Staying ahead in the world of online retailing requires continual research and analysis into customer shopping behaviours. As part of their annual brand strategy review, one of the UK's leading retailers commissioned Flow Interactive to carry-out to a research study with key online customers to help evaluate user needs and desires with regard to grocery shopping. The purpose of the study was to gather customer perceptions of several grocery retailing websites and to identify what user needs are currently being satisfied or dissatisfied. There was also a focus on assessing the proposition of aggregated grocery shopping sites (like mysupermarket.com) and investigating perceptions of grocery shopping through mobile devices.

Flow carried out a set-of contextual interviews with customers of varying shopping frequencies alongside an expert review. Key shopping journeys and tasks were identified in order to focus research on business priorities. During the user research sessions participants were asked to carryout a couple of key tasks across multiple online retailing platforms and were probed to answer questions regarding their expectations and habits of shopping online. The expert review was competitor focused and allowed for several retailing platforms to be assessed from a heuristics perspective. The research and analysis was completed over a couple of weeks and led to strategic findings in relation to how to support different customer mindsets along their shopping journeys.

The insights were shaped into a set of strategic recommendations that will be used by the client to drive the future direction of their multi-channel retailing offer.

What is the main task I have to accomplish while reading an article on the web? The answer is obvious: the task is to read the article.
Yet looking at most websites, only a small part of the webpage actually supports this task. Safari Reader is the latest attempt to help users take matters into their hands. Read more

Registration is a crucial initial step that most online businesses have to impose on people along their journeys. Registration is necessary to check people’s authenticity and start meaningful conversations with them based on the provided details. Capturing people’s details correctly is paramount since storing, for example, an incorrect email address opens the door for trouble down the line. With an incorrect email in the database, not only does the business lose the opportunity to reach out to its customers, but the business’s bottom line may suffer. For example, I have heard about cancelled orders due to mistyped email addresses.

It is no wonder then that registration forms try to make sure details are captured correctly. But how to do it while still preserving a positive user experience? Registration forms basically represent a barrier for people to be overcome before they can do what they actually want to do – finally use the website!

Here is how others have tried to handle this (with varying success):

Confirming entry

I frequently see a registration form that has duplicated Email or Password fields. Now, this is a little bit annoying, especially if both Email and Password need to be confirmed (as below).

In the above example, the person’s interaction flow is significantly interrupted by having to answer two identical questions. As per Don Norman’s model of 7 Stages of Action, answering each single question on a form is a small diverting action on the person’s journey towards accomplishing her goal.

Disabling copy & paste

Harry Brignull wrote about a registration form that does not allow pasting into the ‘Confirm email’ field. Quite creative, but I agree with Harry that it could feel patronising, especially for the more tech-savvy people (who know how to copy-paste). On the other hand, it prevents people (hopefully) from simply replicating a typo made in the first field. And typos are arguably one of the commonest kinds of incorrectly entered details. Now let me ask, why do most websites actually use the wording ‘Confirm your email’? Let’s use ‘Re-type your email’ instead, and it might not be necessary to awkwardly disable standard system interactions like pasting.

Repeating key details before submit

A more elegant solution is not to display the second confirmation field at all. But how can businesses eliminate the eventual errors on forms then? I quite like concept prototypes created by Jonathan Knoll and Russ Unger, that repeat the entered email just before submitting. Jonathan and Russ have produced multiple variants, but variant 5 (picture below) is my personal favourite. It puts the entered email within the person’s locus of attention which is at that point in time on the Submit button.

Unmasking passwords

What about passwords, that are by default masked on most forms (even at registration)? First of all, I believe masking a password does not bring any value in most usage scenarios. Nielsen calls for the death of masked passwords, and I am happy to agree with him. However, as opposed to offering a checkbox to mask the password, as he is suggesting, I think the way to go is actually offering a checkbox to unmask the password. After all, in most contexts security is more important than interaction efficiency.  MailChimp is doing this already, and based on a recent live demo of FontDeck, it seems like we will be seeing this pattern more often.

A pattern for unmasking passwords is also frequently used on mobile devices. This is due to the lack of tactile feedback provided by touchscreen keyboards when inputting a password. Moreover, people also cannot rely on their motor memory (remembering the finger movements like in touch-typing, as opposed to the actual password characters). People often utilize the motor memory to enter passwords with little conscious effort, and this does not translate so easily to touchscreen keyboards as visual identification of keys is needed.

