|Organisation||London Borough of Hackney|
|Collaboration Level||Open To Conversation|
|Budget||£50K > £100k|
|Key Contact||Sarah Palmer-Edwards|
|Phase start||08 January 2019|
|Phase Estimated end||30 April 2019|
You can read our full Local Digital Fund bid here: https://localdigital.gov.uk/funding/london-borough-of-hackney-2/
We want to work together with our partners (Suffolk Council, Stevenage Borough Council, Cornwall Council ) to conduct some further discovery and an alpha project looking at the user journey for council staff who manage and respond to FOI requests and SARs in order to prototype approaches to the unmet user needs we have discovered, and understand whether a digital solution and data standards could better support these needs and allow them to handle requests from the public more efficiently.
We’d like to develop a prototype of an end-to-end open-source solution that includes reporting, and subsequent publishing of non-personal requests. The main objectives would be to assess the potential for increasing efficiency and ease of use for the council, and better meeting the needs of citizens, both from ease of use and response time perspectives, and in such a way that supports reuse by other councils.
A successful outcome would be an end-to-end set of user interface prototypes for a request handling system that have been validated by user testing in multiple councils and a clear recommendation based on a business case on whether to proceed into a beta project for either open-source request handling software, or a set of open standards around request handling. In either case, the user research report created should outline.
We’re in the final stage of the project! The last two weeks has been all go, as we write up our finalised business case and prototype report in detail. We’re reflecting on how well each prototype serves the user stories that resulted from the discovery phase of the project, and how each could be taken forward.
We also spent some time reviewing our draft research findings report, and making final tweaks before we publish them as a set of three.
This week also saw the release of our in-house research on FOI statistics and local government FOI admin that helped us form our approach in the discovery phase of the project, and backs up some of our benefits cases.
And finally, we published our reports and prototypes:
And that’s it! Thanks very much to everyone who’s been following along with the project. We hope that sharing our processes and thoughts like this has been helpful to others who might tread a similar path in the future; and of course it’s also our hope that the final reports will be of wider use.
Our inboxes are always open if you have any questions, comments or just want to discuss our findings. Drop us a line via firstname.lastname@example.org and one of the team will be back in touch.
To get some feedback on the revised questions we decided to use a Google Form, as it was quick to implement.
Of course, it didn’t give any indication of results at this stage, but we were able to use the data collected to manually calculate the results and check that the questions still generated the overall scores we were expecting.
After we were satisfied that the questions were an improvement, we moved on to thinking about how best to show these results.
We wanted the results to help isolate the most important areas to improve within the FOI service. The overall scores made it too easy to use the tool as a “scorecard” to rank yourself against other councils – something we wanted to avoid – so we decided that we wouldn’t show these. The main aim of the scores is to establish the relative focus between the 3 key areas (People, Process and Technology), so we just reflect this with the radar chart.
We also decided that we’d show the scores to each question on the results page, with colour highlights so that it's easy to spot the critical issues.
Building the question pages was fairly quick and simple, as there were no major design decisions to make. We opted to store the answers in sessionStorage so that the prototype didn’t need a backend – something that would need more work and more complex hosting arrangements.
The results page was more complex, but it came together well. The scores from the question pages get calculated and rendered in to a radar chart and we implemented the colour highlights to hint at critical areas for improvement.
You can view the prototype at https://mysociety.github.io/foi-self-assessment/.
We’ve also been building a guide page around Time to allocate. We’ve extracted the beginnings of this from our initial presentations and the prototyping workshop in to a draft, and we’ll be asking for feedback and touching that up ready for inclusion as an example in the Self Assessment prototype.
As well as the prototypes, we’ve been thinking about the business case for taking them forward. We’ve built our first draft financial model of FOI services.
How to pull everything together in a form that’s usable and useful.
Wrapping everything up in a set of final reports.
This week we’ve been iterating on the Self Assessment Playbook, in response to the feedback from the workshop. We refined the questions to make them less about the specific tools or processes Local Authorities have in place. We also tweaked the answer format to be less of a ranking and more of a scale per question. We hope this will be better at pinpointing specific issues that the guides can address.
We also spent some time organising all the user stories we developed in the discovery phase, grouping them around each significant “actor” in the FOI handling system.
A big question we have is how to adapt the MOJ Correspondence Tool across local authorities. One of the issues for Hackney is that they’ve seen a significant improvement in performance from integrating the response collaboration with their current tool. Other councils prefer the system to just manage the workflow, rather than the details of the response. We’re thinking about how we can make this difference configurable for a variety of situations.
