The World’s worst best website – redesigning MoneySavingExpert

Back in October Damian Proctor (Redweb’s Head of Design) and I were asked by Web Designer Magazine to pick a website that we felt represented ‘bad’ design, explain why and offer up a solution of how we’d improve it.

Web Designer Magazine (October, issue 163)

Web Designer Magazine (October, issue 163)

We didn’t want to choose something obvious or easy. We wanted to challenge ourselves by picking a website that was well known within the UK, on some level successful and then look to improve upon it. We’re both strong advocates of the adage ‘good designers redesign, great designers realign’ so wanted to show how our chosen design could be iteratively changed to best meet the needs of its users rather than just redesigning it based on an emotional response to aesthetics. We recorded the process we went through to highlight each iteration and help us to write about it later!

With Damian responsible for the visual design I took lead on the user experience and UI design.  The split in responsibilities and iterative approach seemed to suit a blog post double bill, starting with this one which covers our initial thoughts through to a new prototype UI. Damian has then taken up the baton and run flat out to the finish line leaving tales of visual design, brand development and our final solution in his tracks. No doubt his post over at Randomapricot will be far better than my atrocious metaphor!

Picking a bad design

Having asked friends, family and the wonderful world of Twitter to come up with suggestions we settled on what we felt was the ideal example of a bad yet successful website, enter

According to Hitwise MoneySavingExpert has been the number 1 UK Business & Finance website since 2006. This year it ranked as the 100th most visited website in the UK whilst the forum ranked as the 10th most popular social network. It receives over 7.5 million unique visitors a month, supports 25 full-time staff and annually raises over £100k for charity, in fact raising enough money that they’ve setup their own charity just to handle the administration of the money, and yet this is all achieved without displaying adverts or charging users.

Current situation

With 3 million email subscribers and over 1/2 million forum members they have a social obligation to make the site perform as efficiently as possible. Its clear that there is a large and active community around the site but the design is flawed enough that it hinders rather than helps its users. In effect MoneySavingExpert is successful in spite of its design. - before - before

Clarifying the proposition

In their own words MoneySavingExpert’s proposition is to “save you money on anything and everything by finding the best deals and beating the system… based on detailed journalistic research and cutting edge tools.” They do this through several channels, the main being a weekly email.

The proposition is nowhere near clear enough, the importance of the email is lost amongst the crowded design and confused by the heaviness of content. Similarly the unique way in which they’re funded isn’t immediately apparent.

Simplicity and reducing noise

Following our initial research we wanted to show that with minimal effort we could noticeably and positively impact on the design. One of the biggest problems we faced was the amount of visual noise cluttering the design. with this in mind ambiguous icons and unnecessary graphical elements were removed, the usability of basic elements (such as calls to action) were enhanced and all instances of repetitive content were taken out. Instantly the design seemed cleaner and clearer as well as giving us a better idea of what we had to work with, what content and functionality we had to focus on and how far we could push the design.

Spring cleaned current design

Spring cleaned current design

Improving structure and layout

noise Once the initial clean up was complete we soon realised that, intentionally or not, the design was roughly built on a five column grid. Our first iteration was tweaked to fit and the foundations of our design started to take shape. I also took the opportunity to further improve the layout. Graphical text was replaced, the number of colours used reduced and primary navigation simplified.

Optimised design with grid

Optimised design with grid

Back to the drawing board

With the content stripped back and a basic structure applied, I switiched to my trusty layout pad and started sketching out new solutions. To help retain familiarity with the existing design I kept the prominent right column and grid system but made it more focused by dividing up the content based on whether it was financial advice or related to the site itself (e.g. about Martin Lewis, how MSE is funded). The latter, being less important, was placed within the right column so as not to distract from the main content.

Sketched wireframes

Sketched wireframes

MoneySavingExpert is described not as a website with an email, but more an e-mail with a website, with this in mind I treated email subscription as a primary function so moved it away from the right column and placed it in a more prominent position.

Global elements were next on my agenda. Search was moved up into the header and the primary navigation made more consistent by subtly changing things like the labels, order and spacing.

Within the main body money saving features were grouped together and placed below the email signup to help reinforce the benefits of subscription and support regular visitors looking for the latest financial advice.

