1. Stats, technology and at least one really bad joke about tapirs: the mySociety year in review

    If you’re wondering what a year in Civic Tech looks like, well, here’s the answer:
    plenty of coding, video calls with partners around the world, the occasional conference, and tea. Lots of tea.

    Oh, and then there’s the annual treat of posing for the team photo on a windy winter’s day. That’s us, above. Damp. Cold. Still believing in the transformative power of digital technologies.

    We bundled it all together, with plenty of stats, a couple of jokes, and tweets from some of our happy users.

    Here’s the result. Sit back and enjoy the mySociety Year in Review, 2015.


    Read the 2015 mySociety Annual Report

  2. Boom times for FixMyStreet

    Each of the previous three months has been a record-breaker for FixMyStreet. In January, you made the highest number of reports in the site’s history… until February. And then that record was smashed again in March with over 17,000 reports across the month.

    FixMyStreet has been running since 2007, and it’s enjoyed increasing usage over that time, as you’d expect any site to do organically. The performance in the last few months, though—a 30% rise from the year before—has been notable. We reckon it’s been driven by a couple of factors.

    Grassroots outreach

    At mySociety, we tend not to go for big advertising campaigns (read: we can’t afford them), but you might have noticed that we put quite a bit of effort into spreading the word about FixMyStreet at the beginning of the year.

    Everything we did was low-cost and designed to help us promote the site to as many new people as possible:

    • We offered a number of downloadable posters and other promotional materials (if you haven’t seen these yet, go and take a look; we think they’re pretty nice)
    • We sent our users a stack of branded postcards that they could share with others to let them know about FixMyStreet
    • We also contacted a large number of community newsletters and magazines, serving towns, parishes and villages across the country: perhaps you saw us featured in your local publication.

    Users from council sites

    That all paid off, but there was another source of reports helping us achieve our record figures.

    That source was our client councils, who have FixMyStreet as the primary fault-reporting system on their own sites.

    Eight UK councils currently have FixMyStreet installed, with every report made on via the system on the council site being published on fixmystreet.com, and vice versa.

    Between them they’ve added just over 16,500 reports this year.

    Riding the wave

    So far this year, we’ve seen an overall average of 16,000+ reports per month, and there have been over 50,000 reports since 2015 began.

    Now, let’s hope all those reports get some kind of a response, because as the recent research we collaborated on showed, getting something fixed has the power to turn first-time reporters into conscientious, engaged repeat reporters. And that’s all for the good.


    Image: Jamie Taylor (CC)

  3. How responsive was your MP in 2014?

    This post refers to the statistics we published for 2014. If you would like to see the latest responsiveness figures on WriteToThem, please visit www.writetothem.com/stats/.

    Of course, there are many factors that you’ll consider before you cast your vote in the general election. But we think that one important quality in an MP is that they respond to their constituents.

    So you may wish to check your own MP’s performance on the latest WriteToThem responsiveness league table. Just put in your postcode and you can see how they did in 2014.

    Where the data comes from

    When you send a message to your MP using our site WriteToThem, you’ll receive an automated email two weeks later, asking whether or not you received a response. Every year, we take the data from these surveys and use it to assemble our responsiveness rankings.

    A downturn

    You might think that MPs would be doing the best they can this year, in the run-up to the election. Sadly, that doesn’t seem to be the case: overall, responsiveness has fallen a percentage point since last year, with 46% of emails receiving no reply.

    Some caveats

    You can find all our data and methodology on the league table page.

    We know that messages sent to WriteToThem may not reflect all messages sent to an MP; we also know that not every message will require an answer. However, we think that, taken overall, our sample size of over 36,000 interactions can be seen as indicative.

    Image: Michael Scott (CC)

  4. Vikings, One Direction and Phillip Schofield: mySociety in 2014

    Year in reviewAnnual reports: they’re dry, boring and full of corporate buzzwords, right?

    That’s not really the mySociety way, though.

    All the same, we wanted to share some facts and figures about everything we got up to last year. It’s in the nature of our work that people tend to know about one part of it—say, our international work, or the sites we run here in the UK—but nothing else.

    Well, to give you a more rounded picture, here is the mySociety annual report, featuring, among other things, the pop group One Direction, some vikings, and the TV presenter Phillip Schofield.

    Welcome to mySociety in 2014… and if you enjoy it, please do share it around!

  5. WhatDoTheyKnow’s Share of Central Government FOI Requests – Q2 2011

    The Ministry of Justice have just published their latest quarterly statistics on the handling of Freedom of Information requests by central government bodies.  We’ve crunched the numbers to compare them to the requests made using WhatDoTheyKnow.com

    The graph shows our share of FOI requests sent to central Departments of State jumped to 14.6% in the 1st quarter of 2011.

