USEworthy March 2003
The monthly Newsletter from The Usability Company

Welcome to March's edition of USEworthy. After her first month working with our team, Lisa Halabi provides a short insight to being The Usability Company's latest 'newbie'. We also provide a report of the problems with the online payment system for the congestion charging in London. The report was carried out seven days before the charge was launched and although some of the issues have been resolved the report provides valuable learning for any website developer. In fact, February will be remembered as the congestion charge month at The Usability Company as it sparked a flurry of press activity for the company and our spokespeople.

Our company news looks back over February and records a number of new client wins. This leads us to consider the current state of the market. With war looming on the horizon and increasing talk of a slow down just how is the market for usability services reacting?

Company News

February brought with it a number of new clients for The Usability Company. The British Standards Institute awarded TUC with a project to review and make recommendations about the User Centered Design of the BSI home page. We were also awarded a contract to review the Planning Portal, a government run website providing information about every aspect of town and country planning. February also provided us the opportunity to get to grips with Siebel when Reuters asked to assist with their internal upgrade to the latest browser based version.

After a rocky start at the TFM (Technology for Marketing) show, when we had some of our equipment stolen from the stand during set-up, we ended up having a useful and enjoyable show. The stand featured two aspects of our service set that enabled visitors to experience a Usability Evaluation for themselves and also see some of the latest online research technologies.

We demonstrated a Usability Evaluation using our mobile equipment, which was back in the UK after being used for testing throughout Europe and as far a field as Macau. Visitors to the stand were asked to take the role of a participant and to attempt a number of tasks on a variety of sites. The picture in picture audio video recording of their experience was provided to them on Credit card sized CD-Rom as a reminder of their experience.

In the area of online research we focussed on WebIQ - E-Mission Scorecard, which received a great deal of interest. Roger Beynon of Usability Sciences Corporation, one of our US partners and the developers of WebIQ was over for the show and explained to visitors to the stand how E-Mission Scorecard could measure whether their websites were achieving their mission. A short in and out survey technique is used that generates a high level of 'opt-in' and identifies amongst other things visit intent and success. The service was met with a significant level of interest and we received a number of firm enquiries.

and Finally...

The Usability Company featured regularly in the press during February. As the UK's foremost provider of business focussed usability services and consulting we are frequently asked to comment on a variety of issues. February was no different and we appeared in a variety of publications including the FT where we appeared on no less than four separate occasions. Two of these items related to the congestion charge, which was also covered on BBC Online. We were also asked to comment on the potential for the DRC (Disability Rights Commission) to carry out a formal investigation into the accessibility of websites. This brought Catriona Campbell, our CEO to the front page of New Media Age.

Congestion website congested - Feb 2003

Commuters and other central London road users attempting to pay the congestion charge on the Web are likely to be in for a confusing and frustrating time. With less than seven days to go attempting to register on the site www.cclondon.com seems virtually impossible as response times currently stretch into minutes.

Web service providers are usually very aware that response times exceeding several seconds are likely to deter customers, anyone attempting to use the service now to pay in advance is likely to face a long wait.

Whilst evaluating the site it was never possible to actually pay the charge for the first day of operation as the server was continually unavailable. This is not the National Lottery where people will be gladly queuing up to donate money to the government, but a charge which is likely to be resented and so any faults in the system magnified in the user experience.

Users will also have to have a degree of visual acuity far greater than that required to drive a vehicle. Like many other Web sites the designers of the pages have decided that all its users are capable of and comfortable with reading sub-ten point text. Whereas this might be acceptable for a commercial site, where the user can choose not to continue the transaction, it does seem unacceptable for what is effectively to some an essential service. Having to choose between squinting at a screen in order to meet the deadline, or possibly have to pay a penalty charge, does not seem conducive to healthy eyes.

Two essential pieces of information have to be supplied on the congestion charge form, the vehicle registration number and the date which payment is being made for. Although UK registration numbers strictly follow a number of defined patterns, no attempt seems to have been made to determine if the pattern has been followed. However as non-UK registered vehicles will be allowed to pay the charge this does not seem a point of particular note.

