User experience in software development

Yesterday I spoke at a Software East user experience (UX) event. I’ve not attended any Software East events before, so didn’t know quite what to expect. But the event was expertly organised by Mark Dalgarno and generously hosted by Redgate Software in their swanky offices (which reminded me a bit of Google HQ in London!). Almost 100 people attended, which is quite something for a mid-week event in Cambridge. I was particularly impressed to find out that one intrepid person travelled up from Kent!

I spoke about my experience of embedding UX at Cambridgeshire County Council which I’ve written about previously on this blog. I wanted to get across the sense that embedding UX is a long journey, but that UX techniques are not hard to learn, providing you start with small projects where you can demonstrate the value to your colleagues.

I also talked about how to use data to tell a story about your users and win over stakeholders in the business. And the importance of sharing skills across your team.

Here are my slides from the talk:

There were three other talks which I thoroughly enjoyed.

Things we learned when redesigning the Red Gate website, Stephen Chambers (Red Gate)

Stephen Chambers, the Head of User Experience at Redgate Software spoke about the re-design of the Redgate website. He started by showing the evolution of the site over the years and development of the product page, the most popular area of the site. As the product range grew, so did the page! But without much consideration to layout. Stephen also explained that in the past they’d struggled to get consistency between the main product areas of the website and that individual business areas were protective about their product pages.

Part of the challenge was getting the right environment in which to tackle a complete re-design of the website and dedicated resource. But Stephen managed to persuade his bosses to let his UX team take over a meeting room and work solidly on the re-design for 5 weeks – yes 5 weeks! At the council you’d be lucky to get a room of that size for 5 hours, let alone 5 weeks! One of the benefits of having the dedicated space was that the team were able to  move their computers in and put up all their ideas, sketches, wireframes and visual mock-ups on the walls, which improved transparency about the project. Colleagues could come in anonymously and scribble feedback on a post-it and stick it on a design.

Stephen also said this was the first opportunity the UX team had to work together, rather than on separate projects. And this helped them to make the most of their individual skills. There were some lessons learnt, such as not committing to hi-fidelity wireframes too early on.

The actual development of the site took around 3 months following the intensive 5 week design phase. One thing that struck me was the team’s creativity within the constraints of the project. The team used existing research, but I didn’t have the opportunity to ask Stephen if they’d done any user evaluation during the 5 weeks.

Why did you click there?  How to run 1-to-1 usability testing – Jenny Cham (EMBL-EBI)

Jenny gave an insightful talk on one-to-one, moderated usability testing for the European Bioinformatics Institute (EBI). Jenny explained she had come from a science background and was relatively new to UX. She demonstrated how it’s possible to pick up techniques quickly, but that you learn a huge amount through hands on experience.

Jenny’s domain is very specialist, so her knowledge of genome research and analysis of data was useful in designing realistic user tasks for her participants. This point highlights that in some instances using a consultant from outside your organisation would not be appropriate, because they won’t necessarily have the right domain knowledge.

I really liked Jenny’s use of photos and speech bubbles in her slides and a clip of user testing footage from Silverback. Her presentation really helped to convey some of the subtleties of user testing. For example the importance of good planning up front, during recruitment and when scheduling tests, having the right equipment and putting the participant at ease in order to get good test data.

One tip I found useful was Jenny’s use of a grid to collate findings and post-it notes (on the wall of the testing room) to summarise her research findings quickly. She advocated the use of super sticky post-its for this purpose!

Jenny had a wealth of experience to share, despite claiming to be a relative newcomer to the UX field.

Remote User Testing 101, Rob Kerr & Neil Turner (Cambridge Assessment)

Rob and Neil of Cambridge Assessment did a talk on the pro’s and con’s of remote, moderated and unmoderated testing. This is an area I’ve had an interest in myself and wrote about a while back, but at the council we’ve only had the chance to try remote, unmoderated testing.

Rob and Neil suggested various different testing tools that are either free or very cheap and can be used online. One of the questions from the audience was whether remote tools can be used for desktop software development rather than web. It seems that most tools are web based and therefore suited to web development. However Rob explained that remote, moderated testing was possible providing users remote in to your desktop (which does involve them downloading and installing some software on their computer).