Most mobile interfaces support people by revealing the last character entered for a short time and then masking it, thus giving people the necessary feedback. I am not aware of any website doing the same, but it might be a solution for standard monitor-keyboard setup too. On the other hand, the utility of this short-time revealing is debatable since most people type so fast that revealing the last character and masking it with a time delay is very difficult to implement seamlessly.  Try it for yourself - here is an example of automasking.

Inline validation

Another powerful weapon against incorrect entries is inline validation. Validation can only catch a small proportion of specific errors, but it is generally a good approach since people are notified something is not quite right before they hit the Submit button. Therefore it eliminates the need for the dreadful error messages. “Fatal error - you have not filled in all the details!”. “Oh my god, fatal error - someone actually died!” screams the user in horror.

There are multiple ways of implementing inline validation. Luckily for us, Luke Wroblewski put a few validation variants to the test. Based on his study, validation ‘after’ (after the person indicated that she was done answering a question by moving on to the next one) is the winning option - both in terms of efficiency and satisfaction.

So what?

Incorrectly entered details in online forms are a frequent problem that can cause a lot of hassle down the line. However, when designing forms, make sure you use a sensitive approach to minimising those errors and do not make the people do all the hard work for you.

I would love to hear about your tips for minimising errors in forms.

More and more websites are using collaborative filtering recommenders to personalise their goods and services for you.  For instance, Amazon’s “Customers Who Bought This Item Also Bought,” uses collaborative filtering technology to let you know about other products that might be of interest to you.

Figure 1.  An example of collaborative filtering recommendation demonstrated on Amazon.com

Simply put, collaborative filtering recommenders allow a website to recommend stuff based on how similar your browsing behaviour is to that of other users.  These recommenders will often rely on some correlation threshold value to determine whether you do or do not share mutual interests with various other users.  A really good example of music recommendations based on collaborative filtering would be last.fm (www.last.fm)

One issue with collaborative recommenders is a result of a user's divergent goals - different goals that a user might have when using a particular interface.  For instance, say that a given user is a keen fan classical music (a Mozart aficionado), but regularly listens to Lady Gaga (and other contemporary pop music) when with certain friends because that user knows that Lady Gaga-type music will facilitate a belongingness with those friends.  When that user is alone, s/he wants to listen to classical music and be recommended only this music without having to sort through recommendations based on when Lady Gaga has been selected.

A solution that has been devised to help users with recommender noise that results from divergent goals has been to include product information (e.g., classical versus contemporary pop music) when the recommender filters your recommendations.  That way, our example user only gets recommendations for classical music when listening to Mozart.  Recommenders that include content information with collaborative filtering are called hybrid recommenders.

Another issue for any recommender system is what to do when a new user or new item comes along, commonly known as cold start. For the last four years, I’ve been looking at the relation between people’s music preferences and their personalities, which could be used as an alternative way to help resolve the cold start problem and improve music recommenders.  An associate of mine at Cambridge University, Dr. Jason Rentfrow, does a great job in describing the music preferences and personality research (see http://www.youtube.com/watch?v=29-xYiOOc8w).

Researchers like Dr. Rentfrow have identified relations between the genre of music that people listen to (e.g., rap or jazz) and personality characteristics that those people generally have (e.g., extroversion or openness to experience).  Still, genres can be really vague.  I mean, you and I might both love rock music, but are you going to necessarily love the same rock music as I do?  Instead, to help recommenders to their job, I identified audio features prominent in certain music genres and link these to personality characteristics.  So, instead of saying that extroverts like rap music, I say that extroverts really like music that has a lot of beats that happen quickly together… constantly.  This relation might apply mostly to rap music, but might also apply to certain rock songs, electronica songs, you name it.  Conversely, my research suggests that introverts like music that has few beats in the music, which is typical of classical music, but again, is not exclusive to classical.  As a result, identifying the relation between personality and music preference toward specific audio features can help improve both issues described above by identifying and sorting music according to more precise and objective audio features.

In sum, the work that I have described is still very new and there are a lot of challenges to see through before recommenders can truly become 'personalised' by learning and understanding users’ personality, but there are a lot of opportunities that may result from this type of personalisation as well.

So tell me, do you think your musical tastes describe your personality?  Have you made personality judgements about a person because of the music they listen to?

Also, do you think that there are similar aspects common in most or all of the music you listen to?  Or, do you tend to have a favourite instrument that you like when it’s played in a song?

Finally, what do you think about music recommendation based on personality?

Your comments (positive or negative) are welcome.

