One of the oldest hypertext usability principles is to offer a visual representation of the information space in order to help users understand where they can go. Site maps can provide such a visualization, offering a useful supplement to the primary navigation features on a website or intranet.

A site map's main benefit is to give users an overview of the site's areas in a single glance. It does this by dedicating an entire page to a visualization of the information architecture (IA). If designed well, this overview can include several levels of hierarchy, and yet not be so big that users lose their grasp of the map as a whole.

We define a site map as a page intended to act as a website guide. The site maps we studied took a variety of forms, including alphabetical site indexes, dynamic diagrams, and two-dimensional lists. The term "site map" here thus encompasses a wide array of features, appearances, and names, including "guide," "overview," "index," and "directory."

Two Research Studies

A total of 30 users participated in our site map testing, with 15 in each of the two research rounds.

We tested the following 20 websites, which included a mix of e-commerce and marketing-oriented sites, high-tech companies, B2B sites, content sites, non-profit organizations, and government agencies.

Sites Tested In Study 1	Sites Tested In Study 2
CDNOW (e-commerce) Documentum (high-tech product) Interwoven (high-tech product) Mercedes Benz USA (marketing site for cars) Museum of Modern Art (non-profit) New Jersey Transit (local transportation) Novell (B2B) Salon (online magazine) Siemens Medical Solutions (B2B) United States Treasury Department (government)	Administration on Aging (government) BMW USA (marketing site for cars) Citysearch Boston (visitor info) Harvard Pilgrim (health insurance) The Knot (wedding information/e-commerce) Marriott (hotels, with online booking) Scholastic (children's books) Texas Roadhouse (restaurant chain) TiVo (high-tech product)

In both studies, we first took users to a site's homepage and gave them a task without any special mention of the site map. This part of the research assessed the extent to which users naturally turn to site maps. Later in each study, we specifically asked users to go to the site map if they hadn't already gone there on their own.

Study 1 was conducted 7 years ago. Comparing the two studies thus allows us to assess long-term trends in site map usability.

Site Maps are Used Rarely

The good news is that users can actually find the site map in those few cases where they want to. In Study 2, 67% of the users successfully found the site map when we asked them to "Find a page that lists every part of the website."

Keep It Simple

• Call it "Site Map" and use this label to consistently link to the site map throughout the site.
• Use a static design. Don't offer users interactive site map widgets. The site map should give users a quick visualization without requiring further interaction (except scrolling, if necessary).

A site map is, after all, a map; it should not be a navigational challenge of its own.

As we've found repeatedly, users hate non-standard user interfaces that force them to learn a special way of doing things for the sake of a single website. Site maps should be simple, compact layouts of links, and they should show everything in a single view.

The one small complexity we recommend is to use a multi-column layout. In Study 2, users easily succeeded with 61% of tasks involving multi-column site maps compared to 47% of tasks with single-column site maps.

Multi-column site maps worked better because users needed less scrolling to get an overview of the site's structure. People were more likely to become lost within long, scrolling site maps. They typically scrolled up and down the map multiple times, often accidentally or purposefully skipping content. In fact, users often started with one quick scan of high-level categories, then scrolled back up and did a more detailed search, sometimes repeating this process multiple times with more and more focus each time. In contrast, multi-column site maps made it easier for users to quickly glance at all categories and subcategories, and thereby get a lay of the land before digging deeper.

Why Have a Site Map?

Despite the prevalence of good site maps these days, users don't use them very much. So why bother making a site map for your website? Because it can help users understand your site and what it offers.

I still recommend site maps because they're the only feature that gives users a true overview of everything on a site. One could argue that a site's navigation serves the same purpose. For example, some navigation offers drop-down menus that let users see the options available in each site section. But even with these menus, users can see only one section of content at a time.

A site map lets users see all available content areas on one page, and gives them instant access to those site pages. Site maps can also help users find information on a cluttered site, providing a clean, simple view of the user interface and the available content. Site maps are not a cure-all, however. No site map can fix problems inherent in a site's structure, such as poor navigational organization, poorly named sections, or poorly coordinated subsites.

If site maps required a major investment to design, they wouldn't offer sufficient ROI to be worth doing. But because all of our guidelines call for site map simplicity, making a good one doesn't require a lot of work, and it will help some of your users. More importantly, it will help users at a critical time: When they are lost and might abandon your site if they don't get that last piece of assistance to find their way around.

