https://pds.blog.parliament.uk/2016/05/18/data-and-search-whats-happening/

Data and Search – what’s happening?

I have been the Head of Data & Search for 6 weeks, working in Emma Allen’s Digital Development team. It is a new role, and I’m fortunate because it’s something that I wanted to do for a while.

In these first weeks I’ve been getting to grips with the task at hand. The potential range of responsibilities is daunting, and I know from prior experience that I have a tendency to try to do too many things at once.

So, for now I’m focusing on the first 12 months. I have 3 broad priorities for the year:

  1. Improve corporate data

‘Corporate data’ means it’s internal-facing, often about staff, relating to systems used by HR and Finance among others. This is an area that is causing internal users pain. There are many desired outcomes that aren’t being delivered well – for example when somebody new joins the organisation; or when you want an organisation chart; or when somebody moves team and their telephone number is inconsistent across multiple platforms.

We are in need of a ‘single version of the truth’. This is a challenge that many organisations face, certainly nothing unique to Parliament.

I will be working with colleagues from the Digital Service and from the administrations in the Commons and Lords to improve the situation.

  1. Provide data to the new website

The Parliamentary website Alpha team has formed and is picking up pace. It’s an exciting project, and it’s great to see open, energetic and collaborative work happening here.

We know that we want our website to be data-driven. We have a data architect (Michael Smethurst) embedded in the Alpha team and that’s going really well. I want a similar arrangement for working with our bespoke business systems development too.

We know that we’re going to need early sight of the website’s data needs, and (by extension) user needs for our data too.

Depending on the outcomes of research and priorities, we might have to do some complex work in the next few months in order to deliver for the website. That would be a great opportunity.

It would involve multiple teams: those working on our internal business systems under Matt Stutely that create the data in the first place; the data team; the website team; and multiple colleagues from the administrations in both Houses.

This would be an instructive test of how we’ve changed our delivery culture for the better in the Digital Service in the past year. We would see improved public-facing outcomes as a result of work that has been disconnected from the web and largely invisible in the past. So, fingers crossed this will come to pass.

The website Alpha is also a good opportunity to test the data services that we already have, and learn more about how we need to improve, which leads to…

  1. Building capability

This has been my main area of focus. I split this priority into a further 3 areas:

a) Building the team

I did a fair bit of reflecting on what I’d learned from my last role in recent months, and I realised that you’re only going to get so many opportunities to build a team in your career. This is the part of my work that I like the most.

I’ve already got some seriously great people working with me. To name a few: Robert Brook, Samu Lang, Aidan Morgan, Matias Noe, and Michael.

I have a list of roles that we need to recruit as well. Top of the list are some junior developers, a delivery manager, and a data analyst.

In addition to building new products and services, I have responsibility for supporting our existing data services and related applications. This is part of a wider transition in the Digital Service in how we do support, moving to product-based teams. Some operations engineers will be joining the team in the coming months. I have started work on our process, reviewing the list of current work outstanding, and (crucially) asking our internal users what ‘good’ support looks like to them.

Also under building the team, there’s building and re-building relationships across Parliament and outside as well – harnessing the enthusiasm for Parliament’s data to help improve our services. I’ve made a start. We’ve got a long list of people to talk to!

b) Building capability outside of the team

This is just early thoughts. Being a ‘supplier’ of (say) analytics and visualisation services to Parliament is not one of my goals. I do know that there are areas of considerable data expertise in Parliament outside of the Digital Service. What (if anything) can I do to help facilitate their work? What can I do to help ensure there’s visibility across Parliament of this expertise, and a data community of practice? And how can I help to build data skills in non-specialist areas?

c) Building the next iteration of our data service

I’ve probably spent the most time and energy here. This is about where we go next from data.parliament.uk. It is also about how we serve data to the website when it reaches Beta, if not before.

So far, we think that our service needs to focus on 3 main elements – 1) how we better model and understand the business of Parliament, so that we can better link data from a variety of systems together; 2) a cloud-hosted platform that performs well and is really easy to use; and 3) search, or better, ‘find’ – how can we build the ability to better find our data in from the beginning?

I am also keen to address the provision of bulk data, which isn’t well served by data.parliament at present.

As of 11 May we are in Discovery for 8 weeks until mid-July. Prototyping, talking and gathering and understanding needs has started. We will then do an Alpha build, which we’ll deliver by the end of October. What does the minimum viable data service look like? Something that reflects the 3 elements above, and is also demonstrably better than what we already have.

That’s the progress. We’ll be writing (shorter) blog posts and working in the open as much as possible between now and… forever. If you are interested in any of the above then please get in touch: @dasbarrett or barrettd@parliament.uk.

 

1 comment

  1. Charlie Hull

    I'd be interested to know how you plan to test your search results for relevance - this is something we've been looking at a lot recently, trying to develop a proper methodology for test-based relevance tuning. This series of blog posts might be useful! http://www.flax.co.uk/blog/2016/03/18/get-started-improving-site-search-relevancy/

    Link to this comment Reply

Leave a comment