Martin Lewis, the man behind MSE, has written several money saving books all of which can be purchased via the site. As the proceeds go straight to charity they were worth highlighting but not at the expense of the sites main objectives so they were placed together below the financial advice and offers. Beneath the book shop went any additional supporting content such as free tools and the visitor poll.

Wireframe (Axure prototype)

Wireframe (Axure prototype)

Although it seemed like I’d made loads of changes to the wireframe it still felt familiar.

A new visual design

Once happy with the wireframe I handed across to Damian to work on the visual design. Part 2 of this post explains Damian’s approach and reveals the final design. Read part 2 on Randomapricot now.

But before you go, I’d love to hear what you think. Better? worse? Rubbish approach? Please feel free to leave a comment, as always any feedback is really appreciated.


Master User Experience design an interview with .net magazine

Back in November I was approached by Craig Grannell who asked if I’d like to contribute towards the cover article for .net magazine’s December issue. I jumped at the chance and was really pleased to have been given the chance to comment along-side the likes of Leisa Reichelt, Dan Saffer, Andy Budd and Luke Wroblewski.

With .net describing the article as “demystifying the process (UX) behind web design and development’s fastest-growing and potentially most important industry” I thought it would be worth me posting my responses to some of the questions I was asked.  Enjoy!

.net magazine (December 2009, issue 196)

.net magazine (December 2009, issue 196)

What does user experience mean, as far as you’re concerned?

To me user experience is all about the way in which people interact with a product or service and how they feel about that interaction. It’s about providing a great experience that successfully meets the needs of the people who use it. UX design can often be confused with user interface design but its so much more than that, it encompasses the complete experience someone has with a product and the organisation responsible for it, from the initial handshake right through to the happy ending.

Why does user experience matter? What are the benefits?

Taking a user centered approach to design results in a far richer and more rewarding solution when compared to more traditional methods. For users it means increased productivity and success and in turn gives a company greater advantage over their competitors.

The benefits speak for themselves really. Focusing on the user experience leads to increased revenue (or conversions), helps to reduce cost and promotes customer loyalty. If users are more efficient and effective their chances of converting are far higher. A good example of this was a recent redesign Redweb completed. As part of the redesign we looked at their reservation system and carried out usability testing throughout the project. This resulted in an increase in bookings of 123%.

Costs can be reduced both in the development of the product but also in regards to customer service. By carrying out sufficient research early on the whole team, including the client, are clear on what needs to be achieved and it also limits the chances of unexpected requirements cropping up towards the end.

Once deployed if the product is easier to use then customers have less reason to contact you. If the users are successful and therefore happier they become more loyal, and loyalty generally spreads to friends and family which in turn leads to increased revenue.

How do you approach user experience when creating websites and working with clients?

For me the most important aspect of starting any project is to first educate the client to the importance of a user centered approach. Sometimes clients can be overly keen to see something tangible straight away whereas we first need to fully understand the requirements of the business and the needs of the users. There are many research techniques Redweb use such as collaborative workshops with stakeholders or contextual interviews with  users. We like to continue the same level of openness we show early on throughout the project and involve our clients throughout to help them feel as much apart of the decision making process as possible.

Do you make use of process flows, wireframes and other diagrams, and prototypes? If so, how and why?

Project deliverables differ depending on the brief. Ultimately we like to work collaboratively as a team and openly with our clients and produce artefact’s wherever necessary. As a team Redweb focus on creating a functional prototype that we can use as a base of reference which can also be tested directly with end users and seen and played with by the client. Having said that, any prototype always starts with hand-drawn sketches.

Likewise, do you create user journeys and experiences/work with personality persona’s?

In the early stages of a project Redweb focus very heavily on understanding who the users are and how they’re most likely to interact with the system we’re designing. Personas can be a great tool to focus both the team and the client throughout the life-cycle of a project.

How do you test user experience/interaction and information architecture? What should you look for?

There are many ways to test the effectiveness of an information architecture but it’s as much about knowing what to leave out as it is what to leave in. It’s very easy to take on board everything that a user has to say but a lot of the time its not necessarily what they say but what they do that’s important. It’s about looking at the wider view. Its not so much about concentrating on the little details but trying to understand what they mean on a more global scale; is the tone of voice wrong throughout? Does the site poorly address the users mental model?  Their frustrations may not necessarily be with the interface that you’ve designed but the process they’re being made to go through; 20 questions when 4 will do, or highly personal questions that don’t seem relevant to the task.