It is on the date entry mechanism where good Web design practice has not been followed. The overwhelming majority of websites clearly indicate the month field by using a drop-down list of month names; the congestion charge form has chosen to use the inherently confusing month number technique. So the entry 08/03 might be the eighth day of the third month, or it might just be the third day of the eighth month. The use of a drop-down menu would obviate this potential confusion whilst the provision of guidance in the form dd/mm/yyyy would alleviate it.

In attempting to complete the form further difficulties and inconsistencies were found. An interactive calendar is made available at the press of a button. The majority of Web sites would post the calendar in a floating window, the congestion charge site shows it as a part of the main interface. This makes effective use of a part of the screen that is otherwise not used at all for any other purpose. This does lead to the question of making the calendar permanently available in this otherwise unused area, or even removing the three text fields and relying solely upon it for date selection.

Within the calendar some days are not selectable: any day in the past, weekends and bank holidays. Conventionally these would be shown greyed out to indicate that they cannot be selected. Confusingly on this calendar a range of different colour schemes are used for different categories of unavailable days. The major cue that a particular day can be selected is that cursor changes from an arrow to a pointing finger and the date is underlined, but mostly obscured by the cursor, as it traverses it.

Selecting a data on the calendar causes it to be posted into the text input mechanism. This could be accomplished without any need to communicate with the server, providing an immediate confirmation. The design implemented requires communication with the server, which is currently and can be predicted at peak times to be, unacceptably slow.

Returning to entering the date information from the keyboard, on occasion entering '2' for the month number resulted in a request to enter '02', whilst on later tests entering '8' was accepted. A fairly standard and obvious test, entering the date 29/02/2003, produced the less than helpful error message ' Invalid Date - Enter Date as dd/mm/yyyy'. Whilst correcting this 'error' the message remained continually visible leading to the situation where a valid date was displayed accompanied by a message stating that it was invalid.

Most messages from the system to the user are displayed in small red text against a blue grey background at the lower left of the interface. This is hardly the most noticeable place or format for them and when new users were observed attempting to operate the system many failed to notice them. To confound this problem further, on occasion developer diagnostic messages 'Session variable INTERACTION not found' appeared.

This part of the form was labelled 'Step 1 of 2'. It was not possible to evaluate the communication that ensued in step 2 as, over a period of five days, the server was continually reported as unavailable.

In summary:

1. Soak and capacity testing of a system should be completed in good time before it goes live. Less than 1 week before congestion charging starts, when the system could be considered live, response times are unacceptable.

2. Respect the users customisation of their environment. The browser preferences for text size are ignored by the system.

3. Prevent obvious errors from happening. The use of a pull down menu would reduce date/month confusion.

4. Make effective use of screen space. Having the calendar permanently displayed would make use of space that is not used for anything else and provide an alternative, less confusing and more natural input mechanism.

5. Indicate input formats clearly. The provision of a 'dd/mm/yyyy' indicator on the form would reduce the probability of input errors.

6. Provide clear and consistent feedback. Error messages offset from the center of attention in small font and low contrast colours are likely to be missed. Error messages are for the benefit of users not developers, so diagnostic errors should be removed.

7. Provide clear and consistent feedback. The use of a dialog to communicate with the user in some circumstances and an on-screen message in others makes it even more likely that the on-screen messages will not be attended to.

8. Error messages should indicate the cause and/or the cure for the error. Although 29/02/2003 is an 'Invalid date' explaining why would be a little more helpful.

9. Maintain consistency on the interface. Removing an error message from the interface once the user has attended to it prevents a possible visible contradiction.

10. Respect established conventions. The greying out of a component to indicate its unavailability is a widely accepted convention. The use of an underline to indicate active status is appropriate for a hyperlink but should always be visible, not transiently visible and partly obscured when the cursor passes over it.

11. Less is more. The use of a number of different colour schemes on the calendar to indicate different reasons for unavailability is confusing and unnecessary.

12. Provide timely feedback. The server round trip required by the calendar could and should be removed to provide faster user confirmation.

13. Selection is better than input. As the interface has sufficient space for the calendar component it could be used exclusively to select the required date.

