GOV.UK Verify recently passed one year in public beta. Programme Director Janet Hughes talks about what the last year has involved, and why the team is focusing on the months ahead.
Join the conversation on Twitter, and don't forget to sign up for email alerts.
Transcript
GOV.UK Verify
One year of public beta
It was kind of low key hitting a year because our big focus is about getting from here to live, so we’re not really measuring time on the clock, we’re measuring performance and improvement and iteration and getting to the point when we’re ready to say services can rely on this and they can start turning off alternatives. And that’s what it means to go from beta to live, which we’ll do next April.
Getting ready to go live
We’ve gone from no users to 300,000 users, we’ve gone from one service to 13 services using Verify, we’ve grown the team now so that we’ve got a team capable of taking the service from beta to live. One of our big priorities is improving the completion and success rates for Verify so that more and more people who want to use it can use it. A year in felt like most of the other days, we’re just looking at how can we go on and improve the service today.
8 comments
Comment by Mike posted on
Janet - thanks for such a comprehensive answer. I must admit I was expecting a stock GDS corporate message about how Agile can deliver the world tomorrow, so I was pleasantly surprised to get such a reasoned, balanced view! Thank you, you have restored my faith in GDS! 🙂
Comment by Janet Hughes posted on
Thank you! That's very kind of you to say and much appreciated 🙂
Comment by Mike posted on
Thanks, very interesting. What's your insight into the length of the Public Beta (and the rest of the project)? The Digital by Default approach heralded agile working and getting the MVP live as soon as possible.. do you think there's much of a difference between the timescales of agile projects vs traditional project approaches?
I can understand the argument that the right thing is built because of the constant user iteration, but I wonder if there's a recognition that speed to market might be much the same?
Comment by Janet Hughes posted on
Hi Mike, thanks for commenting.
It might be possible to get to an official 'launch' point more quickly than the speed with which we will go from beta to live, but in cases where that's the over-riding priority it sadly often turns out that what's been built doesn't actually meet user needs so it wasn't really worth the rush. 'Traditional' approaches often compound this by specifying solutions (rather than products) and making it very difficult and expensive to change things once the official 'launch' point has been reached. So you might be able to tick a box to say you delivered on time, but you end up stuck with a poor service. That creates frustrating user experiences, failure waste and more cost.
We're working to get from beta to live as quickly as possible, but that doesn't mean the service will be 'finished' or 'launched' at that point. It's already in use - 13 services have connected to GOV.UK Verify, and it has verified 300,000 identities so far - and we're constantly improving it to meet user needs so we can meet our stated objectives by April next year (see https://identityassurance.blog.gov.uk/2015/10/21/gov-uk-verify-the-latest-on-progress-towards-objectives-for-live/).
I think it's probably also important to say that we're not dogmatically pursuing a rigid set of processes or a particular branded approach - we use whatever combination of tools and processes can best help us meet user needs and we constantly review and improve the way we work based on what we've learned so far.
It's also fair to say that building and scaling GOV.UK Verify is complicated and necessarily takes time. We're not just building a new type of service that hasn't been built by anyone before - we're also trying to build a new market based on published standards and capable of meeting needs beyond central government. That may be more complex in delivery terms than a single supplier delivering a specified solution, but we think it's worth the effort given the value and benefits to users of this approach.
I don't think a traditional waterfall approach could have got to this point this quickly, and I'm certain it wouldn't have resulted in the market innovation and performance improvements we've seen so far during our public beta and the further developments improvements we're expecting over the next few months.
Having said all this, we do of course have a lot of work still to do. We're working hard to iterate and improve the service for users and get GOV.UK Verify ready to move from beta to live, and we welcome ideas and feedback about how we could do better. We publish updates over on our programme blog if you're interested to follow our progress: identityassurance.blog.gov.uk
Comment by Matthew posted on
Do you have any metrics for success rates?
Comment by Louise Duffy posted on
Hi Matthew, we publish success and completion rates on the GOV.UK Verify performance platform: https://www.gov.uk/performance/govuk-verify
You can also read about the work that's going on to improve those rates over on the Identity Assurance blog: https://identityassurance.blog.gov.uk/2015/10/21/gov-uk-verify-the-latest-on-progress-towards-objectives-for-live/
Comment by Tom Ruppel posted on
Typo in 1st sentence, 2nd para: "form" instead of "from".
Comment by Louise Duffy posted on
Thanks Tom, sorted!