Citizens Advice Bureau Sprint Report

During December 2017, the Service Innovation Lab (LabPlus) facilitated a week long collaborative sprint with the Citizens Advice Bureau (CAB) that focused on understanding the customers’ journeys and user needs of people who use the CAB services.

Overview

In this one week sprint there was a particular focus on the CAB's website as they are embarking on a upgrade of their online systems, but face-to-face customer needs were also taken into consideration.

No customer information was shared to ensure privacy of CAB users. The motivation for CAB was to leverage the skills of the LabPlus team in designing their new systems. The motivation of the LabPlus team was to test some assumptions about whether CAB clients would use digital government services even if they existed or were well designed. Testing this assumption was important to understanding policy and service design opportunities moving forward.

The LabPlus team were not entirely surprised to find that most CAB clients would not go to an agency in the first instance due to high complexity needs including clients in vulnerable situations, english as a second language and seniors. This reinforced a need to consider CAB and other non government service providers as part of the user journey, and thus consider how we can support and enable these entities to help New Zealanders. The policy and service implications of this finding need to be further examined.

This sprint report describes:

• how the collaboration came about,
• the approach the team used to explore the journeys
• what was learnt

What is Citizens Advice Bureau (CAB)?

CAB provides services to help people understand their rights, obligations and entitlements and how to use this information to get the best outcomes, provide people with the confidence and support they need to take action, and work for positive social change within communities and wider society. CAB provides a free and independent service to all.

The aims of CAB are to:

• ensure that individuals do not suffer through ignorance of their rights and responsibilities, or of the services available, or through an inability to express their needs effectively.
• exert a responsible influence on the development of social policies and services, both locally and nationally.

The service is provided by more than 2,300 trained volunteers, in more than 80 locations around New Zealand. The volunteers are supported by a comprehensive database and regular learning.

Why collaborate?

CAB is planning to upgrade CABnet (their online systems) in 2018. This includes the enquiry form that volunteers use to detail each enquiry they are involved with, the backend client management system and their website. CABnet also holds their comprehensive database (currently ~2,500 FAQ) that is used by their volunteers when providing advice, and also by anyone searching for information on their public-facing website.

CAB are using feedback from their volunteers and CAB managers for the upgrade to CABnet, however CAB have limited information about the needs of customers who use their website. They are unsure whether the people that come in to visit a CAB physically are the same as those who use the website, and if not, how can the needs and preferences of the people using the website be recognised in the upgraded version.

For the LabPlus team collaborating with CAB is helpful because CAB represents another way that people access information to help them solve problems in their lives. CAB is a non-government organisation so it has a different relationship with customers that government agencies don’t have, which puts them in a unique position to describe how our customers find interacting with government.

CAB have over 500,000 enquiries per annum where they assist customers with a wide range of issues, many of which relate to navigating government to get their entitlements or comply with regulations. Considering the size of the annual enquiries and the CABnet database, there could be opportunities to reuse government entitlement rules and information to integrate with CAB and other non-government service providers.

For CAB, working with the LabPlus team was a chance to experience working in an agile way. CAB team members were interested to see what working in an agile way is like, how it could work for CAB in the future and learn new techniques and tools that could apply to CAB and their development of CABnet.

The approach

We had a small core team of five members, and our sprint length was just one week so we had to make sure that our expectations of what we could achieve in a week were realistic. Also with three of our team members having very limited knowledge about CAB and what they do, a lot of the week was devoted to making sure that everyone on the team had an understanding of CAB’s business, goals, and the varied advice and services that CAB’s can provide.

During the week we completed these activities:

• Sprint planning session to define the work, team availability and charter, identify tasks/ activities for the week.
• Worked through a customer canvas that identified:
  • Customer(s)
  • Problems
  • Unique value proposition
  • Features
  • Channels
  • Metrics
  • Risks and assumptions
  • Experiences and behaviours
  • Needs and goals
• Gap analysis
• User and subject matter expert interviews
• Data analytics review of current CAB website
• UX review of current CAB website
• Create journey maps and lo-fi prototype of a future CAB website
• Sprint retrospective

The journey map we created for the current experience of users of the CAB included customers that:

• physically visit the CAB,
• use the website
• or make contact with the CAB over the phone, email and live chat

To do this we used information that CAB provided (e.g. feedback from NGO’s that use the CAB website and statistics about how many people use different channels etc.). We also visited a CAB and spoke to the volunteers about their experiences, sought input from subject matter experts, and the knowledge of members of the sprint team.

