The Human Touch

The Human Touch

Australia's chief information officers should tack onto their IT budget for 2000 the funds to cover purchase of a Spice Girls CD if they want to ensure their information systems achieve their full potential.

They should then play one song to their IT staff just as they start to specify new business applications. It's the one that goes:

"Stop right there, thank you very much,

I need somebody with a human touch."

Time and again business applications are constructed without the human touch; without regard for the people who will actually have to use the systems once they are implemented. It's a costly mistake, especially as more and more businesses are interacting directly with their customers (humans presumably) via online information systems. In the past information systems were developed mainly for employed staff, who could be trained and then forced to use even the most user-unfriendly green-screen applications. But now organisations are reaching out to consumers via the Internet with applications that must make humans want to use them.

It is a challenge. A survey last year by US-based Zona Research found that 62 per cent of even experienced Internet users had on occasion abandoned their online search for a product or service because of the difficulties associated with finding the required information online. Put simply, potential revenues are walking out the door, because the information systems make it too hard for the buyer to do business with the seller.

It is something which the US banks have experienced first hand, according to Cyber Dialog, a US company which specialises in online market research. A survey in 1999 of US online bank users found that although the absolute number of users grew 100,000, to 6.3 million, 3.1 million US adults had discontinued their use of online banking because of dissatisfaction with the service. Tellingly, only 35 per cent of those 3.1 million said they might be inclined to try it again. The problem, according to Cyber Dialog's manager of finance strategies, Michael Weiksner, was that the services were too complex, or the site did not provide enough customer service.

In comparison, 85 per cent of Americans trading shares online are satisfied with that online experience, and by July 1999 6.1 US citizens played the sharemarket online. "Online brokerages like E*Trade and Schwab have demonstrated real leadership by investing aggressively in marketing and customer service," Weiksner says. "Banks must react in Internet time or risk losing the banking relationships of their most valuable customers." If some companies have such scant regard for the information needs of their online customers, then it is hardly surprising that their own employees also have to battle with the information systems that are supposed to help them do their job.

Dr Gitte Lindgaard ran the human computer interaction laboratory at Swinburne University of Technology in Melbourne until recently, when she accepted a similar position in Ottawa, Canada. She believes that the lack of focus on usability is one reason why companies have so often struggled to successfully implement total quality management, or enterprise resource planning systems or business process re-engineering. According to Lindgaard, there has been too little understanding and acceptance of the link between human performance and company performance. It was a message she endeavoured to get across to telecommunications giant Telstra, when she ran a usability program for the Telstra Research Laboratories during the 1990s.

Lindgaard says that were corporations to fully account for and analyse the costs of new business applications, they would find that "80 per cent of the costs associated with the system come after its release because there is poor, or no, user analysis". All too often IT departments would expect productivity increases as a result of new systems being installed. However, these same departments would have implemented no performance measures, they had no performance goals, they had no idea of what users actually did, nor had they a full grasp of what tasks users had to perform.

Dr Lindgaard recounts one example where she wanted to identify for a client exactly how much time and money was being lost because information systems had been designed poorly and actually made it harder for an employee to complete a specific task. "We worked out that an operator worked 90,000 minutes a year. Eighty-five minutes of each day were wasted per operator," because of the struggle with a computer-based information system ostensibly developed to help the operator deal with a customer enquiry.

Over a year that represented 18,700 wasted minutes of each operator's time. There were 2500 operators, so the total cost of the system's inefficiency came to $12.98 million in terms of wasted salary alone. On top of that was an additional $1.6 million communications bill, $3.7 million of technical costs associated with getting the system tweaked, plus the $10.l4 million cost of the long training course required to help operators understand the basics of the system. Finally, the company spent about $500,000 running a help desk for employees to call when they still couldn't work out how to use the application.

It amounted to almost $30 million of waste. Armed with these figures, Dr Lindgaard secured management's attention. Her study by comparison cost $90,000 and identified minor systems modifications which reduced the waste of the system by 50 per cent. But the company was already fixing up a brand new system, which in a more perfect world would have worked properly the first time; if it had factored in the human element.

Does This Thing Fly?

Computer Human Interaction (CHI) professionals first emerged as "human-machine interaction" experts who helped make aeroplanes somewhat easier to fly. A jumbo jet cockpit is not the most intuitive location in which the human being works. CHI professionals helped refine cockpits and training systems so that pilots would not send planes falling from the skies. Today the challenge for CHI specialists is to keep business systems flying. There is a particular emphasis emerging on the usability of those systems which have been developed to allow ordinary consumers to interact with corporations' information systems.

As the NSW general manager of the Hiser Group, Susan Wolfe spends a great deal of her time consulting to companies which are going online. As she noted in a presentation to the Australian CHI conference in Wagga late last year, "our biggest challenge comes from the fact that the users' expectations have changed. It is no longer rocket scientists interacting with rocket-science applications. It is now you and I doing everyday tasks and having every expectation that they will be simple and obvious to complete. Unfortunately, this is often in conflict with the organisation's ability to deliver."

Wolfe's concerns are borne out by a survey conducted last September by the Australian Competition and Consumer Commission, which randomly visited 250 Australian Web sites and found that few provided even the most basic consumer information to their prospective online clients. Three out of five sites did not provide information regarding refund or exchange policies; 75 per cent had no privacy policy, nor did they disclose how they would deal with a consumer's personal information. A full 25 per cent of the Web sites did not even disclose the business address of the company.

"If consumers are to have confidence when shopping 'online' they must have clear and accurate information about the business they are dealing with, the terms and conditions involved, procedures for redress if something goes wrong and how the trader will deal with their personal information," the ACCC's deputy chairman Allan Asher says.

