Tax Free Childcare (Providers) - Alpha Assessment
The report from the alpha assessment for HMRC's Tax Free Childcare (Providers) service on 10 August 2016.
| Assessment stage | Alpha |
| Assessment result | Met |
| Service provider | HM Revenue and Customs |
The service met the Standard because:
- The service team clearly demonstrated and understood the end-to-end user journey for both child care providers and parents and the interaction between the two journeys.
- Assisted digital support is well defined and the design advanced for the current phase of development.
- The team are geographically separate but have worked well to overcome problems caused by lack of colocation.
About the service
Service Manager: Catherine Pike
Digital Leader: Mark Dearnley
The service allows childcare providers to register their bank details to receive payments via the new Tax Free Childcare service.
Detail of the assessment
Lead Assessor: Thomas Moore
The childcare provider component of the service was presented at assessment, with a further parent component (currently in alpha) making up the overall Tax Free Childcare service.
The service demonstrated was well beyond what the panel would expect to see at an alpha assessment and as such the service surpassed the alpha standard in several areas. This service is a new digital service and does not replace any existing service.
User needs
The team has conducted a lot of research with users and as a result produced personas that can be used by the wider team when thinking about user needs and design issues.
The team has discovered that 7% of their user base are digitally excluded and 12% have assisted digital needs. Consequently the team has also undertaken a lot of work in discovery and alpha to understand the needs of users with low digital skills and those with assisted digital needs, to ensure that the service is available to those who have difficulties doing things online.
Although team members attend research sessions, it wasn’t clear how involved the wider team were in conducting the analysis and making sure that decisions about the outcome of the research were conducted as a team. Going forward the research team should ensure that the research is conducted as a ‘team sport’ so that the whole team can take ownership of product decisions that have been made from observing users.
The team plan to contact all childcare providers and ask they register during the private beta. This is not consistent with a typical private beta, where a small subset of users would be invited to access the service. The panel understands the desire to have all providers available to parents invited to the parent private beta. However, the team should consider inviting a subset of providers for private beta and, when ready, asking parents that use those providers to participate in the parent private beta.
Team
With the exception of a data analyst, the team currently has coverage of all the key team roles described in the Government Service Design Manual. The team works on both the child care provider and parent aspects of the service, and the service manager has oversight of the end-to-end process.
The lack of colocation of the service team was of some concern to the panel. The service team work at multiple sites across the UK, and although this isn’t ideal, the team have implemented measures to ensure daily communication is maintained, and decisions are made quickly. The impact of distance between team members has been lessened through the effective use of collaborative tools, and by arranging regular face-to-face sessions throughout the sprint. The team meet four days out of ten during each sprint for sprint ceremonies such as sprint planning, show & tells and retrospectives, and team members dial-in to daily standup meetings. However, as the service moves along the development cycle the team should look to being co-located where possible so that research and design can collaborate fully.
Technology
The team demonstrated content for planned and unexpected downtime and that the expectation was that any downtime would be resolved within 2 days. As downtime of this aspect of the service would only affect registrations, it should not stop users being able to receive money.
The rest of the technology section was reviewed and covered in the Tax Free Childcare (Parents) assessment.
Design
The name of the service is currently a noun, “Tax Free Childcare” and doesn’t describe the actual task the service performs, or who the service is aimed at. As there are two components to this service, one for providers and one for customers, it is recommended the service name properly distinguishes who and what the service is for.
The team demonstrated they had iterated the design and language of the service to meet user needs. The panel would have liked to see evidence of trying different service design approaches, however the panel recognises that this is a fairly simple transaction. The scope of the service has been based on service functionality. Future features have been categorised into delegation, security and transaction history, and are prioritised based on user demand.
The service has not been extensively tested with users with accessibility needs or had an accessibility audit, however the team stated they had undertaken some guerrilla testing with visually impaired users as well as testing with HMRC staff with access needs. The code is also run through automated accessibility tests before each deployment. The team have also been in contact with the North East Accessibility Centre. An accessibility audit will need to be undertaken, and any issues rectified, before the beta assessment.
The panel would have expected redundant and potentially confusing elements of the service to be identified and iterated as a result of user testing. For example, the “Sign up or sign in to your account” page appears redundant in its current form, whilst the “What to do if these aren’t your details” is unclear. Elements such as these should be examined in early beta development (a review of service content will be sent along with this report).
The team need to develop messaging at the end-point of the transaction. It’s not clear what child care providers can expect once they’ve registered or updated their bank details. For instance, do they need to pass on the bank details to parents? Do they need to provide any further information to ensure they’re paid correctly? The confirmation pages section of the design patterns offers guidance on this.
The service is completely new, and digital take-up will be addressed by sending registered childcare providers letters with a URL and unique user ID. The team are expecting a 99% take-up by live.
Analytics
In addition to measuring and reporting the four mandatory Key Performance Indicators (KPIs), the team also intend to measure the time taken to complete the transaction and referral rates from the Tax Free Childcare campaign site so the effectiveness of the awareness campaign can be tracked.
The team have had initial conversations on setting up a performance dashboard with the Performance Platform team.
Recommendations
To pass the next assessment, the service team must:
- Ensure user research is conducted as a ‘team sport’, with team members attending research and the whole team involved in analysing findings.
- Design a private beta that tests with a subset of childcare providers.
- Test with users to understand issues that may arise relating to authentication and ensure the login process is secure and easy for users.
- Test whether obfuscating most of the bank account number meets user needs.
- Determine a name for the service that describes the task that the service allows users to complete. The “Good services are verbs…” blog gives advice on this.
- Remove superfluous screens from the service to further streamline the user journey.
- Address potentially confusing content issues.
- Undertake a full accessibility audit of the service.
- Test with users with a different range of access needs.
Next Steps
You should follow the recommendations made in this report before arranging your next assessment.
The service should be offered to a subset of child care providers as part of a limited private beta. This will allow the team to test how the service works with real users completing real transactions, and iterate the service accordingly, before rolling the service out to the wider user base. The service can then be offered to all child care providers once the service moves into public beta following a successful beta assessment.