Site maps are a secondary navigation feature — a humble role that they share with breadcrumbs. Indeed, the arguments in favor of site maps are the same as the arguments for breadcrumbs:

• They don't hurt people who don't use them.
• They do help a few people.
• They incur very little cost.

Learn More

Full-day course on navigation design at the User Experience 2008 conference in Chicago and Amsterdam.

Visit the site to read the full alertbox | Feed made by Bruno Torres

The contents of this feed is Copyright © 2006 by Jakob Nielsen

Ad: Hosted by Dreamhost (Sign Yearly L1 plan with 50% off using the promo code BRUNOTORRES)

Web portals have suffered a highly variable existence. Every few years, they're in, and every few years, they're out, with many of last season's darlings filing for bankruptcy or being snapped up on the cheap. It's a different story inside companies: enterprise portals know only one way, and it's up. More and more companies are establishing intranet portals, and they keep improving their features and usability.

It's been 3 years since our last assessment of intranet portal usability. High time for an update. This time, we collected case studies from 23 companies and organizations. The new data supplements the information from the 25 companies in the report's 2 previous editions. Our current intranet portal recommendations are therefore based on the collective experience of 48 companies and organizations over the 5 years since we began our initial portal usability research.

The first new finding is that all 62 previous findings continue to hold. Although much has changed — and we have many new findings (for a total of 117 best practices in the new report) — things don't change much in terms of best practices for user experience. The technology changes and vendors produce ever-more dot-releases, but usability issues move much more slowly because they're based on human characteristics.

For example, we again found that role-based personalization is the way to go. People very rarely use corporate portal customizations, however much they ask for them. (Yet another great example of why you shouldn't listen to what users say.)

An interesting exception here is with university portals, where many users do engage with customization features. Why? Possibly because university staff has a tendency to tinker and to value exploration for its own sake.

Growing Portal Maturity, But Many Newcomers

Despite my vendor-neutrality, I can't resist quoting portal vendor BEA, which stated that the portals market is "positively popping" and projected an estimated $1.4 billion in annual sales in 2011. While I can't speak to the sales prediction, I can confirm that we saw significant growth in portals uptake.

We first studied intranet portals 5 years ago, and the idea is certainly older than that. Even so, many company intranets are only now sufficiently mature that teams can begin turning them into full-featured portals. Some of the bigger and more established portals we studied have reached very impressive levels, but the companies are nonetheless continuing to evolve them as they try to reach single sign-on nirvana (see below) and add new collaboration features.

Our knowledge of intranet portals has expanded along with their increasing maturity over the years. Using the report thickness as a primitive metric for the amount of information we've collected, the page count has grown from 104 pages five years ago to 343 pages now. That's an annualized growth rate of 27% in knowledge about intranet portals and their usability.

From Turf Wars to Cross-Functional Governance

Most companies have embraced cross-functional teams or steering committees as a way to ensure buy-in across departments. This softer approach to portal governance is much more successful than having a (seemingly) arrogant intranet group in the IT department ram a portal down the other departments' collective throat.

Still, successful portal projects can't be run solely by a loosey-goosey assembly of well-intentioned people from across the organization. The portal must be somebody's job. In bigger organizations, a full-time job. Even in smaller places, however, specific individuals must be responsible for the portal as part of their official job description.

It's especially important to realize that an intranet portal is not a one-time project that's finished once it launches. The people in charge of the portal need to stay on the job after launch, or the intranet will suffer portal decay. Ongoing, dedicated resources are required both to integrate new features and maintain the quality of existing features such as search. It's amazing how quickly search quality degrades if there's not a continued push for good headlines and good intranet IA practices.

Single Sign-On: Still Elusive

In our initial research 5 years ago, it was already clear that single sign-on could dramatically improve user productivity and satisfaction, as well as immensely reduce support costs. (A huge proportion of help desk calls relate to password problems.) At the time, single sign-on was more of a hope than a practical possibility.

Our second round of research confirmed single sign-on's potential — and its elusiveness.

True single sign-on (SSO) was and is extraordinarily rare, as our third round of research shows. We can only conclude that it's very difficult to achieve, despite its promise. That said, we're starting to see an interesting, pragmatic approach to what Kaiser Permanente calls "reduced sign-on." Work as hard as you can to reduce the number of authentication requests users encounter each day, even if you can't get it down to 1. You can also reduce frustration by informing users about the impending login request before they click the link that activates the demand.

News and Collaboration: Old and New Portal Drivers

On portals, news has two distinct roles:

• Unifying force: It ensures that all employees are informed and receive a consistent message.
• Narrowcasting: It aggregates and distributes specialized news so that each user gets a filtered view with just the information he or she needs.

