Our deployments taught us to look its limits in the eye.
It depends heavily on context. Gender, local politics, connectivity and communication habits all shape who responds and how. Older people, women in some rural areas and the poorest households are often the least equipped. A digital channel that ignores those gaps deepens the very inequalities it claims to address.
One data point is never enough. What comes in through the bot has to be cross-checked against other sources and fed into a validation and escalation process. No decision should rest on a single channel.
The risk is organisational, not technological. A digital tool becomes a problem when it is designed to replace people where people are indispensable. The bot should handle simple requests so that staff have more time for what matters: targeted visits, supporting the most vulnerable, following complex cases.
Data protection is a red line, not a checkbox. The only fully secure data is data that was never collected. That means collecting only what is strictly needed, informing people and getting their consent, aggregating data and setting strict access and retention rules. Some information should never pass through this kind of channel at all: shelter locations in conflict zones, medical data, political or ethnic affiliation, confidential reports.
No generative AI ever talks to the people we serve. Every message is written and approved by humanitarian staff. That is how we guarantee the information is reliable and the relationship stays transparent.
Before opening a digital channel: five questions to ask
Whatever tool you choose, the same questions decide whether it will help or harm. They are the ones we now ask at the start of every deployment.
- Who will this channel miss? Estimate phone and internet access by gender, age and area, and plan how to reach the people left out: voice, SMS, face-to-face.
- What will we do with the answers? Collecting information you have no capacity to act on erodes trust faster than not asking at all. Name who reads, who decides and within what timeframe.
- What must never go through this channel? Draw the line before launch, with your protection and security leads, not after the first incident.
- How will people know it's really us? A clear sender, a consistent number and awareness through trusted local relays protect against confusion and impersonation.
- How will we know it works? Set a handful of indicators and targets up front, so that the decision to continue, adjust or stop rests on evidence rather than enthusiasm.
The heart of the matter: the cost of information in an underfunded sector
Humanitarian funding is falling structurally while needs keep rising. In that climate, every euro spent on the logistics of information is a euro that doesn't reach people directly. The question is no longer just technical; it is strategic: how do you keep listening to communities when you have less to do it with?
Part of the answer lies in pooling. Building, hosting, securing and maintaining a tool like this costs more than any local NGO can carry alone, and few international NGOs gain anything by duplicating it. Sharing those costs across country offices and partner organisations changes the maths: everyone gets a proven tool without having to build one.
At project level, the return on investment is consistently positive. An annual licence costs far less than the budget needed to collect or share the same volume of information through traditional methods. Donors fund these licences as a way to make monitoring and programming more efficient, and the savings on surveys, spot checks and hotlines can go back into direct assistance.
Over time, a growing network of partners should bring the model to financial sustainability, drive unit costs down further, and turn the tool into a shared operational resource for the humanitarian community.
What we're offering
No single tool will solve access, cost and speed on its own. But some of the building blocks already exist, and they are worth sharing rather than rebuilding inside every organisation.
That is the spirit of what we are doing:
- opening free test lines for organisations that want to assess, in their own context, what a channel like this is worth;
- sharing what we've learned, mistakes included, about collecting and sharing information where humanitarian access is limited;
- building with local and international NGOs a tool that evolves with their needs, rather than a closed product.
If you work on these questions, or want to try this kind of channel with your team, write to us at solisbot@solidarites.org. We'd be glad to talk.
This article draws on "Le SOLIS bot, un lien permanent avec nos bénéficiaires" by Valentin Pistorozzi, SOLIS Bot project lead at SOLIDARITÉS INTERNATIONAL, published in Défis Humanitaires on 30 January 2026 (in French). Read the original article on Défis Humanitaires