What does it mean to be an expert on User Centred Design (UCD)? What does it require to be a User Experience (UX) expert? What kind of educational or experiential background do you require? What differentiates an expert from just a consultant? Is an expert someone that knows the UCD process and is proficient at a large variety of UCD methodologies? Is expertise measured by the consultant’s academic credentials, industrial experience, number of clients, or knowledge of a variety of industries and platforms?

What makes the foundation of an expert is all of those things; education, experience, and a solid knowledge of the processes, methodologies and tools. But what differentiates the consultant from the expert is not just being an expert at implementing UCD in perfect conditions, but the ability to implement UCD in the ‘not so perfect’ context of the client.

As experts we need to be able to assess the client’s current processes, phase of development, schedule, and budget, and then determine what activities and deliverables will provide the best returns within their context. But this isn’t where it should end. As a UX expert, you should be able to deliver not only tactical recommendations, but strategic ones as well. Those strategic recommendations are not just on the user experience of their system or product, but on the activities that can help to move the client towards a more efficient and productive implementation of UCD.

As true user experience experts, we need to propose and implement activities and deliverables for the best return in the context of the client, and work with the client to migrate to a process that will bring to fruition more of the strategic value of UCD.

I’m not going to predict how well it will even perform under one handed operation, however even if it were only as light as the average magazine (like say Communications of the ACM, which I have here on my desk), holding it up with one hand and operating it with the other is going to be a strain even if the arm holding the ipad is supported.

Damn, I think I just talked myself out of buying one… maybe

Personally I’m really quite excited about the iPad and it might well be the first Apple product that I have bought in something like 6 years -  but I’ve had a niggle since I first saw it in action and read about it, which I haven’t seen anyone else pick up on yet (SURELY someone will). That is how inherently awkward double handed interactions with large portable interfaces (or tablet PCs) are.

“What?” you say…

This, I say: that in order to use two handed interactions, you need to put the iPad down on something - sometimes rather suddenly mid interaction - and that to do so is just plain ergonomically unsound.

Let’s review some options for using the iPad:

1. Your lap

Some may remember the days of the NTL set top box and its walled garden internet when internet through the TV was trialled 10 or so years ago. Some clients rushed to get their websites converted for the walled garden so that they’d work on the TV. You had to use certain colours, no stripes, I think it was 640 x 480 resolution, HTML 3 (no frames), table based layouts only, and absolutely no javascript. To interface with this wonderful garden of delights you used a keyboard on your lap.

I didn’t need to conduct any usability trials to work out how this would go. Try it now, put your keyboard on your lap and see how comfy it is to use. It isn’t is it?

To see this in action check this video about how to use an Apple application at 20 – 35 seconds. The narration suggests it's easy but watch what the narrator actually does with the tablet.

2. A desk

Place the iPad on a desk and suddenly you’re craning over at an 80% angle to look at the screen (which is normally upright) in order to see what you’re doing. "Ouch, my neck!"

3. A stand / mount

With the iPad on a stand here are your choices: either the keyboard is at the wrong angle, or the screen is, or the whole thing is at the wrong height, or more likely all three… That's an Ergonomics fail!

4. Your lap II

Legs up on your sofa, knees raised, head rested on a comfy cushion, iPad on your lap. AHA! The one comfortable position in which you can make your double handed interactions; just don’t forget some Velcro to stick your iPad to your trousers to keep it at just the right position.

But wait, OH NO, now my wrists are at a 90 degree angle to my arms. "Ouch, ouch, ouchety-ouch."

In conclusion, the ergonomics of the iPad is not looking good. I’m not going to predict how well it will even perform under one handed operation. Even if it were only as light as the average magazine, holding it up with one hand and operating it with the other is going to be a strain even if the arm holding the iPad is supported.

Damn, I may have just talked myself out of buying one… maybe...

We’re very proud to announce that the Ergonomics Society has awarded User Experience Consultant Frankie Pagnacco their Ulf Aberg Award for her Masters project. Frankie completed the project on sensemaking in the control of Rapid Urban Transit systems in 2008, as part of her MSc in Human-Computer Interaction with Ergonomics, at University College London's Interaction Centre.

The dissertation looked at how control room staff at London Underground’s Victoria Line made sense of the information they received about on-the-ground events through cues from their equipment and from each other. Using field observations, the study uncovered the situations that gave rise to sensemaking, the strategies adopted to ease and speed up sensemaking and the bottlenecks in information-seeking.

The Ulf Aberg award, given annually, recognises outstanding Masters projects in Ergonomics. Projects are assessed on the quality of the research, expertise, originality, clarity and interpretation of results.

