Quick question: how would you describe the experience of interacting with a local authority when moving house? When we carried out research we found that for citizens it is a journey that’s exciting, but often dotted with checkpoints that required the same data from us, over and over again.
Ali Rome, part of the Participatory Design team at FutureEverything, led a three-hour workshop at this year’s Service Design in Government Conference. At the workshop, they talked about this exact scenario and it led to a bigger conversation around public services, data and trust. The following blog is an overview of their talk and workshop at the conference.
Our relationship as citizens with the public sector is based on trust.
Trust is an important topic which lies at the core of public sector goals to become agile. An organisation needs to build trust amongst its teams to change the way they operate. In the context of data, these teams need to build trust amongst citizens so that they feel listened to, and especially throughout this period of change.
In this blog, we’ll talk about big data, automation, algorithms and human-centred design. Specifically how these components, if brought together in the right way, can help us create frictionless citizen experiences. We’ll use the example of moving into a new area throughout, to bring a practical example to our insights.
A little background on the public sector.
As citizens, we place trust in the public sector. Public services in the UK are the main point of contact between a government and its citizens, and the level of impact public services have formed the level of trust that citizens hold in their government.
Currently, our public sector and the delivery of services is going through rapid change. The welfare state was born in 1948; fast forward 70 years, the model for delivering public services remains unchanged in an environment with higher population figures and lower income tax. Less funding, more demand. Couple that with IndyRef, the Brexit referendum and all the subsequent elections and we’re in a situation where the Government needs its citizens’ trust more than ever.
We believe that by creating services that cater to the population, listen to them and answers their needs, this trust can be built. And data can help us, as designers in the public sector, build it.
First, increase trust by helping citizens make better decisions.
Humans make 70 conscious and up to 35,000 unconscious decisions daily. The more decisions you make throughout the day, the harder each one becomes for your brain.
We believe that more choice equals more freedom, but actual autonomy is only achieved when we have the correct knowledge and critical thinking to utilize the information or decisions offered to us. Take a look at local authority websites. It’s often information overload, and slightly intimidating.
This can be done through automation and categorisation as a first step.
Simple remedies could include categorising information on websites for easy digestion and consideration. Filters could be narrowed to help people make more accurate decisions by cutting out information that isn’t relevant. These two things make information easier to process and leave only the relevant information to actually process – they make each decision less taxing on your brain.
Automation is another step we can take to reduce decision making stress. It works by taking data, comparing it and providing conditions for action – for example: if this person lives in postcode area X, text them at 6 pm on a Tuesday evening with the message “Put your green bin out, garden waste is being collected tomorrow morning.”
The principles demonstrated in the Government Gateway program can be applied to local authorities. The Gateway gives users an I-D that allows them to carry their information over between services, saving them the frustration of entering the same data over and over again into different systems.
Think back to moving house. Data, which in this instance would be your name, address and NI number; wouldn’t change between signing up to pay for your council tax and registering to vote, so arguably there’s no need to enter it twice, but it’s something citizens are required to do. This could also be taken a step further – the data could easily be used to automate something like the communication between local authorities and citizens such as text alerts to prompt you to take your bins out (we got this idea from Falkirk Council) or to alert you to any relevant community schemes happening in your area.
Big data is an opportunity that comes with challenges.
We have access to an abundance of data in the public sector; collecting big data is relatively easy but using it is hard. The challenge lies in our ability to draw useful and accurate conclusions, and this ability can be strengthened by our trust in four things. These are:
- Trust in datasets + analysis
- Trust in bodies collecting data
- Trust that data will be contextualised
- Trust that data is anonymised and secure
Also, data without context is just that, data. It’s our job as Service Designers to give this data meaning, and there are two ways in which we can do this. The first is by data analysis; mapping it, comparing and interpreting it. And the second (which we’ll talk about a little later in the blog) is to talk to citizens.
We can create less friction through big data and algorithms.
To make sense of big data we need algorithms, which in their most basic sense define a set of instructions to achieve an outcome. For example, the data shows that people who completed the steps to register for their council tax also registered to vote. So, we can create an algorithm whereby if a person registers for council tax, they’ll then be recommended a link to register to vote. This process alone takes out a few cumbersome decision-making steps, and in turn, generates less friction.
To create an algorithm we need two things:
- Data: Things that have happened in the past.