Using our journey map as a basis we were then able to identify the pros, cons and unique features of each of the channels that CAB delivers service through; in-person visit to a CAB location, phone, Language Connect line, email and online live chat. By analysing each of these channels, we could start looking at how the some of the features that are useful about other channels could be incorporated into the CAB website.

What we learned

Learning about CAB

Before this sprint the knowledge among the LabPlus team of the functions of CAB was relatively limited. While we had all heard of it and had some understanding of some of the kinds of queries CAB handles, we did not have a full appreciation of the range of services they provide. Therefore, the most valuable learning for some of the team was understanding the process of a CAB enquiry, the diversity of CAB’s role and the broader goals of the CAB.

One of the CAB aims is to ensure that individuals do not suffer through ignorance of their rights and responsibilities, or of the services available, or through an inability to express their needs effectively. Put simply, to help people individually. CAB provides free and confidential advice and does not limit the questions that you can ask, or the subjects they can help with. They can answer enquiries about anything from small business advice to how to start the process of adopting a child and everything in between. CAB’s goal is to always provide each customer with an idea of the next step(s) they could take.

The other aim of CAB is to exert a responsible influence on the development of social policies and services, both locally and nationally. More broadly, to help people collectively. CAB takes what they can learn from serving individual clients into the broader social policy arena, uses it to advocate for socially just policies and services.

Some of the issues they have been recently advocating for include:

• better security of tenure for renters;
• better quality rental housing;
• employment agreements for all;
• adequate supply of decent emergency accommodation.

CAB also provides or facilitates the provision of many other services. CAB runs a service for people who do not speak English, or would prefer another language, to call and speak to someone who speaks their preferred language fluently, and get the same CAB service as would be provided in person. This is called Language Connect and is currently available in 16 different languages. Some CAB locations also offer other services such free legal clinics, budgeting advice services, the provision of emergency food parcels, information sessions for migrants and Justices of the Peace on site.

We also learned that it is not simply the volunteers, nor the public using the CAB website, but that other NGOs and charities use the database of information for support as well. Because this is used as key source of reliable information, it offers an opportunity for government to ensure that CAB is consulted as a stakeholder regarding changes in policy etc but importantly identify ways that government content (eligibility/entitlement rules, information) could be shared and accessed via CABnet. With government looking to more digital self-service delivery, supporting customers who don’t feel confident accessing these through personal interactions with organisations like CAB will be important so they don’t miss out on their entitlements. The challenge to government is how we support organisations to give reliable advice to these customers.

Learning these things about CAB plus visiting a CAB location and talking to the volunteers there, helped us to create a more comprehensive current state journey map. This gave us a greater appreciation work of the CAB and the role they play and increased our understanding of the importance of CAB and how crucial CABnet is to the organisation and to New Zealand more widely. This understanding will also be helpful for future work that LabPlus may undertake.

Ideas for improving CABnet

One of the key takeaways for CAB was the analysis of each of their service delivery channels. By analysing the pros, cons and unique features of each of the channels CAB provides their services through, we were able to identify key areas where the website could adopt some of features of other channels to improve the experience of users of the website.

This provided the CAB with the ideas for improving CABnet so that it is more user-friendly for their diverse client base. This was made easier and more detailed because we completed the extensive journey map first to understand what someone does before engaging with the CAB and while using the CAB service, as well as the role of the CAB volunteers. By understanding the experiences of the users of CABnet and other CAB services, we were able to work out what was unique and good about each of the channels of service delivery, and what they were lacking that others provided. This also gave us some idea of why certain channels were preferred over others.

Another key finding for CAB was the importance for other partners they work with (including the future tech partner for the CABnet refresh) of knowing, in detail, what CAB does. Without knowledge of the functions of the CAB and how and why users engage with them, it would be difficult to complete any work for/with them in a way that is helpful and meaningful. The journey map that we created as a part of this sprint may help to provide future working partners with a full understanding of the business of CAB so they can work effectively together.

