- The thing that makes user research unique
- How to run a user journey mapping workshop
- Getting Hired in UX
- Using decision tables to support contextual targeting in your UI.
- UX Brighton 2014: 10% off!
- Some Dark Patterns now illegal in UK – interview with Heather Burns
- So you want to be a UX freelancer?
- User Experience and Jobs To be Done
- Create your own Mac-based usability testing lab with viewing room
- Top posts of 2013
- Dark Patterns and Sports Direct on BBC Watchdog
- UX Brighton 2013: the psychological foundations of design (10% off!)
- Deadly set: how too much focus causes mistakes.
- The slippery slope
- Are you in a teflon-coated UX role?
- Sneaking responsive in under the radar using an mdot site.
- The drunkard’s search
- Powwowapp: for scheduling research
- Assumptive Personas
- Moving on.
There’s one thing that always takes businesses a while to understand about user research – a faulty way of thinking that needs to be overcome for them to deliver good design. In traditional businesses, reports and documentation are expected to depict a competent team in total control, heading in a dignified and controlled direction according to the agreed company strategy, with little graphs that go up and to the right. The thing about user research is that it delivers bad news every time. It depicts reality - the fact that you don’t control how your customers think, that they see things differently and that changes are needed if you’re ever going to please them. It doesn't make you look good. That’s the whole point. If you’ve been involved in user research that shredded all your carefully-made designs and delivered a gut punch that left you reeling for 24 hours, you should be proud. That’s the best kind of user research. Don't be ashamed – bad news is opportunity knocking. If you’ve been involved in user research that didn’t hurt at all, then I guarantee you did it wrong. A researcher who is keen to please the design team is useless. User research is unique because it’s about seeking out the bad news. It’s about finding evidence that can flip your position on firmly held beliefs – the wording on that button, the structure of that page, the order of that journey, or even nature of the proposition. If you want to start getting good news from user research, you just have to keep doing it. It’s like going to the gym. Of course the first visit is horrific. The value comes later, after many repeated sessions. If you want to make progress with user research, you need to get comfortable with being uncomfortable.
I’ve been helping some of my clients with hiring UX designers lately. Here are some tips you might find helpful if you're looking for a role. BEWARE MISCOMMUNICATIONS FROM RECRUITERS If you’ve been lined up the role by a recruiter, be cautious if they make you feel like you’ve been headhunted or that you’re somehow a special candidate. Some recruiters try to encourage you to apply by flattering you, and then they they do the same thing to the employer – telling them how incredibly enthusiastic you are about the role. When the interview happens, both parties can feel weirded out by the lack of enthusiasm on both sides. More than anything else, recruiters want the hire to be successful so they get their payday. This means they sometimes bend the truth a little. If a recruiter ever gets in touch and it turns out you have a direct link with the hiring company, tell the recruiter you don't need an introduction. It'll avoid problems like this and since there's no longer a middleman wanting a cut, you might end up with a higher salary. IN THE INTERVIEW, TELL YOUR DETECTIVE WORK STORIES It's easy to learn how to preach a few design tropes and to trot out the steps of a good design process. The problem with interviewing for UX roles is that anyone can do that. The interviewer has to weed out bullshitters and passengers who have coasted through good projects on the energy of their team members. User experience is not about following process or preaching. It’s about finding the right problems to work on and taking them apart. It’s a journey of exploration and discovery – finding your way past wrong turns, overcoming hurdles and making sense of a complex world. Your challenge is to communicate your this during an interview. Try thinking of your projects as detective stories. Most design projects have aspects that fit into a good story arc. For example, there’s usually a point at which your eyes are opened by unexpected user research findings, or where the project gets into trouble and you have to adapt your approach to get out of it. This is the juicy stuff that the interviewer will enjoy hearing about. If you can't tell these sorts of stories, the interviewer will assume that other people dealt with this stuff and you were acting in a junior capacity - doing what you were told but never really grappling with any big challenges. Prepare for interviews by piecing together true narratives for each of the big projects you’ve done recently. Writing and public speaking really helps because it forces you to think in this way. If you can explain your work using good narratives, this tells the interviewer how you think, which is far more important than how pretty your portfolio is. Narratives are also useful because they're easy to remember and retell, even when under stress. Exactly what you need in an interview. RED FLAGS There are certain responses that I always see as bad sign:
* When asked about user research findings, can you describe at least one surprising finding that led to an improvement in the product? Red flag: if you talk about user preferences (what they said they liked) rather than user behaviours. If you don't have much hands-on research experience, don't panic – the interviewer will already know this from your resume, but you'll still need to show that you're used to attending the sessions, feeling the pain and deliberating with the team about how to action the insights. * When asked about a design decision, can you describe the trade-offs you considered? All design involves considering alternatives approaches and weighing up their strengths and weaknesses. Red flag: if you can't answer probing questions about design decisions, and defer to authority (e.g. "the client wanted it like that") or other externalities like legacy systems. * When asked about a big project, can you name which key problems were left unsolved at the end? Red flag: if you can't name any. Design is never finished. The company learns, time passes, competitors pop up, users mature and the market moves on. You need to be able to talk about what you wanted to do next and why.DON'T BE SURPRISED IF YOU'RE ASKED TO DO A DESIGN EXERCISE IN THE 2ND OR 3RD ROUND Being asked to do a design exercise is a good sign. It's time consuming for the employer to run these sessions so it means they think you're worth it. Some interviewers will email you the materials a few days in advance and then have you present your work in the interview. I personally don't do this - it gives an unfair advantage to people who are able to take time off work and put some serious hours in. But if you get this opportunity as an interviewee, you'd be stupid to not take full advantage of it. Even though it's more stressful, it's fairer to be given the brief on the spot in the interview. The type of design exercise will depend on the client, but it'll usually be a microcosm of a UX project, packed into an hour or two. Some planning, some evaluation, a little whiteboard design sketching, discussion of research options, that sort of thing. If you've not done this before then practice. Practice a lot.