In any case, intranet portal collaboration features tend to be business-oriented and benefit from the accountability that's inherent in having all contributions attributed to an employee's real name. Flames are less severe when you sign your postings and address them to people you have to face in the cafeteria the next day.

Rather than being swayed by fashionable Web 2.0 trends, our case studies indicate the need to make a business case for "Enterprise 2.0" features. Many such features are actually more useful on intranets than on the open Internet. However, you must first ensure that you have a governance structure and rules in place, and that you know the real business value of any planned features. If you can't identify the business case, you're better off focusing on further improvements to the old features you already have.

User-Informed Design

The problem is that teams often use surveys (which record what people say) more than testing (which shows what people actually do). Teams that have tried user testing for their portal projects have become strong advocates for the method and report high value from their tests. Sadly, becoming convinced about user testing's value typically requires a team to actually conduct a test. Having me say that it's important is nowhere as good a motivator as personal experience watching users. Although this is obviously a bit chicken-and-egg, we are seeing more and more intranet portals being subjected to user testing.

ROI Under-Documented

While smaller companies are likely to realize smaller savings, they should still estimate ROI. They can do this at any desired level of rigor; as good as it is, we don't want to hold up Six Sigma as the standard everyone should conform to. It's perfectly feasible to estimate portal productivity improvements using the simpler method we applied to our study of intranet usability in general:

1. Define a number of core employee tasks.
2. Find out how frequently people perform these tasks.
3. Find the loaded hourly cost of an average employee in your company. (Or, for a more advanced approach, segment employees by major job categories and run this analysis for each segment, using average costs for people in that segment, as well as their frequency of use.)
4. Observe and time people as they perform the defined tasks with your current design. For timing, a simple stopwatch will suffice; you don't need special equipment or a fancy usability lab. Indeed, we often collect benchmark metrics for clients by testing in a small conference room.
5. Multiply the following numbers: time on task, each task's frequency, the employees' hourly rate, and the number of intranet users. The result is how much it costs the company to have employees accomplish the tasks using the current design.
6. Adjust this cost estimate to account for the tasks you didn't measure. For example, if you measured 1/3 of the core tasks employees do, you should multiply your measured numbers by 3 to get a decent estimate for all tasks. This, of course, assumes that you didn't focus the testing on the intranet's best-supported or best-designed areas, but rather on a representative and fairly chosen sample of tasks.

Well, this is what I want you to do. With the exception of Dell, the teams we studied justified their portal projects through softer means, such as improved user satisfaction and increased usage.

Next-generation portals increasingly emphasize collaboration features, and measuring community activity forms another argument that features are being appreciated across the company. Many organizations also view improved information access as a key goal. While it's certainly possible to measure changes in employee awareness of corporate information, portal teams currently tend to take a qualitative approach to assessing knowledge dissemination as well.

Indeed, many teams view their portal as such an obvious improvement over the disorganized intranet that preceded it that measuring formal ROI seems like a waste of time. The good news? They're often right. We've seen some failed portals projects — and there are definitely many pitfalls to avoid — but a good intranet portal adds very clear value.

Learn More

Two-day tutorial on Intranet Usabilityat the User Experience 2008 conference in Chicago and Amsterdam.

Visit the site to read the full alertbox | Feed made by Bruno Torres

The contents of this feed is Copyright © 2006 by Jakob Nielsen

Ad: Hosted by Dreamhost (Sign Yearly L1 plan with 50% off using the promo code BRUNOTORRES)

The winners of the award for 10 best-designed intranets for 2008 are:

• Bank of America, US
• Bankinter S.A., Spain
• Barnes & Noble, US
• British Airways, UK
• Campbell Soup Company, US
• Coldwell Banker Real Estate Corporation, US
• IKEA North America Service, LLC, US
• Ministry of Transport, New Zealand
• New South Wales Department of Primary Industries, Australia
• SAP AG, Germany

Half of the winners are from the US, closely matching the nation's long-term performance average of 53%. The remaining five winners hail from five different countries. The southern hemisphere is strongly represented this year, including the first-ever winner from New Zealand. Australia has had many winners over the years, as have the UK and Germany. Spain seems to be an up-and-coming country in terms of quality intranets, collecting its third award this year (earlier awards went to Amadeus Global Travel Distribution in 2003 and Banco Español de Crédito [Banesto] in 2005).