- Definition of Success – The next steps based on that data.
We are in a position of responsibility to use algorithms properly, as there are so many instances in which we can go wrong. Algorithms don’t always mean fairness; when they’re working on historical data they can be biased and repeat past mistakes. We need to check algorithms for fairness and acknowledge bias. Predictive policing is an example where existing biases may be reinforced – if historical data is biased, for example as a result of certain types of crimes not being reported to the police or of racial profiling by the police, then the algorithm and the consequent output could be biased as well.
For this reason, it’s important to use trustworthy data sources and to provide feedback loops to allow the public to demand accountability and be more engaged in the development of services for them.
We can make big-data accessible to citizens.
There’s only so much understanding we can gain from quantitative data. It’s the qualitative data – that cannot be quantified – that really helps us understand behaviour. Qualitative data provides context to the big quantitative data, and we need critical thinking capabilities in our teams to create it. Humanities subjects like philosophy, psychology and ethics give us context.
Often this activity involves us speaking to citizens, and for them to be open with us, they need to trust that the data they give us is being used for the right purposes. This can be done in two ways:
- By ensuring that the data is open and by making it legible. What we mean by this is making sure that not only is the data available to the public to see but that they can also understand it and therefore utilize it.
- Give citizens ownership of data. We can allow open access to anonymised data and encourage citizens to explore it. Encouraging citizens to play with data through open data challenges gives the opportunity for citizens to see what is being collected and for them to come up with uses for that data based on their own needs.
If we’re asking people for their data we can also build trust by being seen to provide a useful service in return. People are of course wary about being monitored too closely but if they understand the reward they’re going to get, they can be more than willing to provide the necessary data. The same could be true if they know it’s for the greater good – for example for changes to public infrastructure.
Human-centred design gives us tools to turn information into knowledge.
Human-centred design is an approach to problem-solving that begins with the people for whom we are designing and ends with new solutions that are tailored toward their needs. It gives us tools to build empathy with people to help understand their experiences, interactions and behaviours around the things we build. To fully create frictionless experiences, we have to understand the perspective of the people we are creating these experiences for.
Big data gives us the hard facts. The how, when and where of actions taking place. Analysis of data gives us interactions to focus on, surges in activity, or the typical way people might carry out certain tasks. But it doesn’t give us the ‘why’ behind the actions, it doesn’t give us context, and that’s where we use human-centered design tools to firstly help us have the right conversations, and secondly to synthesise the qualitative data accurately.
In the public sector, we are in a fortunate position in that we have access to data, the tools to gather analysis and use it to uncover areas where further context is needed. We have the opportunity to gather qualitative research by speaking to the wider public. Done right, we can start to design interactions, services and environments that tackle the challenges they face on a day-to-day basis.
How we do human-centred design.
At FutureEverything, we follow processes outlined in the double diamond.
However, we have modified our approach to align with the ecosystem in which we are designing experiences. This is around the first stage, in which we try to find the problem to solve. Data gives us some of this knowledge, and at the second stage, we go on to determine whether the problem we chose is really the right one to solve.
Let’s go back to the moving house example. Imagine if our goal is to attract people to move to a specific postcode. Data analysis tells us a neighbouring postcode experienced a surge in new households and the initial analysis of the data assumes it’s most likely due to low crime rates. Now, conversations through research (qualitative data) may tell us that in reality, the reason is the rise of independent retailers.
FutureEverything help design citizen data input in IoT project, CityVerve.
The workshop reflected much of the work our team carries out in participatory design projects in Government – national and local. In the CityVerve project, we use quantitative and qualitative data to help develop IoT services that are shaping Manchester and its Citizen experiences. We do this by building trust through value exchange; we create events that citizens will gain value from, and in exchange, we gain the opportunity to ask them questions to help with our research.
We also do a lot of listening on social media, and at local community groups. For us, using human centred design is key to really dig deep into citizen experiences, in that we try to shape the development of some of its initiatives to align with their needs.
In our objective to build trust, we carried out workshops with citizens to help co-create Community Key Performance Indicators for the IoT and Smart Cities. This gave citizens a voice in how the technology is being developed, and, gave them the lead in how to measure success. We have made this process open for local authorities to implement, too.
If you’d like to have any questions or would like to chat, just get in touch.