Ulf Aberg, after whom the award is named, began his career in 1961 and spent over a decade working with Ericsson and later with the National Defence Research Institute. Aberg co-authored the first Swedish textbook on ergonomics and was the first foreign member of the Ergonomics Society and the founding chairman of the Nordic Ergonomics Society.

Congratulations to Frankie!

This may sound controversial but I admit it, I don’t love my iPhone. I realise I could get into trouble for admitting this publicly but I’m prepared to accept that, to get these thoughts off my chest. I was considering going to a self-help group, especially as I am surrounded by lovers of the phone, but instead I am sharing my thoughts on our blog. And I do expect arguments to the contrary.

There are a number of reasons I don’t love my phone (and a few reasons it’s not so bad):

1) I can’t walk down the street writing a text message

Ok, I like writing text messages and I like to send a quick message every now and again. I am also female and I like to multitask. With my old phone (Nokia N95) I used to be able to walk down the street, not looking at my phone and feel the buttons and know what I was pressing and write a message. With my nice smooth-screened iphone I have no clue where the buttons are and lampposts keep jumping out at me so the quality of the experience and my efficiency has decreased.

2) It feels like I’m putting unnecessary stress and strain on my thumbs while I try to hover to write text

Maybe it’s just me (as a 2 thumbed writer of text) but when I type any text into the iphone I find that my thumbs are kind of hovering over the keypad and I take more strain on them. If I’ve been taking lots of notes or writing longer messages I feel my thumbs starting to get tired. I can’t seem to find a nice resting point on the phone without activating a key.

3) I feel like I have to be a robot and hold the phone just at the right angle or it keeps switching between the different views

Now I don’t tend to hold my phone particularly straight when I’m using it, but as a traditional girl I prefer my phone to be upright (in portrait view) rather than sideways (landscape). However, on my iPhone, if I’m just casually looking it, reading something perhaps, then the screen has a tendency to just switch without asking to landscape view. All I do is casually hold it at a comfortable angle in my hand. Now, if I was a robot, then everything would be at neat 90 and 180 degree angles, and I wouldn’t have this problem. But I’m not a robot. (Someone told me how to get it back to portrait, but it doesn’t seem to work all the time).

4) I have to keep the phone switched on for my alarm clock to go off in the morning

I’m the type of person who likes to go to sleep with their phone switched off and charging overnight. Now, I know that the iPhone isn’t capable of this, although I have no idea why. However, back to my alarm - if I switch my phone off it doesn’t wake up automatically and switch itself on and then wake me up. Very rude if you ask me. So now, if I’m to use my alarm I am forced to keep the phone switched on. I know silent exists, so I shouldn’t be disturbed in the middle of the night by incoming calls or texts, but if the phone is there, and I wake up in the middle of the night I might sneak a quick peek to see if I’ve received a text overnight, so, I prefer to have it switched off. And I don’t want to lose precious battery life if my phone doesn’t need to be charged. I also liked the way my old phone would tell me how many hours I had to sleep, something my iPhone doesn’t do either.

5) It takes longer to send text messages

One of my ex Flow colleagues Martin once did a study looking at how fast people could send text messages on different devices. There was a difference between the speed that they perceived they were typing the message and the speed that they actually did type the text message. I haven’t measured myself, so maybe it’s true for me too (although I think not if you know me and have seen me text).

T9 was great. Nice and easy if you’d learnt it, with an option to scroll through the words that might come up as possible combinations. Worked fine. Now I have a qwerty keypad and no choice to revert to the old numeric keypad T9 that I love. I was hunting around for the option to switch this on on my iPhone but there doesn’t seem to be one – of course I didn’t get a manual (as it relies on me being motivated to go and seek one online) so I haven’t been able to check that. Now, I can touch type on a large keypad so I’m pretty aware of where the buttons are on a keypad but having to use them with my thumbs causes no end of problems. I’ve got quite large fingers and thumbs for a girl, but still I’m constantly pressing the wrong key, switching to capitals when I didn’t want to, trying to work out how to not accept the word suggestion it offers baffles me. Though I do like the way that the recommended words account for the fact you probably pressed the wrong keys.

6) I don’t know how many text messages I’m sending

Now, for those of you who aren’t aware, not all phone tariffs have unlimited texts. Therefore there is a big difference in price between a text that is 160 characters long and one that is 161 characters long. My iPhone seems to disregard this, and doesn’t tell me how many characters I’m using. Is the assumption that we all have bottomless pockets or that the number of text messages flying across the world should increase? A simple indication of the number of texts I’m creating would be a real bonus.

