EXHIBIT

(New-economy) customer experience


Baldrigeplus.com exhibits are single-topic publications designed to illustrate and explain key Baldrige concepts. Used with our unique-to-this-site on-line workshop, worksheet, case study and Baldrige Program-sourced material (all available free on the baldrigeplus.com site), exhibits will add value to your performance excellence activities - whether award-oriented or not. The complete set - more than 100 individual documents, well over 1,000 pages - may be previewed and downloaded (from here) at no charge.

Copyright
If printing and distributing, please leave all logos and site branding intact
Great customer experiences develop your brand, increase loyalty and grow revenues. Many customers still find the internet an intimidating, foreign environment where “pages are a jungle of links, buttons, forms, blinking graphics, and baffling jargon. Accusatory error messages leap out at any time” (Ziff Davis Studios, White Paper One, 1999). But the good news, according to Jacob Nielsen (Alertbox, December 12, 1999) is that useability is now generally recognised as an important element of internet success.

More than a quarter of all attempted on-line purchases failed in the past twelve months, according to the Boston Consulting Group (USA Today, March 8th, 2000), and four out of five consumers experienced at least one failed purchase attempt. Why? Technical problems with the site, difficulties in finding products, and delivery problems after the sale, say BSG.

Last Valentines day, users of popular greetings cards sites found just getting in difficult, with site availability ranging from 43% to 73% during the day (Newsbytes.com, 16th February, 2000).

Customer experience and site useability is one of the top five dot-com quality issues (Our View, DCQ Outlook, April 2nd 2000). Questions you may want to ask yourself, your team, or the firm you're considering investing in, include: What's your customer experience like? How do you know? Is it getting better or worse? What about the customer experience of your principal competitors? What's industry best practice? Are there accepted useability benchmarks?” If they want some examples, point them towards Lands' End, where internet sales doubled in the first 11 months, largely due to innovations like 24/7 access to real customer reps. Or to Gap, where shoppers can, for example, compare 'classic' and 'relaxed' versions of the same product side by side.

And customers? If 'quality is what your customers tell you it is,' what are they likely to tell you about their on-line experiences? Not the truth – or at least, not truth you can use. Why?

Because what people say they want, and how they behave when on-line, may be quite different things (according to Jacob Nielsen – sources below). In a focus-group setting, people may get excited about fancy effects, but ignore or find them a nuisance when on-line. Self-reported survey data are three steps removed from the truth: users are inclined to tell you what they think you want to hear; their recollection will likely be imperfect; and there may be a big gap between what they believe they did, and what they actually did.

Surveys are just opinion polls, and likely to be highly unreliable, says Jacobsen. It's been accepted (from survey data) for some time that 28% of users find it somewhat or extremely difficult to locate products on the web – but actual observations show people succeed less than half the time.

Useability panels? Hazardous, for three reasons: first, paid panel members may not be representative of your customers; second, lab-checking doesn't always replicate real-world tasks – like booking airline tickets or finding arcane information; and third, self-reported behaviour may not match what actually happens.

Counting click-through and other automated data? Waste of time, because it tells you nothing about the quality (or effectiveness, or success) of the experience. Automated useability may even be dangerous – it doesn't measure experience, and its availability may divert attention from more valuable data-gathering.

What can usefully be automated?
Response times are important, says Nielsen. You've only got seconds to grad a user. Downloads that arrive like treacle will be dumped before the first elements are on screen. But don't invest in expensive response time measurement services or software, just get the CEO to download the home page from a hotel room next time she travels! You'll hear about treacle.
HTML validation. Clean code makes for smooth and trouble-free site use.
Linkrot. But beware – if URLs are reused (for latest versions of regularly updated content, for example), software checks may report a working link, but the user will get something they weren't looking for. When moving pages, always check that links stay alive, point to what users expect.
Accessibility for users with disabilities should be a given, but may be hard to test – look for specific tests by interest groups?

Nielsen's conclusion – the only valid way to assess useability is to watch real users, on your site, accomplish real tasks. The simplest test. And the cheapest.

Customer experience and New Product Introduction
In the off-line world, NPI is a heavy-duty issue (for a best practice example, go to www.dot-com-quality/solar_turbinesNPI.html). In the immature and highly innovative dot-com world, where speed is everything, it's often a case of throw it at the wall and see if it sticks (and fix the problems in the inevitable re-design).

Sure, speed's important, says Nielsen, “but it is not everything. Customer satisfaction is everything. Launching a bad site with poor usability is a guaranteed way to waste money since it will have to be redesigned more or less immediately.”

Don't burn off your early users, says Nielsen, they'll be lost for good, they'll tell their friends, and your viral marketing will turn malarial.

What's more, getting your site right may not slow your time to market. How do you get on-line with great useability, without endlessly messing around?

Test with only a few users, and test what really matters. The first few users will tell you most of the useability information you need.

Test early – start testing as soon as you have something to test. It costs 100 times as much to make a change in version 2 than it does to get it right in v1. There's no physical media involved, and you don't have to ship CD-ROMs to patch mistakes, so fixing after launch is not so expensive, but on-line customers are harder, and probably much more expensive, to win back.

Customer experience dos and don'ts
At the completion of each sale (says Greg Martin of Evpage Communications), display a screen that says 'thank you,' and don't spoil the effect by advertising on the thanks screen.

Do not force open new browser windows, or require that your customer view new web pages through a frame, “how many times you would return to a restaurant in which the valet parking attendant painted an advertisement for the restaurant on your windshield,” Greg asks.

Do offer a support and/or frequently asked questions page, easily accessible from your main page. If possible, offer free phone support; make it possible for customers to call a real person; offer to notify them of new products – by email or phone, and personalize statements and correspondence – even email.

Footnotes
Material referred to during the compilation of this Note include: Study finds e-tail problems, by Lorrie Grant in USA Today, 8th March, 2000; Dissapointing Valentine's Day on-line delivery times, by Sherman Fridman of Newsbytes.com, 16th February, 2000; Jacob Nielsen's Alertbox for December 12th 1999 (Voodoo Useability) and April 2nd, 2000 (The Mud-Throwing Theory of Usability); Ziff Davis Studios, White Paper One, 1999.