Historically, marketers have given targeting a bad name, with creepy email campaigns and annoying ads that follow you around – but the fact is, targeted messages can deliver a lovely experience if they’re done right. An ideal restaurant is where the concierge knows your name and sits you at your favourite table. That’s good targeting. There’s a big difference between this and having them say “You almost bought an expensive wine with your meal last time. How about it today?”. Let’s work through an example. Let's say we’re designing a mobile app for runners (Imagine we’re the first one on the market to keep things simple). There are certain core activities that users will really care about - in this case tracking running performance and comparing performance with others. Thats pretty much it. However, there are lots of other activities that our business will want users to do, but the users are not yet particularly interested in. For example, if a user is just starting out, they’re probably not going to care much about connecting a heart rate monitor - they’d be crazy to go spend $200 before they’ve even finished their first 5k run. If you run a workshop with the business stakeholders, you'll quickly come up with a long list of actions like this:
* Complete a profile * Connect with friends * Encourage users to engage with friends on social media * Encourage users to publish updates * Make in-app purchases / buy a subscription * Be aware of credits running out (if using a credit-based model) * Connect associated products (heart rate monitors, fitbits, etc) * Review our app * Respond to usage tips (e.g. “You haven’t tried doing a group challenge yet…”) * Use our referral scheme * Buy our branded clothing * And so on…In this list you’ll notice a few things that deliver value to the business but not to the end user. For example, asking users to "complete a profile" is pretty meaningless on its own. Nobody really wants to buy your branded clothing. The action needs to be linked to something that delivers them a clear benefit. This means you’ll need to work on your list to ensure each item answers the users' question of "What's in it for me?". One of my clients refers to their list as the "ladder of engagement". For them it's a short, linear sequence of actions they want users to complete one-by-one (Dropbox uses a similar approach). Other clients have a complex mix of actions and contexts - which can be hard to rationalise into a coherent system. From a user’s perspective, this list is inherently dull and unappealing. Even if you put it in front their noses, they’re not going to want to pick through each item one by one – there’s simply too much noise and too little value. What we need is some way to invite relevant actions from a given user, at a time when they’re in the right frame of mind (in sales terminology, when they’re a “hot prospect”). In doing this we need to make sure the user never feels nagged, annoyed or burned out. This means we need a system to orchestrate the contextual targeting. Luckily, from a development perspective it’s not that complex - all we need to do is create a decision table that maps contextual triggers against appropriate actions. Let’s start with some terminology. Firstly, there’s the UI element in which we encourage users to take an action – let’s call this the “call to action”. Then, there are the locations in the UI where you want these calls-to-action to appear. Let’s call these “slots”. For this exercise let’s assume we’ve got just one slot - a modal dialog that appears when the user successfully completes a run. Post-run modal dialogs in Runkeeper and Strava Of course there are lots of other types of slots you’ll want to consider in your own work, for example:
* INFOBAR: a strip along the top or bottom of your UI that provides information but doesn’t force the user to interact with it (See Google Chrome’s ux docs). * DASHBOARD HERO SLOT: a prominent slot on the dashboard or home screen * ACTIVITY FEED SLOT: if your app focuses on an activity feed (similar to facebook / pinterest), it makes sense to interleave the call-to-action slots with other content. * OS NOTIFICATION: these are powerful but can be overkill. Remember - an OS notification interrupts the user as they go about their normal life and should be reserved for something of genuine importance.Finally there’s the triggering context: each call-to-action needs its own trigger. For example, if the call-to-action is “Connect with friends” then the trigger would be “if they haven’t yet connected Facebook or Twitter”. Before we draw up our decision table, there’s some additional stuff we need to consider:
* WHAT HAPPENS IF THE USER TAPS “NO THANKS” AND DISMISSES OUR DIALOG? We need to be able to mute it for a period, then unmute it later. In other words, we need to define a “snooze” behaviour. * HOW MANY TIMES DO WE WANT TO REPEAT THE SAME DIALOG? if a user taps the close button a few times over, it’s pretty clear they don’t ever want to see that dialog again. Let's say we only want to nag them 3 times on any given action (which means we need to implement a snooze counter for each action). * HOW CAN WE LIMIT THE RATE OF DIALOGS? Pacing is important to avoid user fatigue and annoyance. We need a rate limiter for the system as a whole. Once a dialog has been actioned or snoozed, the system should shut up for a period to give the user some breathing space (e.g. “only show a dialog after every 3rd run”).With all that in mind, here’s the decision table, depicted below. It runs from top to bottom, like a series of if/then/else statements. For example: the first row is evaluated. If the trigger statement is true and the action is not snoozing then the dialog is shown. If those criteria are not met, the second row is evaluated, and so on. When you get to the bottom of the table you loop back around to the top again. This means there is a order to the list - for example, the user will see “find friends” before “turn on notifications”. This gives you a way to express your priorities. Of course there are smarter ways to do this. Microsoft Clippy was originally powered by a bayesian network, and that was back in the 90s. The reason for using a decision table here is because of its simplicity. We're not designing Google Now - we're designing an ancillary feature of an app that has other purposes. It’s important to keep the UX manageable, clear and easy to test. To conclude - I hope I’ve shown that if you’re going to do any contextual targeting in your products, you need a systematic approach that ensures users don’t feel nagged. If you reach for a prototying tool like framer.js, Origami or Axure, you're missing the point. You need to map out your triggers and actions. If you're a UX specialist, this is your domain. Nobody else undestands your users and their real-life contexts the way you do. Given that the Apple Watch and other wearables are being released this year, we all need to get good at designing UIs which show the users the right thing at the right time. Hopefully this article gives you a starting point to think about this stuff.