But the ACCC revealed more than just oversight; rather it uncovered a degree of blatant stupidity from companies hoping to trade online. Although the Internet is inherently global, and therefore consumers can reside anywhere, one in four Web sites surveyed by the ACCC did not even state what currency to use for a payment. This leaves a potential overseas customer floundering as to whether $ meant Australian dollars - or greenbacks, or Singapore dollars or Canadian dollars.

Such carelessness costs a company consumers, and revenues, according to Wolfe. She claims that companies need to give more thought to the psychology of people when designing applications. "How people think, perceive, read, and remember," she says, is important to the way people design successful computing applications.

Hiser's education and marketing manager, Catherine Rosenbrauer, believes that part of the problem is that when companies are designing a computer system they have several groups involved which might all have conflicting agendas.

For example, the IT group might want an elegant computing solution; the business might want a system to improve productivity; and the employee wants something easy and intuitive to use, which allows them to help the customer. The end consumer wants something entirely different. For a successful application, however, the needs of all must be considered. "You need a collaborative approach," Rosenbrauer says.

The Job's the Thing

Ted Briggs is the head of the usability team at the Australian Bureau of Statistics. He believes his role can be simply defined as helping people do their job. "The thing we tell people is that technology is no good unless it helps people to do their jobs," Briggs says.

The ABS' interest in usability was piqued several years ago when it developed several large applications, which were not well received internally. At the same time, the bureau recognised that in the future it wanted to develop a computer-based interface for outside users. "Those combined to make people think we needed to make the systems more user-friendly," Briggs says.

Before taking on the usability role, Briggs ran the ABS' help desk for five years, a career progression he describes as natural. "I remember seeing one senior person in the ABS nearly in tears, because something was not working, and I thought: Why is this designed this way? Systems were just not supporting people's workflow and doing things that were predictable."

Or sensible; Briggs recalls one scripted phone application which was so rigid that an operator could not put in information logically, and as it might arise during a conversation. To ease the flow, operators were reduced to scrawling information on post-it notes, then re-keying the data later on, introducing inefficiencies and allowing errors to creep into the information collection process.

Over the last two and a half years that the ABS has had a dedicated usability team, things have improved, Briggs says. "We have quite a lot of management commitment to usability, but we still get the 'we can't afford it' and we have got to chip away at that."

Dr Lindgaard's main argument for investing in usability is that companies cannot afford not to do it, in order to remain competitive. She cites a Scandinavian case study of an insurance company that developed a more user-intuitive application and claimed a 300 per cent profit improvement in three years, which it credited to the improved information systems. Dr Lindgaard says that an Australian pager service company fine-tuned the information systems used by its operators to become more intuitive and reduced the average 30- second call to 29 seconds. Over a year that saved a total of 1666 hours across all its operators - a productivity improvement which can sail directly to the bottom line.

"Usability is all about people and communication and cooperation. It is about relating to people and understanding people's jobs," she adds. Some technologies such as speech recognition may help create 'friendlier' interfaces, but the underlying application still needs to match the workflow.

"With user-centred design you need to get all the players involved," says the Hiser Group's Susan Wolfe. She worries that too many companies which are developing applications for e-commerce are focused on setting the strategic direction and not really concentrating on the needs of the underlying business. "You need to strike a balance between the user and the business needs," she says. "The back-end processes already in place frequently cannot support the desired front-end experience. Organisations often rush to get their e-business applications deployed first, and then tackle retrofitting their back-end processes after that."

It is not hard to see that with this approach to systems design the e-consumer - the human in the information maze - is often being uncomfortably shoehorned into the application. It's entirely the wrong way to approach the problem, Wolfe says.

When working on a project with a client, Wolfe says they first identify the type of customer expected and who the application is being developed for. Then a dialogue is established between the usability team and the company's marketing department. At that stage a group of people who match those expectations are recruited as a test market and asked about their requirements. The usability team then designs several scenarios which might lead to the application being accessed.

For example: a scenario might be constructed describing how a small businessman might use the service, and how that might change if he were an experienced or an inexperienced computer user. Armed with these scenarios the client is then approached again and asked the question: "Do we all agree this is what we are building?" Wolfe says. Nothing has yet been committed to code. All that has been defined are the usability goals, which are then prioritised. At this stage IT is brought into the loop, and the first design of the systems commences - although it's only on paper.

Throughout the design process the users are involved, in what Wolfe maintains must be a collaborative activity. "We ask the users to refine the design," she says, with a focus on the information structure. Getting that right is what will ultimately decide whether a system passes the usability tests. Wolfe regrets that too many companies embarking on e-commerce focus on getting an attractive presentation for people to look at, when, in fact, "that's one fifteenth of what is important".

This iterative process of design, market test, refine, market test and so on, eventually delivers a paper roadmap of what the final system needs to do. Only then can the actual coding commence. It is, Wolfe says, a radically different business than old-style applications development, where the IT staff were writing applications which would be used probably by IT staff. "In the past, the old systems were developed for expert users. That is not the case today."

Technically brilliant applications which stretch the boundaries of computing have a role, but it's probably not at the consumer coalface. Cue Spice Girls:

"It's cool but you don't even know me. You take an inch, I run a mile."

And when e-consumers run, they're often gone for good.

Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

More about ABS AustraliaAustralian Bureau of StatisticsAustralian Competition and Consumer CommissionAustralian Competition and Consumer CommissionE*TradeSchwabSwinburne UniversitySwinburne University of TechnologySwinburne University of TechnologyTelstra CorporationZona Research

Show Comments