Neil also demonstrated how Loop11 worked as a remote tool for testing user tasks, a tool I’m fairly familiar with having run some tests for our corporate website. I volunteered to do a live demo which involved using the Yahoo site to find out information on a film, whilst my actions were being recorded by the tool.

One of the things Neil pointed out was the importance of getting the task design right and piloting tests, as many of the tools (like Loop11) don’t allow you to make changes to your test once it is published and live.

And so ended a very successful evening. I enjoyed chatting to some new people in the break and after the event.

As a follow on, Mark is planning a UX conference in Cambridge later this year. I’ll be posting more details about the conference as soon as they are available.


Coverage of council meetings by local bloggers

The coalition government has been pushing for greater transparency of Council decisions and spending as part of their Localism Bill. So I wasn’t that surprised to read Nick Booth‘s post about a recent press release from Eric Pickles.

Pickles has told Councils that they should be allowing hyerplocal bloggers to record public meetings. I’ve heard anecdotes that some Councils refuse to acknowledge hyperlocal bloggers or allow them the same access and rights as local media. So it will be very interesting to see how this latest advice from Pickles is taken on board.

I have been really impressed with the way Richard Taylor (unquestionably Cambridge’s most dedicated hyperlocal blogger) has been covering public meetings. Richard has provided an entertaining yet thorough running commentary of recent Council meetings, enabling those of us who are unable to attend to follow proceedings online, or catch up afterwards.


Software East – UX in software development

Just a quick plug for an event happening in Cambridge run by Software East, which is focusing on user experience in software design. The event is on Thursday 3rd March from 6-9 pm at Red Gate Software.

I was very flattered to be asked along by the organiser Mark Dalgarno to do a short talk on how we’ve embedded user experience design at the Council. I’m looking forward to the other speakers too, which include Rob Kerr and Neil Turner from Cambridge Assessment, Stephen Chambers from Red Gate and Dr. Jenny Cham from EMBL-EBI.

I haven’t attended any of the Software East events before, so I’m looking forward to meeting a new bunch of geeks in Cambridge and talking about UX and Agile development.


Service design and localism

I’ve been doing a bit of thinking recently about a approaches to service design in local government in the light of the government’s localism or Big Society agenda.

The way I see it there should be a balance between the following elements:

Service design - business strategy, democratic engagement, insight and data

Business strategy

Let’s face it, business strategy in local government is primarily about cost cutting right now either through streamlining of business processes, use of cheaper channels, selling / transferring assets and working in partnership.

Democratic engagement

Listening to local communities and working with them to co-produce better public services and outcomes is one of the cornerstones of localism.

Customer insight and open data

This to me is the really interesting element in the triangle. Customer insight is data about what different customer groups want and need, what services they already use and what channels they prefer using. Data can be gathered from channel usage statistics and customer research. Open data is all about transparency and making data about council services, assets, performance and spending accessible.

Once data is out there in the open, it can be used to tell stories, which make it more meaningful. In turn this can stimulate ideas and provide new perspectives on how to deliver services.

So how might this model work in practise? If all three elements of the triangle are brought together:

  • Councils would need to clearly define the constraints – i.e. what cuts they have to make and where – and work with partners to explore new ways of working together to deliver public services
  • Communities, council officers, councillors, partner agencies and other stakeholders will come together in public networks, online or offline, to discuss how they want services to be delivered
  • Data will inform conversations and underpin decisions

Let’s assume in an ideal world that service design is an iterative process, whereby services are analysed, prototyped, tested and improved continually. Ideally each element in this model should be considered at all stages of an iterative process, to make sure one element is not dominant over the others. If business strategy dominates services may not meet community needs and if democratic engagement dominates services may be designed around more vocal groups within the community.

I’m not sure what the impact might be of data being a dominant element in service design. Or if it matters. This is just a way for me to think through service design and localism and I’d be interested in what anyone else involved in service re-design in local government thinks.


Did you forget to send the attachment?