    This time round, the top 3 departments were:

    1. Home Office (which includes the UK Border Agency, CRB & Identity & Passport Service) – 254 requests out of 866 – 29%
    2. Department for Education – 81 requests out of 328 – 25%
    3. Department for Communities and Local Government – 59 requests out of 250 – 24%

    Many of the WhatDoTheyKnow users contacting the Home Office & UK Border Agency are trying to find out information about their own immigration case.  We regularly receive emails from applicants asking for help, as they have often been waiting months (or even years in some cases) for an official update to their case, often with the UKBA holding on to identity documents or passport.  Applicants then feel they have to resort to making FOI requests. Many of these are auto-replied by this standard FAQ, and applicants don’t receive a personal answer.  The large 29% share of all Home Office requests suggests that the normal contact methods to keep people updated aren’t working or even that their service is simply struggling with demand.  It’s also likely that they don’t consider these types of requests as formal FOI requests, so it is worth noting that we are likely to be slightly overstating the percentage share figures.

    Free schools were a popular topic for the Department of Education – 9 out of 81 requests were on this subject, and nearly all were refused on the basis that information would be published at some unspecified date in the future.

    To understand the limitations of the data analysis, please see here.

    One interesting trend that has been consistently seen is that FOI requests are more frequent in odd-numbered quarters compared to even ones – if you have any ideas why this may be the case, please add them to the comments!

    – Communities and Local Government
  6. Trying to Practice What We Preach: mySociety Evaluation Reports Published

    Transparent screen by Kathy McLeod

    Transparent screen, by Kathy McLeod, used under the Creative Commons licence, with thanks.

    At mySociety we like transparency – it’s baked into most of our projects.

    TheyWorkForYou attempts to make it easier to find out what your MP has been doing in Parliament. WhatDoTheyKnow tries to make it easier to find out what’s going on inside other public bodies. FixMyStreet and the upcoming FixMyTransport also use transparency to help get problems resolved.

    We think transparency is a good thing for many reasons, but one of its rarely mentioned virtues is how valuable transparency can be for the people within the organisations which are transparent.

    Transparency can be useful because it means people outside an organisation can make critical, constructive suggestions about how you can improve, and it lowers the odds that people in one part of your own organisation will be ignorant of the activities of people in other parts.


    To that end we commissioned Tobias Escher of the Oxford Internet Institute to conduct detailed analysis into two of our major websites – TheyWorkForYou and WriteToThem.

    We were not highly prescriptive in our instructions, and we certainly didn’t ask Tobias to ‘discover’ pre-determined findings. All we did was ask Tobias to find out who was coming to the sites, what they were doing, and whether or not the sites could be considered to be succeeding.  We didn’t do it for a PR stunt:  we did it so we could learn from our mistakes, and so that we could share those learnings with others who might benefit.

    His detailed, quantitative analysis holds the sites up to mySociety’s own stated aims, for the first time. And we’ve published both documents, in full, below.

    Swings and Roundabouts

    It was great to discover that we have, indeed, attained some of our goals by running these sites. For example, one of the reasons we set up TheyWorkForYou and WriteToThem was to make representatives accessible to people who were newcomers to the democratic process. It was therefore heartening to read that 60% of visitors to TheyWorkForYou had never previously looked up who represents them, and two in five users of WriteToThem have never before contacted one of their political representatives.

    But, as you would expect with any properly neutral evaluation, it’s not all good news. Our sites aim to reach a wide range of people, but compared to the average  British internet user, WriteToThem users are twice as likely to have a higher degree and a higher income. It also seems that users are disproportionately male, white, and over 35. These figures and many more are available within these highly readable papers – Tobias did a terrific job in gathering and analysing a huge amount of data, and then making it easy to understand.


    These reports are rich with data, from how visitor numbers boomed during the MPs’ expenses scandal to which MPs most people sign up to receive alerts about. You can also read how a budget airline almost brought a site to its knees in 2007; what part Joanna Lumley plays in our history; and how many visits to TheyWorkForYou actually come from within Parliament itself.

    TheyWorkForYou and WriteToThem have inspired many people around the world to set up similar (and not so similar) sites inspired by the vision of using the Internet to lower barriers to democracy. However, until now we’ve never seen a really clear-eyed assessment of what seems to work, and what doesn’t.

    If you’re at all interested in using the Internet to engage people with democratic systems, Tobias Escher’s excellent research papers will make a compelling read. Thank you Toby!

    The reports

    Download Tobias Escher’s research report on WriteToThem.compdf

    or TheyWorkForYou research report by Tobias Escherpdf

    …and do come back and tell us what you found interesting.