What can designers do regarding testing if they have a really small budget?

These days there are so many options available to designers working to small budgets, you really are spoilt for choice. There are loads of services to help designers with remote testing, from apps like Websort, Treejack and Survs to iRise and ‘The 5 Second Test’. But I wouldn’t recommend relying purely on remote tools. With the growing popularity of remote testing designers can too easily overlook the benefits of meeting and speaking to users first hand. It doesn’t have to be in the form of lengthy and extensive lab based usability testing, it could be as simple as setting up a laptop in a coffee shop and offering people a free drink if they give you 5 minutes of their time to look over your latest design, it can provide invaluable feedback and if you pick a coffee shop with a rewards scheme and speak to 4 users you’ll have earned yourself a free coffee by the end of it!

What do people get wrong regarding UX in web design? What common mistakes do you see or misunderstandings do you find are rife?

From a designers point of view at times it can be easy to focus too much on the little details and neglect the bigger things about how a user arrives at the site, what task are they looking to complete and how do they go about achieving their goal. It’s always worth asking yourself ‘have I done enough to help them?’

With regards to clients they are becoming far better educated than they previously were but sometimes a little knowledge can be more damaging. They can form an opinion on a particular subject and it can be very challenging to move past it. The ‘3 clicks or less’ issue is a good example of this. Although the rule is a commonly held belief there is no scientific proof to support it. A study conducted in 2003 found there to be no correlation between the number of times users clicked and their success in finding the content they were looking for. Analysis showed that there wasn’t any more likelihood of a user quitting after 3 clicks than after 12.

What do you think the next big development in UX will be?

Usability testing relies on people telling us what they’re thinking either whilst they’re working through a task or explaining their actions in exit interviews. It helps us to understand ‘why’ they acted in certain ways but its not the most effective why as it doesn’t uncover subconscious actions or problems that they may have been too embarrassed to admit to. Eye-tracking can be very effective at uncovering what people are focusing on, the priority in which they look at things and the length of time they concentrate for but again it doesn’t satisfactorily tell us how the users feel about their experience and at what points their emotive responses are strongest. I think the next big steps will be in finding ways to measure this sort of response effectively.

Case study within the .net article

Case study within the .net article

If you made it through this far thank you for taking the time, and well done! I’d strongly recommend reading the full .net magazine article if you can get your hands on it as it’s far more interesting and well worth it. If you have any comments on anything I’ve said please add them as I’m always interested in receiving feedback.

techradar logo smallNote (added 13 January 2010): Craig Grannell’s article has now been added to TechRadar. Read the article “The Web Designer’s guide to User Experience” on TechRadar.

Mountain biking, misdirection and peanut butter sandwiches

Recently I took part in a 30 mile mountain bike ride. It was quite a challenge and on reflection made me think about not only my cycling ability but also user experience design… naturally!

This post explains how I’ve made the tenuous leap from mountain biking to UX and then to my thoughts on Jared Spool’s lunch.

My DorsetDirt experience


The annual event is a challenging loop around Dorset starting and finishing in Dorchester. Having grown up in the area the route looked pretty straight forward, the ride not too competitive, and on the day the weather was perfect.

I forgot to pick up an OS map before leaving home but made the assumption, based on previous experiences, that on registering for the event I would be given a route map, so wasn’t too worried. On arriving at the start no maps were available, only written instructions. The directions seemed clear but without a map or prior knowledge of the route I couldn’t be sure. I settled on the tactic of following the herd and for the first half of the ride it worked nicely. I simply made sure I stuck like glue to a decent sized group of riders and let them do the navigating.

As I reached the halfway checkpoint it became apparent that I was near the front and the only people leaving were in ones and twos. My plan to follow the herd looked like it might fail. I latched onto the biggest group I could find. At the first junction we stopped and asked each other ‘where next?’ We started making regular little mistakes, so I turned to the instructions for guidance in an attempt to figure it out for myself. Before long the group had split up and I was riding on my own. I was pretty sure I was on the right course but riding alone made my confidence drop.