Good news everyone - I've secured a 10% off discount code for UX Brighton 2014. Enter "90percent" at the checkout. This code works for all ticket types - if you use it now you'll get an early bird ticket for £116.10+vat (Standard price is £149+vat). Maybe I'll see you there? Buy your ticket now ›
In this article I interview Heather Burns, author of The Web Designer's Guide to the Consumer Rights Directive. SO, WHICH DARK PATTERNS ARE NOW ILLEGAL IN THE UK? The EU's new consumer rights law bans certain dark patterns related to e-commerce across Europe. The “sneak into basket” pattern is now illegal. Full stop, end of story. You cannot create a situation where additional items and services are added by default. No more having to manually remove insurance from your basket when purchasing plane tickets. Hidden costs are now illegal, whether that’s an undeclared subscription, extra shipping charges, or extra items. While the costs are still permissible, failing to advise the customer about them or explain what they are is not. Everything has to be brought out in the open, explained, and clarified before checkout. Even if you are not able to declare a specific additional cost in advance - say, supplemental shipping charges to remote areas - you still have to declare that these charges exist and will be applied to the order. As a part of that, retailer fees and surcharges must be brought out into the open and explained. Retailers can no longer charge “processing fees” in excess of what it actually costs them. Remember when a certain airline used to offer £2 return flights which carried a £45 credit card processing fee? Now, we all knew damn well that the flight was £45 and the processing fee was £2, but there was nothing we could do about it. With the new law they cannot try to swap the figures or surprise you with a £45 “processing fee” at checkout. Forced continuity, when imposed on the user as a form of bait-and-switch, has been banned. Just the other day a web designer mentioned to me that he had only just discovered he had been charged for four years of annual membership dues in a “theme club”, having bought what he thought was a one-off theme. Since he lives in Europe, he may be able to claim all of this money back. All he needs to do is prove that the website did not inform him that the purchase included a membership with recurring payments. WHAT UK LAWS HAVE CHANGED? This law updated and replaced the 1997 consumer rights law, which was laughably outdated. It’s pretty amazing to think that until June this year, digital products and downloads had no reference in trading laws, which meant that consumers had no protection. The new law essentially had three goals. The first was to update those ridiculously old e-commerce laws. It’s a damning indictment of all the UK governments and parties who have held power since 1997 that it took the EU to force us to bring our trading laws out of the Teletext era. The second goal was to harmonise consumer trading laws across all of Europe so that people can do more cross-border shopping. Pour exemple, I love French pop music. If I decide to spoil myself with a bumper order from Paris, I can now do so knowing that I am buying under the same conditions and protections as if I’d gone to my local Fopp. The third goal, and the one you’re concerned with, was to outlaw e-commerce’s worst Dark Patterns. There’s clearly been a lot of good public input into this law. We’re really not used to seeing web laws that deal with real specifics rather than theoretical concepts. ARE SOME DARK PATTERNS STILL LEGAL? The directive only dealt with Dark Patterns concerning e-commerce. Dark Patterns concerning other issues like privacy, information disclosure, sharing and advertising are not affected. We also have yet to see what new Dark Patterns will be invented in response to the Directive! HOW COME SOME E-COMMERCE SITES ARE STILL USING THE SNEAK INTO BASKET DARK PATTERN? ARE THEY BREAKING THE LAW? The law has not been well publicised. Lack of knowledge, of course, is no excuse. In my book I talked about “trading trolls” - people who would surf the web specifically looking for noncompliant sites so that they can place an order, get the stuff, report the site for noncompliance, get their money back, and keep the stuff. After all, if the site is breaking the law, they have no recourse there. I would, of course, never encourage anyone to do that. _*coughs loudly, and winks*_ LET’S GET SPECIFIC - IS SPORTSDIRECT.COM NOW BREAKING THE LAW? At the time of writing, sportsdirect.com sneaks a £1 magazine and mug into your basket with every purchase. As that is adding those items by default, thereby forcing the customer to manually remove them, it is noncompliant. They cannot argue that a magazine and a mug are companion pieces to the items being purchased. They are extra items, full stop. If the magazine is so essential they can simply include it in your shipping parcel like many retailers do. As for a mug, the process of removing one from your basket treats us like one. HOW ABOUT NEXT.CO.UK? Next is stretching the law to its limits. Technically this is legal because it meets the information provision requirements. As with anything, though, just because it's legal doesn't make it right. The Next Directory is not a paper catalogue, it’s a credit programme and a financial service. (I learned this when I signed up for the directory, never received a copy nor the invoice, and then received a late fee notice and a mark against me on my credit record for not paying for a stupid catalogue which I never received.) Financial services are exempt under the Directive, and so next.co.uk are deliberately being as ambiguous as possible because they can. AND WHAT ABOUT RYANAIR? It’s a fair bet to say that the company whose conduct led to this law being created in the first place is going to throw out quite a few examples of noncompliant conduct. In this example, they are still adding the additional payment by default, leaving the consumer to manually opt-out of it. That's wrong. WHAT’S GOING TO HAPPEN TO BUSINESSES WHO USE THESE NOW-ILLEGAL DARK PATTERNS? Quite simply, businesses who don’t comply face a loss of revenue. If you make a purchase, whether that’s buying goods or a service, on a non-compliant web site, you have the right to recourse through your nearest Trading Standards office, in other words, your local Council. Unlike the cookie law, which is dealt with by one UK-wide bureaucracy which has bigger fish to fry, this law is dealt with on a local