In terms of industry sectors, the financial sector is strongly represented with three winners (two banks and one real estate company). This, too, follows tradition: earlier design annuals have typically had a disproportional number of winners from the financial sector (2007 was an aberration, with only one financial winner, JPMorgan Chase).

Financial institutions probably have disproportionally good intranets for two reasons:

• The companies tend to be big and have a lot of money resting on optimal performance. They therefore invest more heavily in IT than companies in most other sectors.
• The companies typically have a long tradition — often going back a decade or more — of taking usability seriously. After all, home banking is doomed unless the user experience is exceptionally approachable. Similarly, the financial sector deals in complex transactions, and training costs for internal applications can skyrocket if the design team fails to truly understand the needs of users in both local workgroups and remote branches. Intranets clearly benefit from the financial sector's above-average attention to user-centered design.

Of course, it might be increasingly unreasonable to view technology companies as more sophisticated than other industries in the use of technology. This year, for example, the retail sector shines with two winners, beating its average performance as it becomes ever more tech-driven. Barnes & Noble might be a bookstore, for example, but it's also a leading e-commerce site. And, as the case study of the B&N intranet shows, the company certainly uses technology to the max.

As with the previous two years, most of this year's winners are big companies, employing an average of 50,000 employees. Still, this year's list includes the first small organization since 2005: New Zealand's Ministry of Transport, which has only 200 intranet users. Once again, we have proof that size isn't everything, and that a small but well-focused effort can produce a great intranet.

Company News

The Campbell Soup Company has an interesting approach to news. Its intranet features an area across the top of the homepage with cells that represent the main business units and corporate functions, letting all users browse each unit's news. While this would be too much functionality for a smaller company's intranet, it can help manage the complexities of bigger companies — especially ones like Campbell that contain multiple strong brands.

Barnes & Noble spends almost all of its homepage on various news areas. Among them is a large central area, Barnes & Noble Today, that features stories from different stores and regions and helps build community among the widely dispersed bookstore staff. A smaller area lists Store Alerts that contain practical information, ranging from when a runaway bestseller will be in stock to recommendations for the gift fixture in bookstore cafés. The intranet team is committed to posting the Store Alerts by a certain time each day, which is critical to its success. News that isn't new gets old quickly, which is why a B&N-like commitment to intranet news is necessary for it to truly attract traffic and interest employees.

Intranet multimedia use has been growing steadily during the last few years, and reached a new high this year. SAP dedicates an entire homepage section and intranet section to SAP TV, with videos on topics ranging from the SAP Cup soccer finals to doing business in Russia.

Increasing Quality and Polish

Take, for example, the staff directory. This is a feature of virtually all intranets, and its basic design guidelines are well known. To go beyond the basics, Coldwell Banker emphasizes finding employees by geography — an enhancement that completely makes sense for a real estate company. Coldwell Banker's employee finder also includes a special Referral feature to help agents find colleagues for referral purposes. The directory's structure of tabs and labels emerged from usability research that revealed how company users think about seeking out their colleagues.

Most of the winning intranets have strong support for single sign-on, which we know from our studies of intranet portals is a strong determinant of employee satisfaction and productivity — and yet something that's hard to achieve and rarely works as well as promised. This is definitely an area where the extra work to get it right improves the experience (and productivity) of users every day.

In general, integration was a strong theme: the impetus for many a winning redesign was to create a single intranet information architecture (IA) with a consistent navigation scheme and consistent user experience. Such redesigns typically replaced hundreds of individual sites that lacked unified navigation and presented a highly inconsistent user experience. This chaotic state remains the norm on many intranets, which suffer reduced usability and lowered employee productivity as a result. In contrast, the new intranets are invariably based on content management systems with special interfaces designed to increase usability for content providers. The CMS interfaces also strongly emphasize the importance of sticking to a few templates with a standardized design.

In an effort to preserve their autonomy, individual departments sometimes resist the move toward a single, unified intranet design. The Campbell Soup Company provides a striking counter-example here. The company's employees are strongly attached to the specific brand they work for, which would seemingly doom any attempts to unify the intranet. However, the design team wisely provided ways for the intranet design to reinforce users' brand connections through customization and "skinning" (changing the appearance to reflect, for example, branded color schemes).

At Coldwell Banker, the company's franchise model resulted in the introduction of "co-mingling" rules in the CMS to blend corporate-level content with content from the local (franchise or office) level. This approach requires some added features, such as an extra CMS field for tagging the importance of each piece of corporate content so it could be appropriately prioritized relative to local content. However, the approach also creates a vastly more consistent (and thus productive) user experience than the free-for-all that's found in many other companies.