We also created a prototype that shows what a future CAB website might look like, based on future features that were identified from both the UX review and user research. A link to this prototype can be found here. The review plus user research helped validate that CAB’s original thinking about their customers and the future website was on the right track.

Reviewing the functions and unique value of the different channels was core to designing wireframes for new potential website functions. Web analytics showed us the types of information users of the website often viewed. Unique values from other channels could be adopted by the website, for example having a person explain a complex answer verbally and with visual aides, such as happens in person, could be brought to the website by creating video content to partner with written content to help those with different learning styles or low literacy access and understand the information more easily.

The search function was identified as a key component for the website, and the value of a powerful search engine that allowed for spelling errors as well as links to pages with related content was also discussed. Feedback from NGOs suggests that they use the CAB website as an authoritative information source.

To help direct those who would benefit from the Language Connect service, we discussed creating a prompt to change language with a drop down menu that links to the page with content about the service in the selected language, making them more visible.

Another opportunity identified was to show case studies that give context to how the information provided online may apply to a real life situation. Many people visit or call the CAB to confirm that they have correctly understood how the information they have accessed applies to their own situation. This may help give confidence to visitors of the website. This is an idea and assumption that would require further testing to validate.

Plan to include customers’ voice

CAB team members noted in the retrospective that the week’s experience had shown that there is no substitute for the customer’s voice. They are planning to have more engagement with customers feedback to have input into the CABnet refresh and other future work. One of the tasks we worked on was to create a plan as to how to get more client input from the customers of CABnet while still maintaining the confidentiality that is integral to the CAB service. The main solution that we identified was to update the survey that CAB uses to garner customer feedback, and to place the survey (or a link to it) in strategic places, to try and encourage more responses.

Another solution that we explored was the possibility of using social media to reach out to customers for their feedback. Some CAB’s operate their own Facebook pages and the team thought that having a live chat on Facebook would be a way to engage customer feedback.

Web analytics insights

We conducted an analysis of the data relating to the CAB website to try and get a better understanding of the users of the website. We looked at data from the period of July 1 2016 - June 30 2017. Over the year that we analysed there were 2.12 million sessions from 1.62 million users with 3.8 million pageviews. The average time spent on the site was 1.41 minutes - the brevity of each visit suggests that most customers find the information they need quickly.

We focused on the user journeys and needs of website visitors people in NZ, not international audiences as both the CAB and Lab focuses were primarily on New Zealander needs. There were roughly 1.22m users from New Zealanders with 400k from overseas which could be the basis of further analysis. Other countries of relevance include the United Kingdom (7% of sessions and 8% of users), United States (almost 4% of sessions and 4.5% of users), and Australia (2.5% sessions and 2.8% users). Thereafter there are less than 1% of sessions and users from (in order), India, Canada, South Africa, Ireland, Philippines and Malaysia.

There were 1.72m sessions from the people in New Zealand

• 7% were looking for a "CAB near you"
• 22% users started at the front page and then went all over the place (see popular pages below)
• A large proportion of people (around 2/3) go straight to a page (not the front page) again reflecting the good google juice of the site,
• 85% of people that go directly to a page then appear to get what they need (don't click through) which is fascinating and may indicate regular users with bookmarked pages.

Some basic behavioural insights below. These are useful in planning into the future:

• Peak times: The CAB website has steady use each week, peaking on Mondays (around 7,200 sessions) and tailing off until the weekend (around 4,200 sessions). There appears to be a lot of people returning every hour or so to see things under the "free advice" page - this was likely the volunteers visiting the website to find the information to pass on to clients.
• Devices: 42% of all sessions are done on mobile phones - 51% mobile users are on iPhones, and 47% are Android based phones (e.g. galaxy)
• Browsers: The other 58% of sessions are on browser based devices (tablets, laptops, desktops), which are predominantly Windows (~85% non phone devices) followed by Macs (~12%) and then Linux (1%). 45% of ALL sessions are done on chrome (970k), followed by Safari (500k) then IE (455k)
• City: Of the sessions from New Zealand, 56% sessions and 58% users are Auckland based, followed by Christchurch (10% sessions and users), then Wellington (9% sessions and 8% users), dropping then to Tauranga (4% sessions and 3% users) and Hamilton (4% sessions and users) before tailing off to sub 2% sessions and users from (in order) Dunedin, Palmerston North, Napier, Whangerei and beyond. Only 1.2% sessions and 1% users had no location set.
• Device languages: 97% of people browsing the web are doing so from english devices (United States, United Kingdom, New Zealand or Australia combined)
• Distribution: There appears to be a very long tail. Only 12% of all sessions were the front page of the website, implying people are linking directly or finding pages directing through search results a lot of the time. This is reflected in 1.3m sessions coming from organic search, 350k are direct links and 60k sessions are referrals. This means any change in website infrastructure will require a good redirection strategy to retain good google juice.
• Popular content: The most popular overall page (apart from the front page) was "find a CAB near you" at 2.44% pageviews (93k pageviews) showing a large number of pages accessed across the site.
• Popular pages (useful to infer themes): The popular page content should give some indication of the content that is most helpful to people. Most popular individual pages apart from find a CAB near you still had less than 1.1% page views. Top 10 most popular pages in order were: legal ages and ID, other ways to sell your house, blocks to travelling (fines and criminal record), relationships property (for division of property after separation), fences and boundaries, being a guarantor, the rights and obligations of landlords, hours of work and wages, trial periods and credit checks and records.

It is worth further investigation to better determine volunteer behaviour because it was a bit tricky to definitively split volunteer behaviour from clients. Analytics assigns each device a unique Client ID, and considers each unique Client ID as a unique user in your reports. Some of the Client IDs had more than 1000 (up to 5000!) sessions in the year, indicating volunteers or others working with the content regularly. There is a pattern of regular sessions from 1000 down to 10 for the top 5000 or so Client IDs.

Agile experience

For CAB, the sprint provided an experience of working using Agile SCRUM methodology. This was seen as useful both for learning how this is done but also provided some knowledge about how a tech partner with CAB (for the CABnet upgrade) will most likely work and what CAB needs to do to work with them. Members of the team who came from CAB also learned some agile techniques that can be taken back to CAB and used in future work as well.

Working well as a team

As a joint effort between LabPlus and CAB, this was a collaboration between government and a non-governmental partner. There was some apprehension about how this would work, especially cramming such a lot into a very short timeline. However, everything ran smoothly and the willingness of the team members ensured that we completed all of the tasks within the allocated time, and provided mutual benefit to both organisations. The Service Innovation Lab was a good environment to do this kind of work, as it provided a more neutral space and relaxed atmosphere that allowed us to feel comfortable and focus solely on the tasks of the sprint.

Challenges

The main challenge of the week’s sprint was how to access the views of the CABnet users that are members of the public (as opposed to CAB volunteers or other NGO’s). We decided that it was inappropriate to talk to CAB users directly, as we did not want to taint CAB’s independence from government by approaching their clients, especially as often times these clients have made a deliberate choice to approach CAB rather than government agencies. This meant that we were relying on the CAB volunteers experience of client views to represent the views of the public. As a result we finished the sprint unsure that we had answered the initial questions we began the week with:

• Are the customers who use the CAB website the same as the customers who visit a physical CAB location?
• Do they find what they are looking for on the website?

While we didn’t fully answer these questions, we are certainly clearer on how we might continue to find the information to answer them and have a plan in place to capture more information from users. The plan mentioned above to capture more of the customer voice will help to answer these questions about whether customers find what they are actually looking for on the website. There is also an opportunity to add a question to the phone script for the CAB phone line asking callers about their website experience to begin to capture some of this information.

What’s next?

Next steps for CAB are to engage with a tech partner and work with them to upgrade CABnet, using our findings from this sprint to inform their work.

From a LabPlus perspective this work will help us to consider CAB and other other non-governmental service providers as part of the user journey in other work that we participate in, including considering how we can support and enable CAB and other similar entities to help New Zealanders. The policy and service implications of this needs to be further examined.

LabPlus will also consider how we can make the best use of the extensive CAB database of information - who else might benefit from this database of information and how can we ensure that government is providing timely information accurately and clearly to allow CAB to keep their database up to date.

Acknowledgements

We would like to thank the following organisations for their help in contributing to the sprint:

Citizens Advice Bureau New Zealand

The CAB volunteers who shared their experiences and opinions with us

Language Connect

Service Innovation Lab (LabPlus)

Interpreting New Zealand