level. A failure to comply cancels the transaction. You can get your money back and keep the goods. If the sale was for a service or a digital download, the contract is cancelled and no further payments are due. SO THIS ISN’T A RE-RUN OF THE COOKIE LAW FARCE WE HAD A FEW YEARS AGO? Absolutely not. The cookie law was the wrong law, drafted at the wrong time, in the wrong way, for the wrong reasons. The Consumer Rights Directive couldn’t be more different. It was desperately needed, it’s common sense, and it reflects the way the web actually works. HOW DOES THE CONSUMER RIGHTS DIRECTIVE AFFECT US COMPANIES WHO ARE DEALING WITH UK OR EU CONSUMERS? The Directive applies to inter-EU sales only. A US company does not need to comply to sell in Europe. Although it would be awfully nice of them. AND FINALLY, TELL US ABOUT YOUR BOOK! My book does what it says on the tin - The Web Designer's Guide to the Consumer Rights Directive. It started out as a blog post, but 11,000 words later I realised it was a bit more than that! Writing the book was my attempt to bring sanity back to the web community. I've been researching and writing about the cookie law since 2012 and it's taught me a lot about the gap between practice and theory. One of the many things I came to realise is that laws concerning the craft of web development are cooked up by offline politicians and then drafted by solicitors for solicitors. They throw an 80 page legalese .pdf onto a web site and say "there's your lot, now comply." They don't think about who actually does the work, and because they are neither coders nor crafters, they very often literally have no idea what they are talking about. The web community needed someone to translate these laws into plain English, break them down into small chunks, and explain how to comply in terms of front-end and back-end implementations, not airy legal theory. And if aspects of the laws are ill-informed or disruptive, we have a responsibility as a community to speak out. BUY THE BOOK ›
People sometimes ask me for advice about getting into User Experience freelancing. Is it enjoyable? Is it worth it? Is it good money? The fact is, the answer to all these things entirely depends on you and the way you choose to run your business. Let me explain… ARE YOU EXPERIENCED? To be a UX practitioner as an employee, you only need to be good at UX. To do it as a freelancer, you need more skills. Most importantly, you also need to be able to cope with difficult business situations without panicking. What do you do when a client tries to get you do work you don’t agree with? How do you deal with a late payer? How do you explain to the boss of your client that they’re wrong? Mike Monteiro's book Design is a Job is a nice litmus test. If you read it and lots of the points are obvious to you then it’s a good sign. KNOW WHY YOU WANT TO GO FREELANCE Everyone has different motivations and it’s important that you understand what you want out of it so you make the right decisions. For example, when I went freelance we’d just had our second baby, which meant I had my whole family dependent on my income. I kept telling myself that money was my main motivation, so when I was offered a long term contracting role at an investment bank, I jumped at the chance. It turned out to be the most hostile environment I've ever worked in, and I quit within a few days. As a counterpoint, one of my current clients is a small charity fundraising startup. They’re awesome and I couldn’t be happier working with them. My point here is that if you’re clear and honest with yourself about your motivations, you will make good decisions and end up happier. Write a rule-book for yourself about the types of work you do and don’t want to do, then stick to it. HOW BROAD IS YOUR SKILL-SET? If you’re very specialised - say you only do qualitative user research or you only make front end prototypes, then you’re closing doors for yourself as a UX freelancer. I see research and design as the yin and yang of UX - you really need a balance of both types of skill. The broader your skill-set, the wider you can cast your net for work and the longer your contracts are likely to be. Being a niche specialist certainly does help you stand out from the crowd, but it’s really useful to be able to turn your hand to anything when necessary. If you’re working as an employee now and considering going freelance, I’d advise you to volunteer to work on projects that are outside of your comfort zone. Focus on your weaknesses because once you go solo, you’ll have nobody to support you. KNOWING WHAT KIND OF INCOME YOU NEED Before you start, you need to have an idea of what your income goal is. If you look on any UX recruiter’s website, you’ll see that UX freelancers tend to earn between £400-500 a day in London. Don’t make the mistake of multiplying it by the number of working days in a year - you won’t be wearing a top hat and monocle just yet. You’re better off starting by thinking of your current income and working out how many days a year you’d need to freelance to hit that. Speak to an accountant, as you may find you’ll pay less tax as a freelancer than as an employee (this is the usually case in the UK if you start a limited company). Your income as a freelancer is primarily defined by your utilisation rate (number of days worked in a period / number of working days in that period). In some agencies I’ve worked, it’s been normal for some billed staff to have utilisation as low as 50%. This isn’t really a problem since agencies tend to charge at least double a freelancer’s rate, and when staff are “on the bench” (between projects) they can turn their hand to other areas of the business like proposal writing, pitching, event organising, blogging, helping out on other projects, etc. Agencies are set up to accommodate bench time, and being on the bench is a pleasant experience. When you’re freelancing, the situation is a stark contrast. 