We’ve made a couple of quick mockups of how we could add response collaboration with either a built-in editor or Google Docs. We’d aim to make this configurable so that the wide variety of editors and document management systems could integrate to the tool.
One downside of an external editor is that it doesn’t include any FOI specific assistance. Fortunately many common editors (including Google Docs and Microsoft Word) support “Add-on” extensions. We mocked up an example of an “Exemption Template” Add-on, where an Information Officer can search for exemptions, view details about them, and add template text, saving time and improving consistency.
We read an interesting blog post from another Local Digital project.
More iteration on the Self Assessment Playbook, Time to Allocate Guide and Correspondence Tool after feedback from the Local Authority partners.
The highlight of the week was meeting up for a workshop to test the early-stage prototypes with the council representatives and gather feedback for the remainder of the project.
Self Assessment Playbook: In the first session we asked the council representatives to use a paper-prototype of the playbook. Up until now they’d only been assessed by mySociety, so this session was aimed at capturing the feeling of the room as they reflected on some tough questions about their FOI processes. We also wanted to establish how “good” the results were – did they reflect reality? The session was really insightful. We found that it was quite difficult to choose some of the answers because they applied in different ways across the organisation. We also talked about how to display the results in a way that doesn’t feel like a “scoreboard”. It's important that appropriate People, Process and Technology make up the overall FOI & SAR system, so we need to try to reflect that in the results generated by the Self Assessment. Though the partners have been influenced by mySociety’s assessment in the discovery phase, the results of the prototype matched fairly well with what we expected. We’re looking forward to the next iteration of the tool, and hopefully trying it with some councils who haven’t been part of the project.
Time-to-Allocate Guide: The companion to the Self Assessment part of the playbook are “guides” that illustrate problems and offer suggestions on how to improve in that area. We’re taking a deep dive in to the Time-to-Allocate problem with the aim of prototyping one possible guide that the playbook can present. In this session we described our understanding of the problem so far.
We then ran a roleplay session where the council partners thought about how they’d question someone from another council to see if the Time-to-Allocate problem is affecting them. We wrote questions on sticky notes, and later grouped them by each attribute – People, Process and Technology. These questions will help us form the basis of the guide on how to notice the problem, and get it resolved.
MOJ Correspondence Tool: In the final session of the day we demoed the Ministry of Justice's Correspondence Tool. Everyone was pretty impressed! We were particularly impressed with the reporting, and the confirmation of allocation – something that would help track important metrics in relation to the Time-to-Allocate problem.
The biggest question that remains unresolved is how to keep the clean, slick feel of the tool while making it customisable for the needs of varying sizes of authority and request volume. We’re hoping to give this some thought over the coming weeks.
Improving the question format in Self Assessment part of the playbook. We think the current questions need more focus on the consequence of the problems. For example, a better scale for assessing the current level of training might be “people don't really know what they are doing and have to ask all the time”, all the way up to “people are really confident and only have to go to the central team in rare complex cases”.
What would we need to do if we took the Correspondence Tool in to Beta. How would we need to adapt it for local authorities, rather than a single ministry?
Using the feedback from the workshop to iterate on the prototypes.
Alex and Gareth attended the Bradford roadshow to present our background research and get some early feedback on our playbook plans. We had some good discussions around our background research and prototype plans, and the other sessions we attended were really interesting too. We spoke to the DPIA project team as there’s similarities in our prototypes, so we’re now keenly following their progress on pipeline.
Martin’s shared a draft of the discovery phase write-up for our comments, so we’re pretty close to getting it published.
Louise is thinking about the business case around reducing inefficiency in the request handling process. We’ve sent off our initial data to the Local Digital team. Louise and Sarah met up to chat about the challenges around procurement.
Sarah hosted a roundtable discussion on the project at Share Digital in Birmingham.
We squeezed in another interview with a member of the team at Suffolk about SAR handling.
We’ve organised to meet in person for a workshop on March 18th in Hackney, where we’ll review the prototypes so far and run some sessions to guide the next iterations.
Gareth and Louise had a call with the Ministry of Justice for a demo and Q&A around their open source correspondence tool. We were pretty impressed, so we’re going to demo it at the Workshop next week.
We did some deeper thinking about the self assessment playbook and found some nice examples.
The SCIO has a self-assessment toolkit, but it covers more the “what”, rather than the “how” that we’re aiming to help with.
The ICO has some nice Data Protection self-assessment tools that highlight “next steps” based on your answers to their questions.
Atlassian’s Health Monitors have a nice interactive format for rating key team attributes and suggesting relevant guides.