    We hope to publish two evaluation reports like this at the start of each new year from now on. Next year’s sites will probably be FixMyStreet and WhatDoTheyKnow. Do get in touch if you’d like to input!

  7. FixMyStreet in Norway

    At mySociety we love our site FixMyStreet – it’s the epitome of a web tool that gives simple tangible benefits whilst generating a little accountability at the same time. Reports through the site were up 40% last year, so it’s clear that users quite like it too.

    FixMyStreet has been copied in many different countries, which makes everyone in mySociety very happy, too, even apparently appearing in a slide deck the White House uses to show innovative services. However, it turns out that the cheerfully minimalist, almost wantonly unfashionable user-interface has an unfortunate down side: most people who copy the site look at it, think “That looks easy!” and then cheerfully start coding their own clone.

    Deceptive simplicity

    Alas – the very simplicity that makes the site good hides the fact that making a site like FixMyStreet really work well is actually way harder than it looks. What will you do when a government email inbox fills up? What about when administrative boundaries change, due to an election or restructuring? How do you know you’re not scaring users away with careless wording? All the hard-won lessons from these questions have been baked into the FixMyStreet codebase, and we’re only too keen to talk to people about them.

    We were therefore particularly pleased when the Nowegian Unix User’s Group (NUUG) came to us to ask if we could help improve FixMyStreet to make it easier for them to install. Over the last month mySociety Senior Developer Matthew Somerville has been working hard with Petter Reinholdtsen and Christer Gundersen of NUUG, and here’s what they’ve managed in just a handful of weeks.

    • The launch of a Norwegian FixMyStreet called Fiksgatami, covering nearly every corner of Norway’s 300,000 square kilometers.
    • Problems reported anywhere within Norway will be correctly directed to any of the 400+ responsible municipalities, thanks to Petter and Christer’s amazing data sourcing skills.
    • As a necessary side-effect of developing this, Norway now has a free, public administrative web service gazeteer – http://mapit.nuug.no. If Norway is anything like the UK this will soon become an indispensable service for many other web sites and mobile tools.
    • The standard mapping is now OpenStreetMap, pulled together by the brilliant Norwegian OpenStreetMap community*. We couldn’t take a technological step backward, and so whilst the site uses OpenLayers if you have JavaScript, the map continues to work just fine without as well.
    • The open source FixMyStreet codebase has been upgraded to make it easier to translate into other languages, easier to use different mapping with, and easier to install. These efforts will continue, as we realise this has been one reason why others have made their own versions.
    • All this has been done without forking, so various major upgrades we have planned for the UK version will be exportable later in the year.

    NUUG’s Fiksgatami is the epitome of what makes civic open source at its best so unmatchably good. It was developed incredibly quickly: just a month to create what is effectively a fully fledged, best-of-breed nationwide e-government service – albeit an unofficial one. Thanks to the hard work of the public servants who fix problem reports, it will make small but meaningful improvements to the lives of a lot of people in Norway.  And it has made the free FixMyStreet codebase better and easier for other people to use to help them do the same thing in other countries.

    I know that at mySociety we are all looking forward to working with NUUG again. And I hope that this story inspires others to look at our code, and to work with mySociety to make FixMyStreet a service that can help everyone who would benefit from it.

    * We’ll be rolling out updated mapping (including OSM) and more in the UK, soon.

  8. Some FOI stats on local authorities

    I recently found these requests by James Muldoon covering FOI statistics for the London Boroughs for 2009.  As we regularly carry out analysis of WhatDoTheyKnow’s percentage share of FOI requests to central Government Departments of State, I thought it would make for an interesting comparison to do the same for the 33 Metropolitan borough councils, plus the City of London.

    Below is a graph of the market share for WDTK.

    Overall, the share for 2009 was 8.1%.  During the year, the share did fluctuate quite a bit, and the requests on WhatDoTheyKnow were significantly lower in the 2nd quarter for some reason.

    Q1: Jan-Mar 2009 – 9.4%
    Q2: Apr-Jun 2009 – 5.1%
    Q3: Jul-Sep 2009 – 9.5%
    Q4: Oct-Dec 2009 – 8.3%

    The City of Westminster has a much higher number of FOI requests compared to the other boroughs, mostly apparently due to a large motorcyclist parking campaign/protest.  73% of all requests made to Westminster via WhatDoTheyKnow in 2009 contained the words “parking”, “motorcycle” or “Verrus” (203 out of 278).

    I will soon start looking for FOI statistics for Local Authorities outside London, either on WDTK, or via their disclosure logs.  The Ministry of Justice encourages Local Authorities to regularly publish statistics on their FOI data.

