Background
SmartStart was launched in December 2016 to give expectant parents step-by-step information about government services, and help them find and apply for the right services for them and their baby. In its first six months, there were 85,000 visitors and more than 5,000 due dates entered.
As part of the Lab+ work in the Service Innovation Lab, the public sector is exploring new ways to improve people’s experience with government. With the shift of many sectors — public, private and non-profit — into digital, our expectations as ‘customers’ of government services has shifted. We expect more.
So what’s a public sector to do? Well, at Lab+ we focused on 2 key ideas:
- Putting people at the centre of service design to better meet their needs.
- Thinking about how we support Government-as-a-platform (GaaP), which is about creating building blocks (or common components) that can be used anywhere in the ecosystem of providing government services.
See Pia Waugh's recent blog post on "Lab+: industry and community needs for building on ‘government as a platform’" for more on the GaaP idea. And if you’ve been keeping up with the Lab+ blog posts over the last few months, you might have read a post on “Lab+: Norming and performing”. In the draft Lab+ work plan Pia Waugh talks about, one of the goals listed is: Conduct the discovery/design for two life events or services.
So we thought taking a fresh look at SmartStart was a great idea. It’s already successful (did I mention the international award it’s already won?) but how could we build on that? Which fits neatly the Lab+ goal to put people at the centre of service design .
And that leads us to a design sprint
The team for this design sprint included Adele Kitto (IR), Beth Davies, Indi Rawiri and me (all DIA). Creative HQ facilitated the design sprint by keeping us on track, asking the hard questions and challenging us when we needed it. To be honest, CreativeHQ’s Brett Calton (Innovation Consultant) and Lingy Au (Programme Manager) contributed as much to the end results as the rest of us so it was a very egalitarian process.
A “design sprint” is pretty much what you’d expect from the phrase: a fast sprint to create something. In our case, it was to build on the original SmartStart's idea of a user-focused service designed for expectant parents. The fast part is to keep the focus on what needs are important — in this case, the parent’s and the baby’s — and not on the government agencies that provide the services.
The design part is using these insights to build a service prototype to meet these customer needs. A design sprint has 5 phases. For us, this meant 1 phase per day — Monday through Friday — with the whole team working in the same room. And a whole lot of sticky notes. The phases are:
Design sprint phases: Map, Sketch, Decide, Prototype and Test.
I won’t spend much time on an in-depth explanation of a design sprint since there are lots of websites that cover it in great detail. For me, the main insights were:
- this is a team activity; everyone has to jump in with ideas, facts, thoughts etc
- imagination is critical; think past ‘what is it now?’ to ‘what if...?’
- you really have to work step-by-step; no jumping around (as tempting as it is) because you’ll just end up with a ‘solution’ that doesn’t fix the problem
- no matter how sure you are on Thursday that your team has come up with the most amazing idea EVER…testing on Friday will bring you back to reality
- the idea that survives testing is the basis for a solution that’ll work.
Mapping
So, back to SmartStart. The team worked out a problem definition, the outcome we were aiming for, and who we thought were SmartStart’s main customers.
Problem: New and expectant parents (particularly those who are more vulnerable such as teen parents) who don’t know what services they are entitled to. And once they do figure that out, it’s really hard for them to apply for services provided by multiple agencies and NGOs.
Desired outcome: New parents can easily find out what services are available to them, based on their particular circumstances. Ideally, they would only have to provide information once (like due date or address) when applying for multiple services.
Target customers: New and expectant parents that are particularly vulnerable, under 20 years old, who are on social welfare benefits (or will be eligible with a new baby).
Sketching
Building on the mapping phase, we spent a day talking through how the system works now, and what issues were likely to be a problem. Since most of us had been in the public sector for a while, we had a deep understanding of the challenges involved in creating a service that depended on agencies’ goodwill to work together. That’s a whole other discussion but from the team view, we made the assumption that agencies would provide joined-up services.
Luckily, our team included people who were involved with the original SmartStart so they had a lot of insights and facts available to answer some of our questions. And 2 people on the team were parents (1 with a 20-month old baby) so they let us tap into their own perspectives and experiences for guidance as well.
Many sticky notes later, we arrived at a sketch of the current system and where we thought our 'SmartStart version 2.0' could improve things for our target customers. Most importantly, that we created a diagram of the current process and worked out where we could improve that process, so that's what we focused on.
Deciding
After we worked out what we wanted to do, we spent a day bouncing around ideas of how to build it. More discussion and swapping stories, thinking of how similar problems were solved, adding some pragmatic ideas, throwing in some crazy ideas, combining where it made sense, coming up with what-ifs and wouldn't-it-be-cools, and talking everything through. By this this stage of the process we had built a good team so we came up with lots of possibilities. Distilling all these into one approach took some effort, but everyone seemed to think we were onto something good.