7) The battery life is bad – I refuse to keep it connected to an energy source all day

Ok, if you don’t use the phone then it’s fine. But as it is, there to be lots of other things on the iPhone aside from the phone – and it’s designed for these to be used better than the simple phone functions (as far as I can tell). It seems a real shame that there isn’t a low energy mode that will conserve power, or I can’t switch the thing off while I’m powering up overnight.

Things I love about it

1) It’s nice to have all texts viewed as a conversation

This is a really nice touch. I’d like to have the ability to order the text conversations alphabetically by person though, so I can easily find a previous one with a certain person, but then I’d have to be able to short link to a letter in the alphabet but as there isn’t a keypad, I can’t seem to do this – hmm…

2) It’s  nice to be able to view my voicemails and play them back

A while back at Flow we were designing a new voicemail system, and people we spoke to said they found great value in voicemails from loved ones, children who lived far from home and relatives who’d passed away, so the idea that the messages live on your phone and you can replay when you like – even on the tube – is a great one.

I’m sure there are more things that I love, but if I was to include 7 here that would seem unfair to my rant.

This is the second in a series of posts about UX Camp London. The first one can be found here.

Back to the Roots: If email is the past, is Google Wave the future?

Ex-Flowster Johanna Kollmann, now doing great things at Vodafone, shared her experience of using Google Wave with a tightly-packed audience. Her main argument was that Wave is a great advance on email, offering us something much closer to natural, oral communication, but with the advantage that it can be stored and traced.

Now, there’s a discussion to be had about what constitutes “natural” communication, and whether what we consider to be natural is just the result of using technologies that we are more used to. But we didn’t manage to have that discussion on the day.

Instead, Johanna gave a demo of Wave, and then took some questions. Though much of the discussion focused on the details of the interaction design (which still seems to have a few kinks to iron out), several people said that they didn't “get” Wave. The problem seems to be that by combining the most useful features of email, instant messaging and virtual conferencing tools, Google may have created a product that, for all its advantages, confuses some people by not being immediately recognisable as one thing or another.

Ground-breaking new products can be baffling at first to people whose expectations are formed by older paradigms, but when we use them they begin to make sense, and we gradually accept them and change our behaviour accordingly (think Twitter, or for those with longer memories, the mobile phone). But on the other hand, some new products are insufficiently well-defined at the proposition level (that is, nobody can quite define what they are for), and our research shows that this inevitably has a direct negative impact on the experience of using them.

It remains to be seen which of these two possibilities applies to Google’s Wave, but I’m impatient to see which one it is.

Johanna’s Slides are on her blog, here.

London’s first UX Camp, a BarCamp-inspired unconference for the User Experience community, happened on August 22 at Gumtree’s offices in Richmond. Over the next few days I’ll be posting my rather belated reactions to some of the best sessions.

X-Ray Listening

Judy Rees, co-author of Clean Language showed us how she teaches people to listen better, using techniques developed in Cognitive Linguistics and Psychotherapy.

I won’t attempt to explain Judy’s method in detail here, as I’m not sure I can do it justice (and, ahem, because my notes aren’t that detailed), but in a nutshell it is a way of combining template questions with the respondent’s own words, to produce endlessly adaptable, open questions. So for example, a template question might be “Is there anything else about...”, onto which the questioner adds a key word or phrase used by the respondent themselves.

After a brief introduction Judy took the group through an exercise. We broke up into pairs, identified some relevant topics for investigation, and took turns at asking questions and listening to each other’s answers. First we did this spontaneously, using our own choice of words, but the second time we used Judy’s Clean Language technique to frame our questions.

The results were striking: everyone said that they felt more comfortable and more ‘listened-to’ when answering the ‘clean’ questions, compared to the spontaneous ones. On the other side, the questioners said that the ‘clean language’ made it easier to formulate the questions on the fly, and elicited more detailed, more honest answers.

At Flow, we spend a lot of time talking to people, trying to ask the right questions, and trying to listen. Most of us find that scripts are too rigid, so we use semi-structured discussion guides to keep us on the right topic, but we formulate or questions spontaneously, using a variety of ad-hoc rules and best practices to get the best results. We are always looking for the best ways to make people feel comfortable, while still getting the freshest and most honest nuggets of information from them.

This brief introduction to Clean Language showed that it is a potentially useful technique for improving both the quality of interview data and the efficiency of the interview process, all the while making respondents feel more at ease. A win-win-win scenario, if I’m not mistaken. I look forward to finding out more about this, and trying the techniques out in a real interview. I’ll let you know how it goes.

The next article in the series is here.