Updated content is a major contributor to intranet quality. The Department of Primary Industries brings two related trends together here. First, it increases usability for content providers to the limit, providing a tool that lets all users submit news items. This, of course, ties in with the second trend — to prioritize company news — because editors need a steady stream of news items from all organizational areas if they're to keep the news fresh and thus engaging.

Productivity Focus

While advanced applications can boost productivity, smaller tools sometimes do the trick. For example, at the Department of Primary Industries, employees (such as mine safety inspectors) often need a department vehicle for offsite assignments. Historically, employees booked vehicles by calling their location's receptionist, who took the information over the phone and entered the details into a fleet management system. Now, a simple, one-page form serves the same purpose, saving time and increasing booking accuracy. (The same intranet also contains an advanced application for geospatial visualization, helping employees manage emergencies like floods, droughts, and exotic disease outbreaks. Thus proving the point that sometimes high tech is appropriate, and sometime it isn't.)

SAP's intranet offers an interesting twist on the productivity focus, providing special personalized pages for the company's Executive Board members. Top-level executives are busy, and their time is expensive, yet they're often among the least-trained users as their duties are focused elsewhere. It thus makes perfect sense to dedicate special attention to improving usability for this small but important group of users.

Knowledge Management

Bankinter probably had the most impressive expertise finder, with a profusion of graphical tools for visualizing the distribution and location of knowledge across employees.

Diverse Technology Platforms

The most-used products were SharePoint and the Google Search Appliance. Other frequently used products were Red Hat Linux, Lotus Notes and Domino, and Oracle databases.

No single product made the list of most-used products for all of the four most recent Design Annuals (2005-2008). This simple fact reinforces the point that intranet platforms still have a long way to go. That said, the following products made the most-used lists more than once during this four-year period:

• 3 of 4 years: Google Search Appliance, Microsoft SQL Server
• 2 of 4 years: Apache, Documentum, IBM WebSphere, Java 2 Enterprise Edition (J2EE), Lotus Notes and Domino, Oracle databases, SharePoint

Intranet Branding Gets a Lighter Touch

This year's winners again include generic intranets and several traditionally branded intranets, with names like Discover, InSite, and Flagscape. However, we also have a prominent showing for a third option: give the intranet a name, but one that's not a strong brand in itself. Such names can borrow strength from the organization's main brand and typically include a plainspoken description of the intranet's function. These names include Employee Self Service, my Campbell, Coldwell Banker Works, and US Retail Inside.

Whether or not an intranet is branded, it needs internal marketing to familiarize employees with its many features and new areas (particularly if it's a big intranet). Several winners this year had attractive and effective ways of promoting intranet features.

Intranet Trends

• Increased personalization
• Integration of information sources, often resulting in a single "one-stop shopping" page
• Emphasis on mission-critical applications and information (such as sales targets)
• Improved event and project calendars
• Special sections to help orient new employees
• Prominent display of stock quotes and other financial information
• Integration of external and company news, often in the form of customizable feeds
• Integration of alerts with the main intranet to inform users of important messages
• Redesigned and improved search features, which often went from horrible to good and generated ecstatic user feedback

ROI for Intranet Usability

Bank of America collected detailed measurements of the time required to navigate from the homepage to 11 different intranet destinations. After the redesign, the average time decreased from 43.6 seconds to 21.7 seconds, cutting the navigation time in half. This corresponds to a 101% increase in post-redesign productivity, because users can get slightly more than twice as much done in the same time.

In our research testing a large number of intranets, we found that the productivity gains from a major improvement in intranet usability were likely to be 72% on average. How does this research result square with the 101% increase measured at Bank of America? The difference in findings is easily resolved with two observations:

• The 72% productivity increase is an average; some redesigns will experience smaller improvements, others higher.
• Bank of America has an award-winning intranet, so it stands to reason that its ROI will be higher than average.

At Campbell, the number of intranet visits per day increased by 727% after the redesign, but the number of actual pages viewed per visit decreased from 9.12 to 1.43. As a result, the grand total number of page views only increased by 30%. Overly simplistic use of Web analytics might focus on this latter number and conclude that the redesign had been only a modest success. On the contrary, it's wildly successful, as the two other statistics show:

• Employees get many more things done with the intranet, as shown by the 727% increase in visits.
• Employees are much more efficient each time they visit the intranet, corresponding to increased task productivity. Ideally, productivity is measured as time on task, which directly translates into the amount of work done per hour. We can't quite claim that the decrease in pages per visit corresponds to a 538% productivity increase because people might spend more time on each page. Still, there's no doubt that intranet efficiency — and thus employee productivity — increased immensely after the redesign consolidated, on personalized homepages, the information that users need.