Prototyping
The ever-resourceful Creative HQ team brought Kelcey Braine (Design Consultant) in to magic our stacks of scribbled sticky notes into a set of smartphone screen designs that looked fantastic. These allowed us to test the ideas out by using illustrations that really captured our idea. The approach was to ask expectant parents a series of questions to determine what services or benefits they might be entitled to. The prototype screens show the 8 steps we envisioned for the improved process:
- Welcome to SmartStart. Where you can begin by adding your due date, learn about pregnancy phases, look at a To Do list, find out what services or benefits you might be entitled to, or to register your baby online.
- Are you getting all the help you need? This allows you to choose the types of support you want including emotional, financial, health, housing or to stay in school (for younger expectant mothers in secondary education).
- What is your relationship status? The first in a series of questions to determine what an expectant mother might be eligible for; criteria for some benefits are based on what kind of relationship the parents have.
- Are you a New Zealand Citizen or Resident or Other? There are some differences in services that are based on the parent's residency status.
- Thanks! Based on information provided, this screen shows the services that the expectant mother may be eligible for; some services will require verification of answers.
- Awesome. This screen gives expectant mothers the option to apply for the selected services; this is the point when parents would need to register with SmartStart to apply for the services.
- Hi, could you tell us a little bit more about yourself? This screen would get the expectant mother started on providing additional information to verify eligibility, based on the applicable business rules; for example, some benefits are only available to people working part- or full-time.
- Great, thanks! In our ideal world, once an expectant mother answered all the questions, she would be given the chance to consent to SmartStart applying for those services on her behalf directly with the agencies that provide them. An alternative to that would auto-complete the right forms with the details she provided so that they could be printed, signed and sent off, reducing the need to repetitively provide the same details to different agencies eg address, date of birth, full name.
The prototype screens show the 8 steps we envisioned for the improved process.
Testing
Friday. The big day for trying out our idea. A bit daunting to be honest, but exciting as well. To test our prototype, Creative HQ recruited people to try and get the best matches to our target audiences. So we had 5 pregnant women take the prototype on a test drive. Beth Davies from our team was our official test leader. To keep things as consistent as possible, all participants:
- viewed the prototype on the same computer
- were all interviewed by Beth
- were asked the same series of questions
- were encouraged to tell us about their thought process; for example, "I would click on this button to because I'm most interested in financial support."
The rest of the team watched the testing from another room. That way we could see and hear everything but wouldn't make people uncomfortable. Overall, the feedback from was people was positive.
Lessons and surprises
We did have some unexpected responses to some of the prototype designs.
- Plain English. While we thought that we'd written plain, straightforward text for the prototype, we confused some of our testers. So we'll need to spend more time crafting the text to avoid any misunderstanding.
- Context is everything. Some of this confusion was that we didn't give people enough context. For example, we just leaped right into asking questions to determine their eligibility for services. But we never said that. We need to create a 'guided journey' to services not a 'DIY journey'.
- Explain what's in it for them. Building on these points, we could include some sign posts, warnings and opt-in/opt-out points. For example, we could have:
- started with a statement, "If you answer these questions, we'll let you know what services you and your family might be eligible for." so that people are clear what's going on.
- added some reassurance, "Your answers will not be shared without your permission."
- let people know the reason for SmartStart, "Once you know what you're eligible for, you can sign up for SmartStart and we'll help you apply."
- point out the benefit of answering questions, "If you don't want to sign up for SmartStart, you'll still have a list of services and providers so that you can apply directly."
- moved some questions to after registration; most applications include a request for bank account information to deposit any financial benefits. We should have skipped this in the pre-registration questions because it made people suspicious.
Wrap up
Personally, I learned a lot during this design sprint. Not just about SmartStart but about service design, how a design sprint works, user interfaces and a whole bunch of intangibles. This week was equal parts amazing, frustrating, exhausting, exhilarating and ultimately a great experience. If you have the opportunity to do this yourself, grab it.
But beyond how valuable this experience was to me or anyone on the team, the real win was that we ran a proof-of-concept process that resulted in some pragmatic insights on how we can improve services for New Zealanders. Which gets us back to where we started: putting people at the centre of service design to better meet their needs.
Lab+ is housed in the Service Innovation Lab, which is an experiment carried out under the leadership of the ICT Partnership Framework’s Service Innovation Group. It's managed by the Service Innovation Team in Department of Internal Affairs in partnership with Assurity Consulting.
Check out earlier blog posts about Lab+ and the Service Innovation Lab.
Follow us on Twitter at @NZLifeEvents.