50% utilisation can be very stressful. Your cash-flow becomes an obsession. When I worked agency-side I used to write a lot of proposals and do a lot of free pre-sales consultancy. I know a few freelancers who have carried on doing this after making the break to freelance. Beware of this mistake. You’ve no longer got the cushion of an agency’s day rate to behave like this. Sales work is vital for you, but you need to recognise the difference between sales and free consultancy or spec work. PREPARING BEFORE YOU MADE THE BREAK I was an academic researcher for many years at the beginning of my career, so I’ve always been into knowledge sharing - writing articles, public speaking, and so on. This has helped with my visibility, but it’s easy to over-estimate the value. A few hit articles on Hacker News don’t translate into clients queuing up. One article I wrote last year on Dark Patterns got 50,000 uniques overnight. Not one genuine sales enquiry occurred as a result. It’s worthwhile doing all that stuff, but it doesn’t solve your sales problem for you. What helped me more was my work history – not in terms of my résumé but the personal relationships it gave me. When I started out I made a list of all the client names I’d worked for in the last 10 years. It wasn’t actually that long – about 50 individuals. Then I started thinking about all the other people I’ve crossed paths with during that work. All the project stakeholders, and all the junior and mid weight people on the teams. The nice thing about time passing is that people spread out into other organisations and your peers take on senior roles. If you made an effort to build a positive relationship with them, they’re much more likely to hire you. My fundamental point is that you shouldn’t always chase after the people who are currently holding the purse strings. This may sound obvious but it’s important to maintain relationships with the people you like from your prior clients' organisations and make sure you meet up with them socially every month or so. One honest friendship is worth a hundred sales meetings. KNOW WHERE YOUR NEW BUSINESS WILL COME FROM If you haven’t got a pipeline of work coming in from direct clients, then you’ll probably end up using a recruitment agency. This isn’t necessarily a bad thing, but recruiters have their downsides. Firstly, they’re pocketing about 20% of the day rate they’re charging the client, which would otherwise be going to you. Secondly, they have a habit of bending the truth to you about the project and to the client about your skills. The most common horror story about recruiters is that you get mis-sold a project. That wonderful blue-sky vision project turns out to be a wireframe production line role. If you go to a recruiter and they line you up with a design agency, you now have 2 layers of middlemen sitting in between you and the client. Most London UX agencies charge their clients around £1000 a day. The client sees you as a very expensive resource and if they’re nervous, they’re going to quibble over every hour you bill. This means it can be quite high pressure and stressful in comparison to working for the very same client in a direct relationship. All that said, there are lots of good reasons to freelance in design agencies. The agency might have an awesome team you get on with, skills and methods that you want to acquire or great projects that you want to be involved with. To sum up – going freelance is not an easy decision to make. I hope at least now I've given you a few things to think about while you're making up your mind.
In the field of UX, we’re all pretty familiar with the concept of behavioural personas, but not everyone is aware of the parallels between this and Clayton Christensen’s "Jobs To Be Done" theory which became famous in his 1997 book The Innovator’s Dilemma. He explains it quite nicely in this Press Publish interview:
Here I am. I have characteristics that slot me into demographic segments. I just turned 60. I’m 6 feet 8. We just sent our youngest daughter off to Columbia. I have all kinds of characteristics. But none of these characteristics or attributes have yet caused me to go out and, say, buy the New York Times today. There might be a correlation between particular characteristics and the propensity that I will by the New York Times – but they don’t cause me to buy it. What causes us to buy something is that a job arises that we need to get done, and we buy or hire a product and pull it into our lives to get that job done. What’s important is that understanding the customer (as a set of demographic attributes) is the wrong unit of analysis. It’s the job that we need to understand, because the job itself is very stable over time. If you keep focusing on the job, then you can weather through the ebbs and flows of technology as they come into your industry. I’ll give you an example. Let's say 'I need to get this from here to there with perfect certainty, as fast as possible'. We all find ourselves needing to get this job done on occasion. Actually Julius Caesar had the same job to do, and the only thing he could do to get the job done was hire a horseman and chariot. Now we have Fedex, but the job itself is just the same. The way you get the job done has changed over time. If you define your business by technology or customers (as demographics), you can get blown out of the water with regularity. And so Western Union would be hired to do this job during the time of Abraham Lincoln. They framed the business as 'long range telegraph', and so when the telephone came, Western Union was just blown out of the water. But if you focus on the job, then when new technologies come along you can look at them and say say “holy cow, that would let me do the job even better, so you buy into it in a way that keeps the enterprise going." – Clay Christensen, Feb 2013 (paraphrased and brackets added for clarity)To elaborate, the Jobs To Be Done theory basically says that any company which describes itself as “technology X for demographic segment Y” is eventually doomed. If you tie yourself too closely to a technology, you'll get blindsided by the next big thing. Similarly, if you define your customers by their demographic attributes, you learn very little about how you can help them better. On the other hand, if you get under the lid of the jobs they are trying to do, and understand their Psychology – i.e. their needs, goals and mental models around these jobs – then you’ve defined the problem in a way that brings the solution into focus, and it isn't tied to a particular technology or approach. This is what we're trained to do in the field of UX, which is why we're so well placed to have a seat at the product strategy table. As a UX practitioner, it's easy to let yourself become pigeonholed as someone who just provides a tactical service like wireframing or usability testing. This is disempowering - you should think bigger, even if the people around you aren't. A theory like Jobs To Be Done can help you bridge the gap from tactical to strategic thinking, and it can help you reframe the way you talk about your role at work.