    Data caveats

    • Brent – excluded from totals & comparison as the underlying FOI request is still outstanding.  The ICO is apparently investigating.
    • Camden –  Q1-2009 data excluded from totals & comparison due to partial refusal to the FOI request by Camden (FOI Act Section 12, costs of complying too high)
    • Southwark – excluded from totals & comparison.  They said in their FOI response:  “due to a serious malfunction of our reporting database we have no access to the data stored centrally”.  The data has been re-requested by James to see if the malfunction has been fixed.
  9. WhatDoTheyKnow’s Share of Central Government FOI Requests – Q1 2010

    The Ministry of Justice recently released the latest statistics on freedom of information implementation in central government for the first quarter of 2010.  We can use this data to roughly calculate the share of FOI requests made via mySocety’s Freedom of Information website WhatDoTheyKnow.com.

    According to the Ministry of Justice 6,857 requests were made to “Departments of State” between January and March 2010, of which 707 were made via WhatDoTheyKnow – a 10% share.  Defra and DCSF had the highest percentage of WhatDoTheyKnow requests in the first quarter, at 23% and 24% respectively.  We have made the departmental breakdown of the statistics for the past two years available as a Google Spreadsheet.

    Precise calculations of the percentage share are not possible, due to the way each department defines a valid request.  For example, some FOI requests to the UK Border Agency (included in Home Office data) or the FCO made via WhatDoTheyKnow would not have been counted as FOI requests, but instead as “routine requests for information”, falling outside the scope of data collection for the statistics produced by the Ministry of Justice.  In addition, requests for data readily available (i.e. exempt under Section 21 of the Freedom of Information Act) are not counted either by the MoJ, but will be included in the WhatDoTheyKnow statistics.  Overall, these form a minority of requests made by WhatDoTheyKnow, so the percentage share we quote is fairly accurate.

    WDTK share of central departments’ FOI requests

    WDTK share of central departments’ FOI requests

    The graph shows that the total numbers and percentage of requests have been fairly consistent over the past year, around 10-13%.  Share of requests increased significantly during 2008 as the site grew in popularity following its launch.

    The large increase in total FOI requests between Q4 2008 and Q1 2009 may partly be due to the increased popularity of WhatDoTheyKnow and also increased visibility of Freedom of Information to the general public through more mentions of FOI law in the media.  Key stories in the press included Jack Straw’s veto against disclosure of the Iraq War Cabinet Minutes and Parliament’s aborted attempt to hide details of MPs’ Expenses. Requests made via WhatDoTheyKnow accounted for around 20% of the total increase in FOI requests.

    It’s worth bearing in mind that the above graph only covers the 22 central Departments.  WhatDoTheyKnow.com lists over 3,750 authorities including local government councils, police forces, NHS Trusts, non-ministerial departments and many more, and also covers Scotland’s public authorities. There is no national monitoring of how many FOI requests are made to such bodies, or how well they perform when responding to requests.

    In total, 6,565 requests were made via WDTK in the first quarter, 87% of which were to non-central Departments.  It’s unlikely that the site will ever get full coverage as the majority of FOI requests originate from companies carrying out market research, journalists, political parties and charities, most of whom prefer to make requests in private in order to maintain exclusivity over the released data.  WhatDoTheyKnow would like to include these companies in future, possibly by offering a managed FOI service on a fee basis, including an embargo period prior to the released information being made fully available on the internet – more information is available from the WhatDoTheyKnow team.

    Articles based on previous statistics releases:

  10. Hansard Society report on MPs Online

    The Hansard Society have just published a report entitled MPs Online: Connecting with Constituents. I’m only going to talk about one part, the part that mentions the mySociety project WriteToThem in a section on MPs’ use of email.

    We’re surprised and disappointed to see our methodology for collecting data on how well MPs respond to constituency mail being called “unreliable”, especially from a paper that makes a number of simple mistakes of its own in just a few lines on one page.

    • On page 5, they state that WriteToThem has been “tracking responsiveness to emails via their website for three years”. Most importantly for the theme of the report, we don’t just send emails – we send faxes to a number of MPs (and other representatives) who do not accept or want messages via email.
    • The figures given for survey responses in the table are backwards; 2007 and 2005’s figures should be interchanged – how could we get more survey responses than messages (again, not necessarily emails) sent? 🙂
    • They claim there is “no quantification of the response categories provided” – the raw data used to automatically generate these categories is given in the adjacent column (“very high” simply means a response percentage of 80% or more, for example; our code is all open source).
    • We exclude MPs with very small sample sizes, and take a range of steps to make sure the data is not abused.
    • We have four years of statistics now, not three; our stats for 2008 were published nearly six months ago.

    The Hansard Society, to the best of my knowledge, never got in touch with us to request any clarification or ask about our data or methodology, which we would have been more than happy to supply.