https://insidegovuk.blog.gov.uk/2015/05/11/what-were-working-on-11-may-2015/

What we’re working on: 11 May 2015

Regular readers will be familiar with our fortnightly “roadmap update” posts. In response to feedback we’ve dropped the jargon and will call these posts “what we’re working on” from now.

Having been dutifully quiet for the election period, here’s a bumper list of everything we’ve done since the last update at the end of March and what we’re working on next.

What we’ve done since 25 March

Running and supporting GOV.UK

To maintain a reliable service and in response to requests from users and government colleagues, we’ve:

  • made lots of changes to centrally-managed content including:
    • annual uprating of fees and charges across 150+ URLs
    • changes relating to the new tax year including rules for pensions, capital gains tax, marriage allowance, community sports clubs and national insurance
    • changes to several smart answers including overseas passports, births and deaths, adoption, UK visa checker and marriage abroad
    • clarified guidance about changes to child car seat rules
  • launched a new countryside stewardship grants finder
  • supported a few organisations’ name changes (the orgs formerly known as TSOL, Highways Agency and NMO) and updated our guidance for how to rename, merge or close organisations
  • improved the security of the blog platform by making it more difficult for unauthorised users to publish content on it
  • changed the Cymraeg link in the site footer to make it clear that it doesn’t toggle the currently-viewed page into Welsh (it now reads Rhestr o Wasanaethau Cymraeg, list of Welsh services)
  • dealt with a bug affecting publisher user accounts (upgrading part of the application in the process)
  • trained hundreds more people around government in user needs, writing for the web and our publishing tools
  • fixed a bug which meant documents scheduled for publication in mainstream publisher didn’t recognise British Summertime
  • improved search in specialist publisher so it’s no longer case sensitive and searches both slugs and titles
  • improved the publishing interface on specialist publisher to making options to publish and withdraw documents clearer
  • improved the print stylesheet for HTML publications

We’ve also continued tying up loose ends from the transition programme. We’ve:

  • created a new organisation type for judicial bodies, to allow the high courts to have a presence on GOV.UK
  • updated messaging on the air accident reports finder to reflect that they have now transitioned
  • removed transition progress bars from the list of organisations
  • republished a bunch of MAIB documents so that the organisation name shows correctly in search results
  • scraped and imported legacy competition commission cases into the CMA case finder and improved the filters so users can make multiple selections

Improvements to the manuals format

The manuals format is one of the newest additions to the site (read more about the format here), and recently we spent a bit of time improving it. We’ve:

  • added the ability to search within a manual - here’s an example
  • exposed manuals to external search engines
  • improved the publishing interface of the specialist publisher (used for manuals and other specialist formats) to reduce accidental publication

Improvements to feedback loops

We’ve also made improvements to how we collect and act on user feedback. We’ve:

  • launched an A/B testing framework which means we can run experiments more easily and more often
  • started testing a new version of the anonymous feedback link at the bottom of a selection of pages of GOV.UK, to see if we can gather more actionable data
  • created a weekly CSV report on the most problematic content on GOV.UK to help identify areas which need improvement
  • researched what publishers need from the user feedback explorer tool and have started work on new features which include filter by organisation and export to CSV

Readying GOV.UK for the election

To make sure GOV.UK meets users’ needs through the transition from one government to the next, we’ve:

  • built and deployed ‘history mode’, which means policy-related content is clearly labelled and down-weighted whenever there’s a change of government
  • automatically applied history mode to all policy-related content from the 2010-15 coalition, with the ability to manually override it
  • added history mode to the blog.gov.uk platform, so bloggers can apply it selectively to their posts
  • replaced the policy format with simpler aggregation pages, removing the policy narrative (which users distrusted) and allowing content to be grouped both by a policy and its underlying sub-policies.
  • improved how GOV.UK handles changes to the machinery of government, and updated guidance for government publishers on all aspects of managing content through an election

Making it easier for users to find things

The problem of how users find what they need remains a very high priority, and we have a team dedicated to solving it. Recently we’ve:

  • completed a thorough assessment of everything we know about users' problems finding things on GOV.UK, and developed improvement plans to solve them
  • upgraded the search engine (ElasticSearch) to version 1.4 to allow us to improve search result quality
  • begun the process of identifying gaps in the categorisation system of the site (topics and subtopics), to make sure that all content can be categorised appropriately
  • begun linking mainstream browse through to topics and subtopics, so users can find all the GOV.UK content by subject (where content has been tagged to topics and subtopics)
  • migrated legacy detailed guidance categories to subtopics, so we can have one fewer browse mechanisms on the site
  • moved tag management from a legacy app (panopticon) into collections publisher and put it in content store as part of moving to the new publication pipeline
  • improved the spelling correction suggestions that are given for misspelled searches

Improving our software and processes

We’ve continued building out a new single publishing pipeline so we can simplify our backend software architecture, and we now have one format (case studies) using the new pipeline end to end. This is a big piece of work and will be ongoing for some time.

In our content team, we’ve made use of the quieter pre-election period to try out new ideas and improve how we work. Among other things, we’ve spent time thinking about tough content design challenges (like long lists, forms, letters and the increasing use of mobile devices), and worked up plans for how we can proactively review content and make improvements theme by theme.

Things we plan to do next

We’ll blog soon about our long term goals and the missions we’re working to achieve this financial year.

Meanwhile, in the coming weeks we expect to:

  • add more filters to the new policy pages and tidy up technical debt from election-related changes
  • complete the switchover from classic Google Analytics to Universal Google Analytics on 2 June
  • continue to replace detailed guidance categories with topics and subtopics, and link mainstream browse through to topics
  • make some ‘quick win’ improvements to site search using features of the newer version of ElasticSearch
  • define user journeys to work on as part of content improvement theme work for Immigration and Passports
  • run a planning workshop on the Scottish rate of income tax
  • complete final outstanding HMRC post-transition content improvements
  • work on 3rd country updates to the Marriage Abroad smart answer
  • finalise and publish updates to the pay and leave for parents calculator as well as a new version for adoptions
  • finish scoping user needs and content for the Planning Portal transition
  • add EU logos to two funding finders

As always, if any of this is unclear, or if you have feedback on whether we’re prioritising the right stuff, please do comment on this post to let us know.