The Madison Public Participation Playbook allows user generated content in the form of highlight comments.
We distilled the examples from our research to the key workflow that we’re aiming to prototype:
Some of the council partners are attending the Local Digital Agile for Teams training for three days.
We’ll be kicking the week off together in Hackney for our workshop.
Had an extended conference call (it's a shame that not everyone can access video calls) where we chatted about the discovery phase and proposed prototypes.
We were pleased to hear that the discovery phase had helped to frame the challenges faced by authority partners and gave them a way of thinking where to focus improvement efforts next.
We also discussed which direction the remainder of the project should take, with broad agreement that we should focus primarily on prototyping an FOI playbook, and thinking about the Time-to-allocate problem.
The idea behind the FOI playbook is that there’s currently no clear path for understanding and improving the people, process and technology areas of an FOI service. Our idea is to build a tool that helps councils self-assess to give them some indication of how they compare to others, and offer suggestions on how to tackle the most likely problems that occur based on what emerges from the self-assessment.
Reducing time-to-allocate is critical in giving enough time to gather information and compile an approved response. We think there’s a gap in the tooling that masks this problem. Improving visibility of time-to-allocate would help pinpoint workflow problems – hopefully ones that we can start to address through the playbook.
In the remainder of the week we started digging into what prototypes of these might look like.
We thought about what questions you might ask during a specific workflow: https://imgur.com/a/Z2cFm8i
We made some super early sketches of how a playbook might look: https://imgur.com/a/I3Y7iuF
Louise has been thinking about the business case in terms of social, user, financial value.
We’ve been looking at how other organisations have approached the concept of a playbook, to take inspiration from the better ones. Here are some we liked:
We’re also trying to figure out the right “depth” for the playbook: just how much detail is useful, and how much becomes off-putting or impractical?
An in-person workshop with attendees from partner councils would be a great way to start answering some of these questions, and to brainstorm ideas for guides/self assessment questions.
Some of the self assessment could be around a concept of standard comparator requests: if answering a FOI request is hard/expensive, that’s an indicator.
When thinking about self assessment tools we’re also considering the variance between service areas – different service area responses might indicate that cultural change is required, eg FOI comes up in the weekly meetings.
Katharine (Information Governance Lead at Hackney) sent over some analysis she’d previously done on overdue FOI requests. The stats are very interesting. One striking question that occurred to us is “how do you know what you don’t know”? If you have to ask several colleagues before you can ascertain that the information is not held, or not held in a returnable form, then those requests are going to be the most time-consuming, but you might not know that at the beginning of the process.
Alex and Gareth are attending the Bradford roadshow to present our background research and get some early feedback on our playbook plans
Martin’s aiming to finish up the discovery phase write-up.
Sarah is hosting a roundtable discussion on the project at Share Digital in Birmingham.
Louise is thinking about the business case around reducing inefficiency in the request handling process.
We were fully focused on developing the research and problem definitions last week, so the weeknotes fell behind, hence the joint update!
Back from the mySociety retreat, we spent time developing our ideas from the session we had there, putting them into a presentation, so that...
...we could then coherently present the research and theories so far to the Show & Tell group.
We refined and recorded versions of the Show & Tell presentations so that we could share with council representatives who couldn’t make the Show & Tell calls. It also allowed us to address questions that came up from the initial presentations:
12th Feb – our research on the similarities and differences between the partner councils: https://youtu.be/EDeeSAW_QF0 (10 mins)
19th Feb – the main problems we found, and the prototypes we’re suggesting we could progress with: https://youtu.be/V_6u6UAbGG8 (6 mins)
Sarah and Gareth co-led a session at the Local Digital Roadshow – here are Adam’s notes. It was good to get some external eyes and ears on our findings so far and confirm that we weren’t way off the mark.
During the discovery phase of the project we noticed that there were opportunities for small, single-purpose tools. One observation we made was that several councils create “due date” sheets – calculating the due dates of requests based on the date they were received. We hacked up a quick prototype in to Alaveteli – the software that powers WhatDoTheyKnow – and it looks like its something we can provide for everyone:
Universality: what are the main problems that broadly apply to all councils when dealing with SARs and FOI requests?
We’re starting to think about the business case for an open source FOI platform.
We’re hoping to present our background FOI research at the Bradford Roadshow
We’ve presented what we think are the main problems we’ve discerned in current request processes, and asked the group to rank them, together with some of our prototype ideas to address them. The next Show & Tell is an extended call so that we can deliberate how to take the concepts forward. Should be fun!