Since I started using email back in the early 1990’s there have been countless times when I’ve forgotten to attach a file to an email. I know I’m not the only one who suffers from this very common phenomena known as a post completion error.

The crazy thing is that errors like this should be considered in the design of software that supports such common tasks as sending an email. Early ATM machines used to give you your money back before your card, which resulted in many people leaving their card in the machine. This was resolved by simply changing the order of tasks, so now you get your card before your money. Which, after all, is the thing you went to the machine for in the first place!

So why in the last 15 years or so has this common error not been tackled in email clients? Well I can’t answer that one, but I was delighted to find this morning that Mozilla seem to have put a neat solution into their Thunderbird email client. Here’s how it works…

When you are writing a new email and mention words like ‘attached’ or ‘attachment’, Thunderbird automatically detects this and pops up a non-invasive reminder at the bottom of the email (highlighted in red).

If you still forget to add the attachment, despite the reminder, and click ‘send’ a prompt pops up in a dialogue window:

A pretty simple but elegant solution to a very common usability problem.

Ironically the way I found out about this was because I was in the process of replying to someone who’d forgotten to send an attachment to an email!


#ukgc11 session on Agile lessons learnt in local government

In advance of this year’s fantastic UKGovCamp (aka #ukgc11) I pitched a session idea on lessons learnt from using an Agile project management method to manage web development. My idea for the session arose after some interest in a blog post I wrote recently and the discussion on the UKGovCamp forum thread.

So on the day I came armed with my slides and stood in line to pitch my session. I was joined by Stefan Czerniawski and Catherine Howe, who pitched a second session to explore the potential for applying Agile principles to policy. This is an area I am very interested in, but as I have limited experience of working with policy makers I ended up taking more of a back seat in their session, to observe, absorb and blog some loose notes.

Thankfully Catherine has written up a brilliant blog post about the session. Which is a relief, as I was flagging a bit at that point and was finding it hard to stay mentally alert. I think I used up all my energy facilitating the first session and also felt a bit ropey from having stayed out too late the night before – my only regret about the whole day!

The session went far better than I could have hoped for. It was well attended and resulted in an open and fascinating discussion about the practicalities of implementing Agile methods and applying them effectively.

Photo by Paul Clarke

There was a real mix of participants. Some with loads of experience and some who knew very little about Agile, but were curious to find out more. In preparing for the session I hadn’t really considered that some of the participants wouldn’t be familiar with the Agile manifesto, terminology or methods like Scrum and the roles involved. So I did my best to explain the concept of Agile and the Scrum method in simple terms in limited time and some gaps in my own knowledge. But I realise that pitching the session at the right level, to address the varying experience and understanding of those in the room, was a challenge. I look forward to Julia’s notes on the session (as mentioned in her write up of the day) as a relative Agile newbie.

I explained that we have used the Scrum method in my organisation (a county council)  for about 18 months for web development projects, with varying degrees of success. Here are the slides I presented in the session:

Notes and reflections from the session

The following notes are written up from memory, as I was too busy talking (as usual) to write down anything during the session. I haven’t attributed these insights to individuals (in case I do so incorrectly!), but they were a combined effort of the participants who attended.

It can be difficult to introduce Agile methods when senior managers are sceptical. I recommended starting small by applying an Agile method to development of a mature platform, rather than choosing a project with significant risks and unknowns.

Someone asked if you can do Agile covertly. And the consensus was – yes absolutely! Agile can fit within existing, more traditional project management processes quite neatly, e.g. within the ‘product delivery’ phase of Prince2. It was even suggested you could do Agile within the ‘implementation’ phase of a waterfall method and by the time you get to the ‘verification’ phase, you’ve already done the testing you need to. Business people don’t want to know how you are delivering something, just that you deliver results.

But…and there are two big buts…

  1. You do need the product owner involved. It can be very hard to persuade product owners to be involved in sprints and attend daily scrum meetings. Good planning ensures key individuals in the organisation are available when you need them to be.
  2. The scrum master has a key role to play in negotiating with product owners and other senior business stakeholders who insist on the delivery of particular features and refuse to compromise over the scope or prioritise their requirements.