There’s no one-size-fits-all solution to usability testing labs. The UX team at The Telegraph is Mac-based, which puts Techsmith's Morae out of the question. Telestream’s Wirecast would be a decent Mac alternative, but it was really unreliable on the MacBooks I tested it on, putting it out of the picture. After some experimentation, I ended up breaking the problem into two parts: (i) recording and (ii) transmission into a viewing room. For the recording part of the problem, I ended up going for Screenflow, Reflector and an IPEVO Ziggi-HD USB Document Camera. Screenflow serves to record the screen, in-built webcam and audio from any Macbook. It also runs happily alongside Reflector, so we can wirelessly mirror an iOS device on the Macbook while recording with Screenflow (pictured below). Finally, if we’re testing using a low power device like a Kindle, or if we want to record our participant doing something non-digital (e.g. paper prototypes or sketching exercises) then we can use the Ziggi-HD document camera. It's basically just a webcam, but it has fixed focus so the camera doesn’t keep refocusing on the participant’s hands as they do stuff. It's also mounted on an anglepoise-type stand, making it easy to point at your participant’s device. The bundled app displays the camera's footage on your desktop, so Screenflow can be used to record it along with the Macbook's inbuilt webcam. I'm probably going to buy a second Ziggi-HD, remove it from the stand and fix it to a Mr Tappy (or a similar rig) which would allow the participant to move the device around as much as they like while always keeping it in frame. Moving onto the viewing room part of the problem: VNC, logmein and other screen sharing tools aren’t particularly suitable for this as they can be quite flakey and give a fuzzy picture. This is forgivable if your participant is in a remote location, but it’s silly if they’re just down the corridor. Plus, the last thing a researcher wants is to be interrupted mid-session to be told that the viewing room connection has dropped out. Connecting the viewing room directly to the testing room by dedicated cables seemed like the way to go, but I couldn’t find any gaps under doors or dividers that would allow us to do this. In the end I realised that we had a raised floor, so we could actually run the cables underneath the floor tiles. This is a really useful tip - if your office floor has those flapped boxes in the floor then you probably have a raised floor too. We had our buildings managers take up the floor tiles and put long HDMI cables between two meeting rooms, leaving the terminals poking up out of the floor boxes (If it's a long distance between the rooms, it's cheaper to use a CAT-5 HDMI extender). Rather than trying to do picture-in-picture, we just decided to have two TVs in the viewing room (see above). One of the TVs is used to mirror the test macbook’s screen (above right), the other is used to display footage of the participant’s face, and play the audio of the interview (above left). After looking at a few camcorders, I ended up taking a punt and ordering a GoPro 3 Silver. It turned out to be a good choice - the quality of the picture is amazing, and the wide angle lens means you get footage of the user’s head, their torso, the device they’re holding, and even the interviewer. It’s worth being aware that the cool-sounding WiFi feature of the GoPro is rubbish for remote viewing. It’s laggy, blury and it can’t be used as the same time as HDMI-out. I'd advise keeping the WiFi turned off and just connecting the device via the micro HDMI port on the side of the camera. It’s surprisingly awkward to mount a go-pro onto a tripod: you’ll need to buy a tripod mount and a frame, then you’ll need to use a small hacksaw to cut a large enough opening for both the HDMI and USB power cables. Still, it doesn't look too bad when you're done and since the camera is so small, participants quickly forget it's there. So there you have it: a simple and reliable Mac-based usability testing lab that can be used for desktop, mobile and pretty much anything else you can throw at it. - INTERESTED IN WORKING IN THE TELEGRAPH'S NEW UX TEAM? I'm consulting there at the moment. Lovely team, great brand and lots of exciting new things happening right now. Why not pre-empt the job ads – send your CV & portfolio to Tim Goodchild.
2013 has been an intense year for me, having left Clearleft in April to start my own business as an independent UX consultant. I've got to say, I've been loving every minute of it. When you're running your own business the risk makes everything so much more interesting and exciting. Pitches are more exhilarating, debriefs are more satifying – even the admin work like running your CRM and doing your book-keeping has a certain edge to it. The downside is that I haven't had much time to blog, something I plan to set right next year. Here's a round up of the most popular posts from 2013: THE SLIPPERY SLOPE. Just as I thought people were getting fed up with articles on Dark Patterns, I got over 50,000 uniques on this article within 24 hours of posting it. If you're interested in this subject, do get in touch. We've got a team of 5 people working on darkpatterns.org now - myself, Marc Miquel, Jeremy Rosenberg, Joseph Dollar-Smirnov and James Ofer, and we're looking to grow it this year. It's a loosely bound collective where we all throw in a few hours here and there to update the site and its content. THE DRUNKARD’S SEARCH. People often only look for what they are searching by looking where it is easiest. Are you making these mistakes in your own work? ARE YOU IN A TEFLON-COATED UX ROLE? This article is dedicated to all the bad UX designers out there who get away with doing very little. COMBINING REFLECTOR AND SILVERBACK FOR IOS USABILITY TESTING. This is my standard operating procedure for iOS usability testing now. Reflector works well alongside Sliverback, Screenflow and many other screen recorder apps. SNEAKING RESPONSIVE IN UNDER THE RADAR USING AN MDOT SITE. Since writing this article, I learned that this exact strategy has been used on BBC News and The Guardian. Nice to know. SUBMITTING A TALK PROPOSAL TO A UX CONFERENCE THIS YEAR? READ THIS. A few handy tips if you're trying to get a speaking slot at a UX conference. IS USER-CENTERED DESIGN BROKEN – OR IS IT JUST US? Fashions come and go, but the value of involving users in your design process will always stick around. DEADLY SET: HOW TOO MUCH FOCUS CAUSES MISTAKES. People often don't talk about the risk of being "too focussed" in our work. It's actually a documented problem that human factors researchers have to account for when analysing accidents. Thanks for being a reader in 2013. If you haven't already, why not subscribe to new posts by email?