The last instruction on the sheet before the penultimate checkpoint was to turn down a steep hill towards a Friary. After 2 miles of cycling I realised I had taken a wrong turn. I blamed myself for it and headed back in the opposite direction. After a short distance I met 2 other riders who had made the same mistake. I felt better knowing that I wasn’t the only one. We turned down the next junction we came to and rode for another couple of miles. We soon came across another group of riders all studying the instructions. We were all lost and couldn’t figure out how. Within a couple of minutes 2 more riders turned up from a completely different direction to the rest of us. We were all pretty tired and being so lost was starting to annoy all of us. We were in a steep sided valley and through one wrong turn we were having to ride up and down the valley to find the correct path out which made the ride all the more tiring.

We read back through the instructions one last time and I realised that the instruction to ‘turn right down a steep hill to the Friary’ was actually just a comment ‘there is a right turn down a steep hill to the Friary’. We had all made the same mistake and the error had cost us time and energy and had negatively impacted on our experience.

Clear sign posting

In retrospect my experience had a lot in common with issues I often see people face when taking part in user research.  Each time I made a mistake, however small, I instantly blamed myself and with each error my frustration grew.  I blamed my poor preparation, which in the case of the ride is probably quite fair, but conversely if my preconceptions were commonly held, that preparation wasn’t necessary,  then it should be catered for. An good online example of this is the application for a UK Driving License. It’s a legal document and you’d expect to have to submit some official information, but not necessarily what sort. The first step of the process deals with this well, it clearly explains who can apply and what is need to complete the application.


DirectGov lays out clear instructions at the start

During the ride the errors at first were slight and easily corrected but as the ride progressed my confidence dropped and I had a strong feeling I was the only person that had made mistakes. I’ve witnessed this regularly in usability testing, however much a system is at fault people will often blame themselves and believe that they’re alone in making it. It’s often not until you witness someone else make the same mistake that your confidence is reassured. When I took the wrong turn I felt far more confident when I saw another rider, even if they didn’t know the right way either, at least I wasn’t the only one!

“Any good global navigation scheme should, at a glance, answer the top three questions every user has at the back of their mind.” Derek Powazek

The lack of a route map wasn’t the only problem, the route wasn’t sign-posted either which meant even when you were on the right track there was still a sense of uncertainty. Derek Powazek in his A list Apart post ‘Where am I?‘ explains that for navigation to be successful it needs to answer 3 questions:

  1. Where am I?
  2. Where can I go?
  3. Where have I been?

Without the reassurance of clear sign-posting I became paranoid that if I needed to, I wouldn’t be able to find my way back and although ultimately I completed the ride and was therefore successful in completing my task I had more work to do to get there. The extra effort wasn’t necessary and negatively impacted on my overall experience.

My diminished experience wasn’t the fault of the organisers, they did a great job, they just hadn’t necessarily taken context into account. After 20 miles or so a lot of people would have been tired and hungry, and if they weren’t familiar with the course their chances of going wrong would be increased. When mentally and physically tired its far easier to make a mistake, especially if in my case it looks like the easier option!

Lessons learnt

In summary, there are 5 lessons that can be learnt from my experience and applied to web design:

  1. Understand your users preconceptions
    What preconceptions will your users have of your service or product? If they don’t match make sure its clearly communicated to the user.
  2. Instructions are only needed if the journey isn’t intuitive
    Optimise the interface and make sure instructional text is only used when absolutely necessary.
  3. Make sure each step of the way is clearly sign posted
    Where am I? Where do I go next? How do I get back to where I started?
  4. Consider the context
    Accessing a website at home in a comfortable environment with no distractions is a totally different experience to accessing the same site in a busy work environment or on the move via a mobile device.
  5. And lastly, provide free tea and biscuits at the end In other words make sure the finish-line is obvious and if there is no direct reward (e.g. a successful purchase) make sure there is a sense of completion or achievement.

Peanut butter and Jelly sandwiches

In writing this post I was reminded of a workshop technique that Jared Spool wrote about a while ago on the UIE blog which helps to illustrate the importance of clear instructions. In my next blog post I’ll be looking at Jared’s techniques and the use of instructions within web design.

If you’d like to read the next post as soon as its out subscribe to the RSS feed or follow me on Twitter.