When a development team learns their velocity over time, introducing someone new to the team can throw things off balance. Estimation poker can help the estimation process, because developers have to discuss and resolve why there are differences in their estimations.

Developers can sustain their pace when working in sprint cycles. Following a waterfall method can result in a huge rise in stress levels as you approach a big bang launch and everyone is working overtime to get a product finished. Instead there are small rises in stress levels at the end of a sprint, but developers should be able to leave work at 5 pm and maintain their pace in a sustainable way.

The velocity of the team (progress over a series of sprints) can be greatly improved by having a UX designer produce designs in advance of the sprints.

Agile involves developers in business strategy, so they work towards achieving business objectives.

Although the Agile principles promote face-to-face communication and co-location of teams is often cited as being important, it is possible to use geographically dispersed teams  if you make good use of technology. An example was given of how a team which is distributed worldwide use Skype to manage daily scrums.

Not everything gets finished at the end of a sprint. Developers must be prepared to throw code away.

Be flexible about the method and techniques you use e.g. user stories could be written up on cards and stuck on the wall, kept in a spreadsheet or a specialist application.

Agile can be applied outside IT contexts. One participant said that his organisation has applied Agile principles to business strategy projects.

I’d like to extend my thanks to Dan Hardiker, Andrew Woodward, Catherine Howe and Sharon O’Dea who all pitched in with excellent insights based on their experiences of working in an Agile context. I personally learnt loads from the other participants.

Which goes to show if you’re willing to share your own experiences, you get back a bucketful of new insights and knowledge from others in the process!


Dissertation research: perceptions of using social media for community engagement

I’ve talked about my MSc dissertation research before on this blog. In fact I originally set up this blog to explore some topics related to my research. Having had a bit of a break after completing my dissertation in August 2010, I have decided to publish it here, spurred on by the kind words of a fellow academic researcher Catherine Howe.

My research question was:

How do the attitudes and perceptions of citizens, Council officers, Councillors to the use of social media for community engagement compare and contrast?

My Master’s Degree was in Human Computer Interaction. If you have a particular interest in research into social media and civic engagement (and quite a bit of time on your hands), I’d recommend the full dissertation (PDF, 2.4 mb).

But if you are a local government officer or someone with less time and patience, then I’d recommend the 10 page (that’s the smallest I could manage!) Executive Summary (PDF, 50kb).

I also wanted to add a little disclaimer. The primary research data was gathered from semi-structured interviews with 18 participants. For purposes of confidentiality the data is not included within either document. Because the research question focused on a relatively new research area, it was challenging to find participants with significant experience of using social media, let alone those with experience of using social media for civic engagement. Whilst the collection and analysis of data followed rigorous qualitative research methods, the quality of the data collected was not as high as I had hoped for. I would therefore advise some caution in the interpretation and application of these research findings.

Please note that the usual Creative Commons copyright license I display on this blog does not apply to the two documents linked above.

You can find me on Twitter if you have any questions, comments, or would like more information about my research.


In the hotseat…

This Thursday 13th January between 1-4 pm I will be in the hotseat, quite literally! I’m hosting a online question and answer session in the Local by Social Online Conference about the social media project I have been involved with in Fenland, Cambridgeshire.

The project aimed to improve engagement with customer groups who were hard to access in Wisbech. The project has been delivered with funding from the Local Government Delivery Council’s Customer Led Transformation programme.

You need to be a member of the Communities of Practise to participate in the hotseat discussion. To become a member of the Communities of Practice platform you will need to register and  join the Customer Led Transformation CoP. You can then join the hotseat discussion thread where you will find the full project case study and some vox pop videos with people who contributed to the project.

If you would like to know more about the project or ask me anything about using social media to engage customers, do come and contribute to the online discussion on Thursday 13th January, between 1-4 pm

By way of a little background, I’ve written about the project before in these posts:

Update on social media project in Wisbech

Fenland social media project

Digital engagement framework adapted for local government

Digital engagement governance

And you can also visit the project blog or take a look at a presentation I did for the Local by Social online conference last November.