This is probably of greatest interest to UK readers, but the sneak into basket dark pattern was featured on BBC Watchdog last week. This dark pattern is going to become illegal in the UK next summer, as our implementation of the EU Consumer Rights Directive (2011) finally comes into action. Sports Direct, National Express, Skype and Monarch were all singled out as offenders. Sports Direct sneak a mug and/or magazine into your basket when you checkout. Their response was "Following feedback from our customers, we have simplified the process to remove the magazine from an online basket." Er, here's a suggestion - how about you stop sneaking stuff into the basket in the first place? Let's take a look at how it works today: try ordering something from sportsdirect.com. When you add something to your cart, you'll see as clear as day that your selected item is the only thing in there. Proceed to checkout and, if you have your wits about you, you'll see something suddenly has been added to your basket: To remove it, you need to click "back to bag", where the sneaked-in item will appear - even though it didn't before. Finally you can remove it. Phew. If this is simplified, I can barely imagine how complex it must have been before. Sports Direct go onto elaborate: "Many of our customers are pleased to receive the magazine with its sporting and fitness tips, celebrity interviews and product reviews and advice". It just goes to show that if a big player like them has the audacity to keep doing this even after being named and shamed on national TV, then the change in the law is not coming a moment too soon.
Design is a branch of applied psychology. If you don't know the first thing about the psychological foundations of design, then you're not likely to be a good designer. I'm going to UX Brighton on November 1st, and you should too. I've negotiated readers of this blog a nice deal: enter the promo code 90percent at checkout for 10% off »
Aviation Psychologist David Beaty on the phenomenon of 'Set' (1991):
'Set' is a survival characteristic we have inherited. The human brain evolved to help individuals live and survive circumstances very different from our own. It predisposes us to select our focus on that part of the picture paramount at the time – a vision often so totally focused that it ignores the rest of the environment. Once something is identified […] it takes on a reality of its own and sticks in the mind like a burr which is difficult to dislodge. […] The mind becomes tunnelled on a particular course of action. Add to that the ingredient of fatigue and it is not difficult to see that a 'set' as hard as concrete can result. Furthermore, 'set' is infectious. There is a follow-my-leader syndrome. So it is easy to see why most aircraft accidents are caused by 'silly' mistakes in the approach and landing phase. […] 'Set' has been a factor in many aircraft accidents. [In a case in 1972 over Florida], the crew of a Tristar were not sure that their undercarriage was down. The accident sequence was begun by a burned out light bulb in the system which is designed to show that the undercarriage is down and locked. […] the crew examined every possible of finding the trouble. The flight engineer crawled down into the nose, while the captain and the first officer tried every combination of switches and circuit-breakers. […] the three members of crew did not notice that the autopilot had become disengaged and the aircraft was sinking […] eventually crashing into the Everglades. Because they had become preoccupied with an unsafe landing-gear indication, they failed to monitor the critical altimeter readings. Ironically, the air traffic controller noticed on his radar that the aircraft was losing height, but instead of pointing this out simply asked diplomatically "How are things coming along there?". The crew, still obsessed with their landing-gear problem, assuming he referred to that, for they could thing of nothing else, replied seconds before the crash, "Everything is all right!"– Paraphrased from The Naked Pilot by David Beaty (Ch.6) Aircraft accidents make grisly reading, but they are one of the most accurately documented and analysed areas of collaborative behaviour in humans. It's vital for us to understand how and why we make mistakes - not just in safety critical systems but in all walks of life. When I read that passage above, I see parallels with so many of the mistakes I make on a daily basis at work and at home. I can see myself in every role: the captain, the flight engineer, the first officer, the air traffic controller. You should too. It's so disappointing to see the way the field of UX has latched onto Psychological research findings recently. Take cognitive biases for example: instead of seeing our own weaknesses in them, we've decided to use them as tools of manipulation for the people we design for. "Only 5 items left; sale ends today; untick this box if you don't want to receive emails, but tick the next one if you do…" - This sucks. Intelligence Analysis is another field like Safety Critical Systems design, where bad decisions cost lives. If you pick up a textbook on Intelligence Analysis, the chapter on Cognitive Biases will be written from the perspective of the analyst's reasoning abilities. Understanding your weaknesses helps you avoid making mistakes in the future. This is the way we should be thinking about thinking. Let's analyse ourselves for a change.
Maybe you're a UX researcher, passing choice insights into the product development machine. If what comes out the other side doesn't seem right, you feel free to bitch and moan. "Didn't they listen to the findings I gave them?" Maybe you're a UX designer who works at the early stages of the design process, doing the discovery phase, running workshops, producing concepts, sketches and setting the vision. Again, it's so easy to pass the buck and feel vindicated when the quality at the end of the process is low. Sometimes it feels like you're a doctor advising a sick patient to give up smoking. If the patient keeps at it and eventually dies, it's not your fault, right? Working agency-side makes this point of view even easier. You often don't even see the end result – your agency gets paid and you move on. Being teflon-coated feels safe, but in reality it's quite the opposite. You'll stop improving your skills and sooner or later people will realise you are delivering no value. UX people already have a bad reputation for delivering formulaic rhetoric and not delivering the goods. It's bad for you, and it's bad for our industry as a whole. Fight to make yourself more accountable. Critique, don't complain. Work out how to fix the process. Remember, design isn't just UI. Organisations are designed. Workflow processes are designed. You already have the analytical skills needed to make change happen, you just need to step up to the plate.
