Text and Voice Messaging Service for Newcomers
TeamMichael-Owen Liston: Project Lead, Research, Design
Manu Kabahizi: Project Co-Lead, Research, Design
SkillsResearch, Project Management, Mobile Service Prototyping
ContextResearch and Service Prototype for Immigration, Refugees, and Citizenship Canada
An independent pilot project for Immigration, Refugess, and Citizenship Canada
We’ve been prototyping a service to connect refugee newcomers to volunteers via SMS text and voice messages. This project was undertaken as a response to an RFP from IRCC for innovative activities to support Syrian newcomer settlement in Canada. We were selected to explore the potential of an SMS messaging service in this context, as part of a rapid month-long research and prototyping sprint.
Why was this necessary?
By March 2016, over 25,000 Syrians had arrived in Canada as refugees, over a period of several months. Of these people, over 17,000 were Government Assisted Refugees, meaning that they had been identified by the UNHRC as being extremely vulnerable, and in greatest need of humanitarian assistance. The recognition that this population of newcomers to Canada would have particular challenges accessing resources to facilitate settlement—and might well have particular needs unmet by current services—led IRCC to put out a call for proposals to explore the space, try to better understand the challenges, and prototype new ways of addressing them.
We proposed using a text messaging service as a platform, based on the success of similar programs worldwide, and our shared experience developing interventions for underserved populations using mobile technologies.
How does it work?
Newcomers can send questions about anything related to settlement or daily life—like who to contact about employment programs, or where to find specialty foods—and volunteers who are currently available will receive a notification, and the opportunity to respond to the question with a text or a voice message.
The Story of the Project
Where we began
At the outset of the project we were interested in the potential of an SMS-based messaging service as a potential intervention. We imagined that text messages could provide an accessible, “low-barrier” way of interacting with an automated service that could respond to requests for information in Arabic, 24-7. We drew inspiration from similar projects around the world, including a number that we ourselves had been a part of previously.
However, our research rapidly overturned one of our fundamental assumptions—that the Syrian newcomer population currently entering Canada is overwhelmingly tech-savvy and comfortable with text messaging. There is indeed a truth to this popular conception of “the Syrian refugee”, but it in fact highlights inequities that have existed in Syrian society for some time; the demographics of the people entering Canada under the Goverment Assisted Refugee program were vastly different from those crossing Europe over land, or arriving in Canada as Privately Sponsored Refugees. This population has typically below high school levels of education, no spoken or written English or French capacity, and—critically, for our assumptions about text messaging—are often illiterate in Arabic, as well.
Where we ended up
While text messages could still provide a valuable medium of access for many of the potential participants in our proposed service, it was clear we needed to find ways of meeting the needs of people, regardless of their literacy level.
The platform we were using for rapid prototyping the service, TextIt, has robust support for creating Interactive Voice Response (IVR) services—the familiar telephone customer service “voice menu”—that can be deployed in parallel with a text message system. We were able to quickly prototype an asynchronous voice message service within the platform, that could be used as an alternative to sending and receiving text messages.
Another opportunity that we identified in our research was the extent of the existing pool of knowledge about settlement issues that lived within the community of volunteers, settlement workers, private sponsors, and cultural groups at large. The resources are out there, but the challenge is one of awareness and access for newcomers who may have high needs and extremely limited resources. Rather than proposing a fully automated database of information as a starting intervention, we saw an opportunity in the human “database” that was already deployed at scale—how might we multiply the points of contact between community knowledge and newcomers?
By the end of this research and design sprint, our prototype was focused on testing the potential of anonymized messaging between community participants and newcomers, with relatively little automation. We believe that this would allow a valuable service to be deployed much more rapidly, with greatly reduced upfront development cost. This service could be used to train machine learning and natural language processing models, however—through use, such a service could build the capacity for the system to be able to handle some informational requests automatically.
Who we are
We are Michael-Owen Liston and Manu Kabahizi. Between us we have extensive experience in interaction design, software development, technology consultation, and projects in the community development space worldwide.
Please contact Michael with any questions or inquiries: firstname.lastname@example.org
Is this a real, working service now?
Our mandate for this project was not to bring a fully working service to market within a period of two months. Rather, our goal was to use our project proposal as a kind of provocation, and learn about the potential for such a service from the people who would use it. It was exciting for us to see government taking this kind of iterative, design-driven approach to such a complex problem space, rather than committing from the start to a preconceived, infrastructure-heavy “solution”, even before the dimensions of the problem is understood.
What are you doing next?
Response from participants in our research and prototyping was very positive. At time of writing we are exploring opportunities to continue development and implementation of the service for the Syrian newcomer population, but we are also keen to explore potential interest in this model of service platform for other newcomer populations, as well as more diverse audiences.
If you have a look at some of the project pages in the links below, you’ll see there are lots of similar-but-different ideas about how this kind of service could be used to help people—from finding settlement services to accessing mental health resources—and some great examples of like-minded projects “in the wild” already!
Links to other projects using low-barrier technologies like SMS in this space, news articles, and resources about chatbots, conversational interfaces, and natural language processing in general.
- Refugee Text Service (Denmark)
- SMS’UP (France)
- Welcome! (Sweden)
- Services Advisor
- Welcome Ontario
- MCIS: #Interpreters4Syria
- Crisis Text Line
- How cellphones are redefining the Syrian refugee crisis
- New initiatives emerge to help refugees
- Smartphone use on the refugee trail
- How to make apps that help Europe-bound Syrian refugees
- Text a Syrian: a New Service Will Help Refugees Navigate Life in Canada
- The app that connects refugees with Swedish locals
- Mobile for Development — Partnerships between MNOs and Innovators: Ideas from MWC
Chatbots and Conversational UI in General
- Raw Data Podcast, Episode 11: So… What’s On Your Mind? (includes segment on Crisis Text Line)
- R U There? A new counselling service harnesses the power of the text message. (Crisis Text Line)
- The search for the killer bot
- Conversational User Interfaces
- All Talk and No Buttons: The Conversational UI
- Chatbots Magazine (Medium.com)
- The Next Phase Of UX: Designing Chatbot Personalities
- Invisible Apps (Chatbots.org forum)
- Smartphone-Based Conversational Agents and Responses to Questions About Mental Health, Interpersonal Violence, and Physical Health
- The next hot job in Silicon Valley is for poets
- Bots, the next frontier
- How to build a better bot
- Conversational interaction design: constructing context
- Chat bots, conversation and AI as an interface
- As Messaging Apps Boom, Brands Tiptoe In
- IPA research reveals how mobile phones improve reproductive health knowledge
- SMS Based Disaster Alert System In Developing Countries: A Usability Analysis
- How to Prevent a Plague of Dumb Chatbots
- When do bots beat apps? When context and convenience matter most