Going undercover

Do you sometimes feel like you are the only one who thinks it’s important to find out how your customers use your organisation’s website? Have you ever read user experience design books where you think “yep that all sounds great, if you’ve got loads of time and a big budget to boot, but how am I going to get agreement from stakeholders to spend time and money on that?”

If you feel like you’re struggling to do user experience design *properly* then you should probably read Undercover User Experience Design by Cennydd Bowles and James Box, who both work at the well respected Clearleft, published within the New Riders Voices that Matter series. It’s an easy, enjoyable read and if you’ve ever struggled with introducing user experience design methods or culture into your organisation, you’ll get loads out of this book.

The authors provide clear descriptions of a number of UX methods and deliverables and demonstrate how they can be used in context to better understand your users and design for them. But they also tackle organisational culture and how to work with other stakeholders, including project team members and senior managers. And no matter how good your research or designs are, you won’t succeed if you can’t work collaboratively and influence your colleagues or clients.

I’ve worked in the private and public sector, agency side and in-house. I’ve always worn multiple hats in project teams. I’ve been the project manager, the developer, the user researcher, the information architect, the user experience designer and the product owner. I found this book really pragmatic because Bowles and Fox don’t assume you are a UX consultant working agency side (aka an ‘outie’), but also provide advice for those working in-house (an ‘innie’).

What they really drive home is this: if you’re passionate about user experience and you care enough to try to make a difference, you can do. By going undercover, being disruptive and getting results.

On the companion website to the book they have re-produced the book’s manifesto from chapter one, which reminds me of the Agile manifesto. I particularly like the statement ” UX is a mindset, not a process—it lasts all the way until the site is live, and after”.

“We believe in going undercover. We don’t mean you should skulk around in the dark. As an undercover user experience designer, your mission is to get people excited about UX without them realizing what you’ve done. Unless you’re an expensive consultant or a senior manager, you won’t do this by knocking on the CEO’s door and demanding change. User experience design is disruptive. It asks difficult questions. Good-enough managers in good-enough companies don’t want you to rock the boat; they’re busy worrying about meeting next month’s targets.

We believe in introducing UX from the ground up. Sneak UX into your daily work, prove its value, and spread the message. Results are more persuasive than plans.

We believe change comes through small victories. Putting users at the heart of a business is a huge cultural change. It takes years. But you’ll be surprised what you can achieve with focus, patience, and persistence.

We believe in delivery, not deliverables. Some people practice user-scented design, not user-centered design. They churn out documents—sitemaps, wireframes, specifications—but they’re not interested in what happens next. UX is a mindset, not a process—it lasts all the way until the site is live, and after.

We believe good design today is better than great design next year. There’s no such thing as perfection in design, particularly on a medium as fluid as the web. You’re not here to impress other designers; your job is to make your users’ lives better.

We believe in working with people, not against them. Just as we empathize with users, we must respect and understand our colleagues. We reject elitism and accept that compromise is healthy. Passion is fine; zealotry is not.

We believe in action, not words. Introducing UX into your company is a lot of work. No one will do it for you, so you’d better get cracking. Remember, it’s often easier to get forgiveness than permission.”

If you’re a seasoned UX professional then you may feel you can do all this and more standing on your head. But I would highly recommend Undercover User Experience Design as a good read nonetheless. And for those ‘innies’ among us, particularly those who often feel like a lone voice championing the user, this may well be the book for you and your colleagues. I’ve already lent my copy to two people in my team!


Using personas for web and service design

I’m a fan of personas. In the user experience (UX) field, personas are fictional profiles of your users based on research data. Personas can bring your users to life and help guide the design process. Giving your personas names, pictures, personal profiles and using believable narratives will help everyone involved in a project to empathise with user goals, behaviours and motivations in a very tangible way.

Personas have been used in web UX design for a number of years. Alan Cooper has long promoted the use of personas as part of his goal directed design method. I have recently been reading an excellent book by Steve Mulder on personas. Turning research data into personas can be overwhelming and I’d recommend Steve’s book if you want a really good understanding of how to develop and use actionable personas.

Examples of personas
Photo by CannedTuna on Flickr

Continue reading