Testing our prototypes in Northern Ireland
We are testing how we can identify more people who are affected by emergencies and need support. You can learn more about how we identified this problem from Harry’s blog post.
The team tested prototypes with people who experienced the 2017 flash floods in Derry-Londonderry. Following our testing in Cumbria and Barking we identified there were six stages to the process. We need to get each of these six stages right for our service to work well and I have broken down our learnings under the different stages to better show the user journey.
1. Becoming aware that support exists
We have been testing different prototypes along Alpha phase to see how we can make people aware that support is available to them.
We iterated our ideas from Barking (read the blog here). This included taking ‘British’ out of our branding to fit the local context. We also learnt from previous rounds that people would share information with other people they knew who had been affected. So we also changed the text of the Facebook advert to make it better for sharing, for example by including the support line number.
We shortened the Radio ad considerably and repeated important information, such as phone number, to make it easier for people to copy it down. In our discovery phase we learnt that people want support from people they can relate to. So one of our local Emergency Response volunteers recorded the radio ad to try and make it sound more familiar.
Our first hypothesis was people are more visible if we put adverts out about support on local radio. The feedback on the radio ad was still relatively inconclusive. A couple of people really liked this option, particularly those who didn’t use social media. Our second hypothesis that people are more visible if we put targeted Facebook ads out was a lot better received compared to the Radio ad. We learnt that people share and receive information about support available through word of mouth. By including the most important information on the Facebook ad, it became easier for people to share with those who otherwise wouldn’t see it.
“A lot of people find information on Facebook and would be more likely to come across this information rather than a Radio ad”
2. Considering whether you want the support available
When people became aware of the support available most people said they would call the number, however for the test we asked people to click on the link to our landing page.
This time we included more descriptive text on what support you might find at the rest centre. We also added images to show people what a rest centre and call centre might look like. We were testing the hypothesis that people need information about support to want it. We quickly learnt that perhaps there was too much information, and key details needed to be at the top of the page. For example, people wanted to have the contact details and information about the rest centre at the top of the page.
“There is just so much information you can take in after an emergency, that it’s better to have the contact details at the top”
3. Having your needs assessed and deciding on next steps
We were assessing the hypothesis that people want to speak to someone. In our previous rounds of testing most people’s preference was to speak to someone about their needs, therefore we iterated the support line. Once again it was well received. However we also had one interviewee who wasn’t comfortable making the call which shows the need to have different options available.
The updated call handler interface included a first page with the most basic information required when beginning a phone call,. The second page was for recording more specific details. To try and minimise typing for the call handler we added an auto-complete function, allowing them to focus more on the call itself. This would also help to retrieve records if a person had called before. The function helped us to avoid the situation of a person having to tell their story multiple times.
We were testing the hypothesis that people’s needs can be assessed over the phone. With the exception of one person who was not comfortable making a call, this hypothesis proved broadly correct. However, we learnt a lot about the structure and the content of the call. We learnt that people would generally be phoning with one or two priority needs and that we shouldn’t try and ask too much on the call.
“First call, you do not need too many questions”
To remain person led we have recognised it’s better to provide some support without people having to ask for it. For example, if there is cash assistance available people should opt out rather than opt in to receive this. The early aftermath of an emergency is a chaotic time and by asking too many questions we would be breaking our principle of designing for stress. Therefore, a follow up call to discuss other needs would be much more appropriate
We learnt that some people might not be in the headspace to talk in the early days after an emergency.
“I wasn’t speaking to anyone for days after”
We also know from previous interviews that people’s needs change substantially over time. From practical needs such as ‘where am I going to stay tonight’, to emotional support needs which may emerge later on. We need to design to ensure that people’s full range of needs can be met, at the appropriate time.
4. Receiving confirmation of next steps
Our hypothesis that people want confirmation of support and next steps led us to further iterate the confirmation email. Again the confirmation email was well received and people liked to know what to do next.
However, as with other findings above we found that less can be better with the amount of information provided. Some people we spoke to didn’t have an email address and therefore would need to receive this information in another written format. Some people would be happy with just receiving a confirmation text message.
“The last thing you want is to check your email at that stage, maybe down the line”
Overall, the format of sections divided to next steps for you, actions for us at Red Cross and sharing information with other organisations have been reassuring to most people. People liked being able to call back and would prefer to have a follow up with the same person since that naturally allows the emotional support to take place longer down the line.
5. Having the right resources allocated to meet your needs
Most people liked to have a reliable source to give advice; a way finder to reduce confusion. Some people heard contradicting advice after the flooding which made it harder to make an informed decision on what to do next.
“I would have liked to have had all this information back then”
6. Having your information shared with the right organisations to provide the support you need
For us to be able to action support it is important we are able to share information with the organisations that can provide the support people need. We tested the hypothesis that people are ok with their information being shared with other organisations. Most people have been fine with us sharing information with other organisations in order for them to receive the relevant support.
“I don’t care who supports me, I just want help from everyone I can”
We included the option to opt out from sharing as part of the confirmation email so people remain in control of what information is being shared about them, and with whom.
Some of the community groups we met expressed an interest in having access to some of the information collected so that they could provide a more effective local response. This is something we will think through more in the next phase of the project.
We are now coming to the end of our Alpha phase and are preparing for Beta.