Let's face it, there's the right way to do design, then there's the pragmatic way to get things done within your organisation. The two are often not the quite same thing. Let's say you want to create an elegantly minimal responsive site that focuses on the core UX and privileges the reading experience over ads, cross-links and clutter – but can you achieve it? In some organisations you'll have to pry the above-the-fold advertising real estate out of the cold, dead hands of the senior execs. As Leisa Reichelt said a few months ago: "Politics and egos are the main reasons that great design goes awry" […] 'Show, don't tell' is a design principle that seems to work well." Some organisations just have too much invested in their full-fat desktop sites. A messy tangle of revenue streams, too many job roles and too many egg shells make disruptive change feel almost impossible. The mdot site seems like a perfect candidate for bringing in change. It typically doesn't get much attention as it's not a great source of revenue. It's often dated, and the organisation knows "something" needs to be done. This allows you to quietly go about creating a beautiful responsive site. You're free to streamline the user journeys, ditch all the crap that's accumulated on the desktop site over the years and do things the right way. It makes sense to start with a focus on smaller viewports, and gradually expand your attention to larger sizes until one day – maybe, just maybe, once the benefits have been proven with your mdot testbed, you can flick the switch and turn off that old desktop site completely. Credit to Jeremy Keith and Josh Emerson for the ideas in this post.
"There is the story of a drunkard, searching under a lamp for his house key, which he dropped some distance away. Asked why he didn't look where he dropped it, he replied 'It's lighter here!'. Much effort […] in behavioural science itself, is vitiated, in my opinion, by the principle of the drunkard's search" - Abraham Kaplan (1964) It may be an old story, but it's something we're all doing somewhere in our work. The real challenge is finding out where and dealing with it. For example - * Tinkering with the details on a single high traffic page, because that's what's easiest with your MVT package (Button colours and labels don't push the needle far, but it's so hard not to scratch the itch). * Never going back to challenging design problems because "the decision has already been made", and instead turning your attention to the new stuff because it's shinier and more exciting. * Focusing on areas that you've already been running research on, on the logic "we're looking here already, there must be some gold here if we keep trying!" * Looking at just one small area of the site because that's what the client asked (UX problems are rarely polite and tidy enough to stay within the area originally scoped). * Always reaching for the same research tools (e.g. just analytics), because that's your comfort zone. I've said this before, but we designers are very good at turning our lens of analysis on others, but we often forget to turn it on ourselves. If you're doing your current project with the exact same approach your the last one, chances are you've missed the opportunity to do something better.
This is neat. Powwowapp is free little app to help you schedule research appointments. If you work in a UX agency then you're probably used to paying about £70-£100 a head finders fee for some recruiter to trawl their database and make a few calls. It's worth it if you've got a tough screener spec, but the rest of the time you've got to wonder if your cash is being well spent. With Powwowapp you hook it up to your Google calendar, create your empty slots and you're given a public facing URL for you to share. People can then pick a slot, book themselves in and give their details, which appear directly in your calendar. The neat thing is that the participants don't get to see any other entries in the calendar, nor any of the other participant's details. I imagine it would work pretty well in combination with a targeted mailshot (if your client has an established user base), a targeted ad campaign, social media or a combination of all of them. It doesn't give you the ability to pre-screen the participants (i.e. have them fill in a form and prevent them from signing up if they aren't suitable), but you can achieve that with a bit of Wufoo magic if you need it. Powwowapp was made by Viget as a side project to scratch their own itch. It is in alpha but it seems pretty solid to me. The real downside is that without a revenue model or any explicit commitment to running it long term, I'd be reluctant to use it on a big client project. With a bit of support from the UX community, I'm sure they might give it some more attention – why not drop them a line to show your support.
It seems that assumptive personas are getting fashionable again, thanks to Lean UX's Proto-Personas and Gamestorming's Empathy Maps. Getting stakeholders to think about their users is a good thing, but it's dangerous when you start treating them as facts rather than hypotheses. Maybe it's time to trot out that old argument again.
Yesterday I quit my job at Clearleft to become an independent UX consultant. I'm going to miss this place. What I've loved about Clearleft is that it's just so different to any other agency I've worked at. There's no company process - everyone's encouraged to experiment and try different techniques to suit the client's needs. There's hardly any internal meetings. I've never once had a conversation about my billing efficiency. The focus is on quality, and profitability is almost seen as a by-product. You're encouraged to share your learnings externally rather than keep them in-house. Everyone's trusted and given a lot of independence. I've been a UX consultant of some form or other since the early 2000s. Even though I say it myself, I've been pretty good at "standard" UX practice for many years. It was working at Clearleft that made me realise that those skills alone are not enough. Achieving good design for a client takes more than just good design. It's providing education, therapy and facilitation. It's about getting a client's work environment ready so that good ideas have a place to grow and flourish after your project has finished. Your attitude to UX changes when a client hires you to make good UX happen and to see it through, rather than just to turn the handle of a small cog in the machine like user research or prototyping. In Dan Saffer's Interactions talk last year he joked that UX designers are good at "keeping it vague" and that they focus on the "easy, fun part". I'm happy that doesn't apply to me anymore. So if I've got so many good things to say about the company, why am I leaving? It's simple really. I've always wanted to do this - it's not a sudden change in heart. This is the point in the post that I was expecting to give you a sales pitch and plead "hire me, I'm available!". In fact since announcing it yesterday afternoon, I've already been booked up until July this year . It's heartening to know that now I've made the jump, the water is warm enough after all. Thanks everyone. Oh, and don't be a stranger, connect with me!