Digital Self Assessment (Opt-In Messaging) - Live Assessment
Digital Self Assessment (SA) is one of HMRC’s flagship online services: over 85 percent of SA customers chose to file online during 2014-15. Despite this, HMRC still send large volumes of paper to SA customers. HMRC want to improve this by giving them the option to receive electronic rather than paper communications. Digital SA allows customers to say "yes" to paperless and choose to receive digital alerts instead of paper.
Department / Agency:
Date of Assessment:
Result of Assessment:
After completing the assessment, the panel can confirm the Digital Self Assessment Opt-in and messaging service has shown sufficient evidence of meeting the Digital by Default Service Standard and should go Live as a Digital by Default service on GOV.UK. The service can now remove any Beta branding.
The service was assessed against and has met all 26 points of the Digital by Default Service Standard.
User needs and research
The stated user need for the service was "as a Self Assessment online user I want to sign-up for digital alerts so that I can receive Self Assessment communication electronically".
Sprint based research into the usability of the service has led to frequent improvements during the beta period. Once improvements are identified they are fed into the backlog and prioritised. During the private beta the opt-in was offered to 100 users, with 75 choosing to take up this service, however some of those thought they were signing up to a digital self-assessment service, rather than to receive their messages digitally. This was quickly identified and the language and the layout of pages have been iterated regularly to make it clear what the service is offering. When the service moved to public beta there was initially a 53% take up. This improved further when the user journey changed so that new users signing into the My Tax Account service were offered the opt-in as part of the process at the very beginning. Currently there are over 1.2 million users signed up, with a very low contact rate of 0.3% (most of which relate to the process of tax returns rather than the service itself).
The use of A/B testing allowed the testing of graphical icons like checks/ticks to the opt in process, showing an increased uptake of the service by drawing users in to the content, increasing the comprehension of the service. So far only a tiny fraction of people have chosen to opt-out after signing up to the service. Most of the research has been carried out in testing labs with a task based approach. It was good to see that user research is available to the team to observe, the live user research demonstration was a first in a Digital by Default service assessment.
The team have most of the expected roles in place, either directly part of the co-located team or as part of a shared resource through HMRC’s common technology resources. There has yet to be a Digital Analyst recruited and much of this work has been carried out by the Deputy Service Manager. The Service Manager is empowered and responsible for the service, an example where the business wanted to push marketing material to users through the service was resisted to protect the interests of the users.
There is an identified reliance on contractors, but a plan is in place to reduce those numbers and share knowledge. As an exemplar service there has been support from the GDS Transformation team during early beta but that resource has decreased as the service team have taken over. Roles are clearly defined with clear demarcation which is especially important with the reliance on the shared services offered by HMRC.
The service team use showcases and ceremonies, and have stand ups led by scrum master. Non-functioning components are prioritised in the sprint and the team use a variety of agile tools to help manage the work from a backlog. Originally the team ran with 2 week sprints but are now running weekly. An example of the agile process working was the changes to the verification email step resulting in a reduction in contacts asking what the verification link was, the time scale from identification to production was about a month including user research and testing. While introducing agile into the organisation was initially tricky there are now expectations for very quick delivery of products and services built this way. The team average 2 releases a week, but demonstrated the ability to make changes as frequently as they needed. Some dependencies are across the common platform and take some more time to deploy, however most changes now are around the content and can be quickly released, an example of a bug that was identified and removed within 2 hours down to the regular deployment process having no downtime.
The assessment panel were encouraged to see the service is sharing the knowledge and learnings from this service to other teams in HMRC, particularly around the content of paperwork being able to be reduced and made clearer.
Development and security
The service team have engaged with all the right people across HMRC from the outset and continued to engage at each important change, for example a CLAS consultant was in the same building and the team frequently involved them. There is a good understanding of the risks posed by the service and the design has taken appropriate measures to defend against them, for example of the risk of phishing attacks.
The GOV.UK Verify service is not currently being used because it does not yet provide all the features needed, the service will be looking to switch over as soon as appropriate. A large proportion of the code is open source and being developed in the open. Some libraries are not yet open source but the team has a roadmap for how these will also become open source in due course.
Design and content
The service has Google Analytics and the team have set up some funnel visualisations to identify dropout points, although there was not a clear explanation of the funnel. User research informs the look of the service (based on GOV.UK design patterns), an example the team gave around design included removing a blue banner that had messaging directed at the user, this had a direct impact in increasing users engagement (most users were ignoring it). Content changes to the opt-in service can be made quickly.
The service is yet to fully meet all the AA standards, an outside independent body is to test th the service in April, changes are needed are HTML tweaks to help with screen readability and ordering. Any changes should be prioritised.
Assisted Digital and digital take up
Personas were created dependent on different circumstances, those who struggle tend to go through charities rather than directly to the HMRC. The team demonstrated that assisted digital support does not need to be provided at this time. If the scope of the service changes (eg new transactional services are added or the user base changes), the team may need to undertake research with assisted digital users and design, test and provide appropriate assisted digital support which meets user needs. As this service is part of Your Tax Account service that service has responsibility for providing Assisted Digital (the use of Digital Self Assessment is optional and user will need an email address before signing up).
At the time of the assessment there were no plans to mandate paperless services (the assessment took place before the 2015 Budget), however the take up for this service has been higher than initially expected. Contact centre staff have been asked to encourage users to opt in, followed up with IVR messaging.
Analytics and performance
Google Analytics Premium has been instrumented on the service and goals set up to provide completion rates for the Performance Platform. SPLUNK has also been used to provide the data for the A/B testing with support from the Behavioural Trials Programme at HMRC.
Testing with the Minister
The service was tested with the responsible minister.
The service has met all points of the Service Standard but the assessment panel have highlighted the following recommendations to support the service’s continued development.
While the service has met the user need for many of its users there is some scope for further thought about the options for notifications, it may be that important call to actions in messages could be backed up with additional notifications via other digital channels (such as SMS messaging), more research into this would help understand this better.
It is recommended that when recruited the digital analyst should be co-located with the service management team and be responsible for managing the improvement of the Google Analytics configuration (i.e. views, filters, goals and funnels), further optimisation of the implementation and the provision of actionable data insights for consideration for further design iteration or AB testing. This responsibility should not sit with the Service Manager or the Deputy Service Manager.
The team should share the research into the use of visual prompts (ticks/checks) that led to increased take up, this is potentially useful for other services. The reduced word count in the messages to users leads to a clearer message, it’s now important to ensure users understand the content and can easily act on any calls to action, more research should be conducted on the style of the language used.
Digital take up
The service team will need to continue to encourage users to opt-in to digital communications, the option to print out communications will help some users to transition across. More research should be carried out to find out why some users do not sign up and others choose to revert to paper based communications. It will be useful to look at benchmarking the digital take up against other financial services.
This messaging service, one of many parts of My Tax Account, has been given approval to go Live as a Digital by Default service. We look forward to seeing what the team will deliver in the future for Self Assessment.
The success of the service so far in public beta and the reassurance of this pass result should encourage all users to make the choice to switch to digital messaging with HMRC and the Self Assessment service. 2.2 million pieces of post have been saved so far saving £600k.
The assessment panel thanks to the whole team for their time and clear answers during the assessment, this service demonstrates the kind of change of approach for all communications needed across government, it exemplifies the Digital by Default approach.
Digital by Default Service Standard criteria