My First Month - Lisa Halabi

I'd like to take this opportunity to introduce myself. I'm Lisa Halabi, the new addition to the happy gang that is called The Usability Company (TUC). I hark from a background in usability and ergonomics. I approached TUC already knowing Marty Carroll from our shared time spent on the Usability Professionals Association Committee and Catriona Campbell from various speaking events. It really is a very small field, but getting bigger by the day. It was obvious to me that TUC are going places and have a good reputation in the field.

After a brief orientation I find myself already working on a large Usability Evaluation for a major central government web site. Nothing like being dropped running then!

So, it's always a bit daunting when you start at a new company. It's the little things that can make you look really daft, like not knowing where the toilets are or how to operate the door's intercom system. This got me to thinking: it's a bit like usability really. Technology has the amazing ability to make users feel incapable or embarrassed, especially when trying something for the first time. Yet I'm always amazed at the ease with which people will blame themselves for not being able to do something. It is human nature to believe that we must be at fault or that we are in some way below average intelligence if something doesn't work like we expected. I have a mantra; there are no bad users, just badly designed interfaces. If a user can't easily navigate their way around or find what they're after quickly and efficiently, then the chances are the design is at fault, not them.

It's a good job I'm so passionate about making things simple and easy to use. I often find myself getting increasingly frustrated with new technologies, which just don't seem to work the way I expect them too. Often they're too complicated and too much bother to figure out. It just doesn't seem worth it sometimes. In our busy hectic lifestyles most people don't have the time to read the manual, look at the help pages or even go for training. Even if they did, who wants to! Wouldn't it be better if things were so intuitive and easy to use that they could be picked up as you go along? Good design is invisible, you don't notice it when it works, but when it doesn't you soon know about it. My job is to make technology invisible so you don't even notice how easy it was. Then I know I'm doing my job well. If you're not thinking about it, then it's worked.

So I'm happy to be aboard and look forward to working with some of you in the future.

Market Report

After a difficult January The Usability Company found February to be considerably more buoyant and we received an increase in enquiries and in overall interest in the company. We won a significant number of new accounts, some mentioned earlier in this newsletter and some won under strict NDA. However, in the course of winning new business there were the inevitable losses. As it is two years since The Usability Company was incorporated it seemed appropriate to share our thoughts.

Without doubt, since the company was founded usability has gained in recognition in the general market. Outside of usability practitioners there was very little recognition and understanding even of the meaning of usability. As a result we found ourselves spending the vast majority of our time with potential clients educating them about the process and benefits of usability. This created its own issues as it was unclear in many organisations just who would or should be responsible for usability or the customer experience.

More recently we have found companies contacting us, and asking us to help solve their problems of low conversion rate or a website simply not achieving its mission. We find our enquiries extend significantly beyond the area of Intranet and Internet, and although this has always represented a significant portion of our business the difference now is that the enquiries are coming not only from specialists within organisations, but also from business owners. There has certainly been a significant shift in understanding and as a result the market is improving in the face of a potential slowdown.

So why is this? Unlike two years ago, there are now far more cases of proven benefit as a result of employing usability within the development process. We were probably the first usability specialists to talk with any real conviction about return on investment and drive our clients to measure it. Now that tangible benefits have been realised organisations are far more willing to invest in existing channels to market knowing their investments will be returned 10 fold. If there are fewer sales to be made, as an organisation you need to make sure you are winning a larger proportion of them than your competitors. This is without doubt one of the reasons may of our clients still work under non-disclosure agreement; they see improved usability as providing them with a competitive advantage.

So what of the lost opportunities we have had? As with any business we lose business sometimes because we have misunderstood the brief, sometimes because the client is testing the market and has an incumbent supplier or for a variety of other reasons. Lately we have found ourselves forgetting that the market for usability is still embryonic. Recent activity from organisations that now understand usability has blinded us to those that still don't and we have sometimes not focussed on the need for education. This sees us losing business to market research companies because we haven't helped the client understand the difference. They evaluate us against criteria that we cannot meet, and were we to try would devalue our offer. The lesson for us is that the education must go on.

Click here to print