It was mySociety’s annual retreat this week, which saw us meeting up in a big house in the country and working face to face for once. This gave us a chance to talk through where we’ve got to so far and what the next steps will be.
During the retreat we designed a rough framework for comparing councils’ FOI handling capabilities.
We also started thinking about which prototype ideas will have the biggest impact.
How does a differing focus on People, Process and Technology affect an FOI system?
Where does knowledge lie in the council organisations, and how is that best extracted in response to information requests?
What prototypes should we go ahead with, in which of these three areas?
Presenting our preliminary research findings on the weekly Show & Tell call.
We’ll be at the Local Digital Roadshow.
More interviewing, by phone this time, and across the whole range of councils! We had really useful chats with Dumi and David, our main project contacts at Stevenage & East Herts and Cornwall, then moved on to no less informative conversations with other staff too: three from Cornwall, four from Stevenage & East Herts, and one from Suffolk.
In between all that, we were writing up our notes from these interviews and last week’s in-person sessions in Hackney.
The overlaps in each council’s process when it comes to dealing with requests: the processes all seem reasonably similar, but we discovered very different levels of contentment expressed within different authorities. Might headcount be a factor?
Whether there’s a size threshold where a council needs to move from lightweight processes to something more prescriptive.
What the final format of our report should be — we want people to read it, so it has to be accessible and not too dry.
What tool improvements could help improve processes for everyone.
We’re interviewing two more staff members, from Suffolk.
It’s mySociety’s annual retreat, a couple of days where we all go away together to work intensively, face to face. We’ll be holding a session on this project to get together and review all our notes from the interviews, observations and research and define the scope of what’s next.
On Monday we spent an enormously useful day, speaking to eight people from Hackney who work with FOI and SARs.
We came back on Tuesday to run six observation sessions, taking notes on how staff approach the various stages of fulfilling a request.
We also had a call with Alex Nash about the open source correspondence tool developed at the Ministry of Justice. It sounds like there might be some shared interest, so we’re getting the tool up and running to have a look at.
Then we spent the rest of the week writing up our notes, and starting to digest the information we have so far.
One theme that emerged from our sessions in Hackney was that of reporting FOI activity to the wider world. The Scottish Information Commissioner requires some data to be submitted for performance monitoring, so we should look into this as a starting point.
We also noted that minimising the time to assign to the people who can provide the information asked for in FOI requests is key for giving the responder enough time to gather the information on time.
We’ll be interviewing again, but this time remotely. We’ll be talking with staff in Cornwall, Suffolk and Stevenage.
This project involves several people working in different locations around the UK. It will generate a large quantity of information as we learn about users’ needs and refine the project scope. It’ll move fast and it’s really important that we’re all kept in the loop.
Enter #weeknotes! Via this simple format, the aim is to record the project’s progress (and the process) where everyone involved can see it. We also hope that they’ll build into a permanent record that any other councils interested in the decisions and processes we went through will be able to access.
Please feel very free to let us know your thoughts and comments.
Kick-off! A call between mySociety and all partner councils.
We’ve scheduled a weekly Show & Tell call where we can demonstrate progress and receive feedback.
mySociety presented our draft user research plan, and we’ve got the ball rolling on scheduling interviews and observation. We want to get this underway as soon as possible in the next two weeks (scheduled for 21st & 22nd January). We’ve got a day of interviews and a day of observation.
In the interviews we’re aiming to talk to FOI team members from all partner councils, as well as individuals from other council departments who routinely take part in the FOI/SAR processes.
The observation sessions will cover Hackney Information Governance staff using Infreemation.
mySociety has been rereading our existing internal research on FOI statistics and local government FOI admin, as well as process documentation sent over by Hackney Council. This is helping us narrow down the questions to ask when we come to interviews.
Gareth and Martin took all the post-its from the earlier workshop during which people had shared Fears, Hopes, Pain Points, Users and User Stories, and typed them up.
How to make best use of the Trello board where we’re organising all the project to-dos.
Whether those “other councils interested in the decisions and processes we went through” mentioned in the intro to this piece, actually exist… or will find these notes if they do.
How to get this post added to Pipeline!
Sarah from Hackney is aiming to get the project set up on pipeline.localgov.digital as a central place to post updates. (As you’re reading this, obviously she succeeded!)
mySociety will complete work on the User Research plan, laying out a few different options on the scope.
Process documentation continues to come in from partner councils, so mySociety are reviewing this to aid preparations for the research interviews.
We’ll finalise our plan for working in the open on this project. We’d like to share previous user research and prototyping/testing reports with partners, and publicly.