Pre-positioning a network of field specialists: RapidPro for Emergency Response

Launched: October 2016

By Elliot McBride Innovation Specialist, UNICEF Pakistan

Flooding, earthquakes, attacks and complex personnel displacement are all common emergencies that have plagued Pakistan since its birth as a nation. Mobile phone usage in Pakistan is high, with SMS making up a large percentage of user communication. In 2014 alone, 302 billion (with a b) SMS were sent within Pakistan. Emergency situations demand innovations that support real time information to bridge the gaps between affected communities and humanitarian actors. In Pakistan, the UNICEF team saw the potential for RapidPro to fill this gap in the event of a crisis. To achieve this and foster responsible prototyping of real-time information channels in emergencies, a new order of operations has been developed within RapidPro programs.  In so doing, we’ve been able to better utilize the real time information component of RapidPro and apply it to delivering vital information about emergency situations from relevant sector experts directly to decision makers. In doing so it has created a highway of information about emergencies from field sector experts directly to decision makers, in real time.

Flip the script

In the past, registration of RapidPro users occurred immediately prior to (or during) beneficiary registration. This is where relevant survey data is taken such as the questions which took users hygiene kit feedback from flood affected regions. Typically the user’s participation is then over. Results are sent back to Islamabad for analysis and the next user registers. In the case of mapping a baseline of WASH access in urban areas, RapidPro registers user’s information at the beginning of the survey regarding WASH access and use, RapidPro then deactivates the user and collects the data.

Though this model has proven effective, we’ve recently been exploring a new approach to pre-position  ‘mobile responders’ by registering field workers across Pakistan as a preparedness strategy. Registrations have occurred in conjunction with trainings by IOM and UNOCHA, where UNICEF has developed an activity that can be added to trainings of any nature, for field workers who are proficient in emergency first response across various sectors (WASH, Food Security & Livelihood, Cash, Health, Shelter) to self-register on RapidPro and join the growing network of digital responders across the country.

This activity, together with a short informative brief, registers participants to a group in RapidPro called Emergency_database.  The registration survey contains only the details necessary for targeting specialist emergency workers in specific field locations, Within this instance for Pakistan, it was determined that name and gender of the emergency specialists are irrelevant to field reporting and are therefore not collected.

The RapidPro flow collects information about workers region and work sector for follow up after events. This is a screen grab of the two critical stages used to a) map and define workers to the database, then b) retrieve specific engaged specialists based on location and specialisation. For example, flooding in Chitral can be met with a situation report poll which is sent only to field officers in WASH sector who are based in Chitral.

a) Defining and storing emergency workers by field. (You can find out about how to create a registration flow here)

(Above) there must be a new group for each sector for follow up. Also collect responses for ‘other’ sectors as analysis can determine if you need to add more later.

(Above) there must be a new group for each sector for follow up. Also collect responses for ‘other’ sectors as analysis can determine if you need to add more later.

b) targeting select groups by sector and region after an emergency. (You can find out about how to use dynamic groups for targeted communication here)

b) targeting select groups by sector and region after an emergencyb) targeting select groups by sector and region after an emergency

b) targeting select groups by sector and region after an emergency

Note: it is important to leave the Other option after the split by expression blank with no follow on connections (above), this way, no other messages will be sent to workers who are not Chitral WASH experts. These JSON files can be delivered on request (

When information reaches Islamabad innovations, it will be cleaned and passed to officers who implement emergency response plans. This information will help triangulate a more accurate picture that will allow existing UNICEF emergency responses with a unique perspective in order to understand beneficiary needs typical to each disaster. Ultimately it is the goal of this project that emergency response will become a more efficient practice for UNICEF, with assistance that meets the priorities of the beneficiary before the outset of response, rather than addressing all emergency situations with a similar approach and discovering nuances that could save lives once arriving at the disaster site.

Further advantages of this design include that it is not situation limited, random emergencies can be factored into the programme on a needs basis by accessing the pre-identified database of field workers depending on the emergency. This process differs dramatically from other RapidPro project formats, where the survey information is gathered simultaneous with the participant registration.

After a pilot round of training, 89 emergency field officers were registered on the RapidPro emergency database:

Pre-positioned field specialists, registered with RapidPro, UNICEF, Pakistan - 19/10/2016

Pre-positioned field specialists, registered with RapidPro, UNICEF, Pakistan - 19/10/2016

There are many ways in which RapidPro surveying can be used as an emergency response platform. Below shows this format which aims to extract information most valuable status updates in the present tense to give the most accurate picture of a community, post emergency.


Toward a pre-positioned  future of ‘mobile responders’

The initial goals of this project will be to begin the process of developing a large and dynamic database with registration containing enough information to educate workers about the importance of responding via SMS during an emergency. UNICEF WASH and Disaster Risk Reduction have both developed the database concept to fit the emergency prone needs typical to Pakistan, as such if this is database is scaled to other officers, it will need to be tailored to each country's specifications. Engagement strategies and education surrounding accurate registration must be strengthened in order to develop a platform which is accurate and reliable. The longer term goal has become normalizing the practice of SMS use as a valid tool for emergency response. This can achieved through effective upscaling, repetition and education strengthening.