From providing households with toilets in urban disaster zones to mobile apps to trace lost children, the Humanitarian Innovation Fund is paying for new thinking in emergencies.
Kim Scriven, its manager, spoke to Thomson Reuters Foundation at the International Crisis Mappers Conference in Nairobi.
Why does the humanitarian system need to change?
It needs to change because of changes in crisis context - increasing hazards in severity and frequency - but also more engaged affected populations, wanting different types of assistance.
There are changes in the nature of humanitarian crises, like increased urbanisation. The Haiti earthquake really demonstrated that the programming approaches that agencies have been using have been developed in order to respond primarily in rural areas.
Those responses aren't really suitable.
With increased mobile phone ownership in the developing world, with increased access to Internet and FaceBook and with more disasters in middle-income countries, the profile of the people affected by disasters and conflict has changed.
Syrian refugees in Lebanon, or people now affected by the typhoon in the Philippines, are much more active in the way they engage with agencies.
And are aid agencies failing to respond?
How the system has tried to change traditionally is through codes and standards, through increasing accountability.
It hasn't led to the kind of transformational change that we need to see if we are going to have a humanitarian system fit for the 21st century.
Over the last 15 or 20 years, there has been a process of standardisation and bureaucratisation of the aid industry. As organisations have grown, they have become more and more concerned about having effective business practices. And that has probably lowered the risk profile of some organisations.
We need a much more open model that looks at capturing ideas that are being disruptive at the edge of the system and engaging with people outside the system and funnelling those ideas in.
That's what we try to do, encouraging people to share risks, encouraging them to collaborate and share costs of research and development.
Who are the best innovators?
Applications have been most successful where they demonstrate a level of collaboration. We have seen a lot of success with humanitarian agencies working in partnership with academia or the private sector, where they are able to bring in specific knowledge and skills, but also new ways of working.
There's something really powerful about bringing together those different skill sets and experience and using them to create novel ideas. We are trying to structure and broker those partnerships.
How much money do you give?
For our large grants, which are about $250,000, we do two calls a year and give out between four and seven grants in each call.
The funding that we provide is very much for testing and experimentation and comparing new ideas, new products and services with existing practices.
What successes have you seen so far?
Things that we have funded that have shown success are now going on to be replicated elsewhere. For instance, we funded a component of a beneficiaries' communications programme that the Red Cross ran in Haiti. It could do automatic voice recognition, like when you book your cinema tickets.
It was a telephone line that people could phone and could get information on a wide variety of issues like healthcare, sexual and reproductive health, disaster preparedness, access to aid distributions. It evolved based on people's needs. It was a free phone number that received over a million calls in its first year.
We funded a UNICEF project in Uganda with refugees from the Democratic Republic of Congo that's now being used in South Sudan and more recently in the Philippines. It's a rapid family tracing and reunification system.
Information used to be collected on a paper form. A child might cross a border and go to a transit camp and be registered by UNHCR or the Red Cross and get on a bus, with the paper form, to a refugee camp. It would take a number of days for that form to be found to re-identify the child.
What they did was create a mobile phone app so, whatever agency you are with, you could put the data directly into the mobile.
It would be centralised and allow the information to be shared much more quickly. And it could also have photographs attached.
Instead of that child getting off the bus and being on their own, there would be a representative of a relevant aid agency waiting for them. So they could immediately be assessed and be given appropriate care.