Full Report

Learn More

Visit the site to read the full alertbox | Feed made by Bruno Torres

The contents of this feed is Copyright © 2006 by Jakob Nielsen

Ad: Hosted by Dreamhost (Sign Yearly L1 plan with 50% off using the promo code BRUNOTORRES)

Information architecture (IA) poses a tremendous challenge in designing any navigational system. Historically, intranets have had little in terms of systematic IA efforts; designers typically "structured" intranets according to the organic growth of pages and features provided by different departments. Employees suffered the consequences, repeatedly getting lost in confusing structures with inconsistent navigation options.

Luckily, many companies have begun taking intranet IA seriously, launching systematic efforts to maintain consistent navigation systems using a deliberately designed structure, rather than one that evolved haphazardly.

We initiated an effort to document intranet IA processes and the resulting designs, both in terms of the visible user interfaces and the underlying structures.

We analyzed the IAs for 56 intranets. Our analysis encompassed a wide range of organizations in 12 countries:

• 33 companies from a variety of industries, including financial services, utilities, and technology
• 11 government agencies
• 5 healthcare providers
• 4 educational institutions
• 3 non-profits

Architecture without Architects

Among the organizations in our study, only 25% had a full-time person dedicated to the intranet's IA. We can assume that companies participating in a project about intranet IAs have an above-average interest and commitment to the topic, so the percentage across all intranets is no doubt much lower.

All is not lost, however. While it's better to have a large, multi-disciplinary intranet team, a smaller team can do good work if some members take on multiple roles. People in other jobs can learn the basics of IA, just as they can learn the basics of usability. Just as all design teams should do user testing — even if they don't have a specialized usability professional at their disposal — all intranet teams should take IA seriously and take systematic steps to improve it, even if they lack an official "information architect."

On such teams, a designer, usability specialist, or writer typically takes on the information architect's role. Given a little training in the most important IA principles — and the resources to base their design decisions on user data — teams can certainly achieve great IA results without full-time, dedicated information architects.

One of the most important goals on an IA project is to institute a consistent user experience for two key elements: the visible navigation user interface, and the underlying — invisible — structure (where things are found on the intranet). To successfully achieve this, teams must:

• Decide to proactively design the IA instead of letting it evolve.
• Ensure that management supports the central IA designer's authority to provide guidance and structure to other departments' intranet work.
• Ensure that management won't second-guess the design team and impose the awkward structures or navigational terms that individual executives happen to like.

Generalist Tools, Simple Research Methods

For example, only 4% of organizations used computerized tools for card sorting studies. This certainly makes sense: you can get most of card sorting's benefits by sticking to the simple technology that the name implies (that is, actual index cards).

Our study's teams used only four research methods frequently: surveys, card sorting, traffic analytics, and user testing. These are certainly the most important methods; any organization that employs all four will have a sound foundation of empirical data on which to construct a good IA.

Only a few organizations used more advanced methods, and most would have benefited from a significant increase in systematic user research. For example, only a few organizations conducted field studies(also known as ethnographic studies), where users are observed in their natural environment, such as an office or the factory floor. A few organizations used search log analysis to great effect to improve the IA and identify short-cut menu candidates. After all, when users search for something, they: (a) want it, and (b) don't see it in the navigation.

The good news is that most organizations in this study employed some kind of user research as a foundation for their intranet IA. This is a huge advance over the early days of intranets.

Common Categories

Only three topics attained top-level navigation status on the majority of the intranets:

• human resources (HR) information (66%),
• company information, (63%), and
• news (59%).

When we started this project, we had hoped to produce a recommended IA for intranets. Although structural diversity ultimately made this an impossible goal, we did identify an IA skeleton that projects can use as a starting point and adapt to their local circumstances.

Many intranets follow several general patterns. Certain types of companies also tend to follow particular trends. For example, manufacturing companies often include a product-related category in their top-level navigation, whereas companies with a focus on intellectual property often present a top-level knowledge management (KM) category.

Personalization and Customization

Users should be able to customize the Quick Links list. In our study, some intranets definitely made doing this easier than others. It's always hard to get users to customize features, so it's worth spending time to make the default presentation as useful as possible right out of the box. And, if your customization UI doesn't have incredibly polished usability, it might as well exist for geeks alone, as no one else is likely to use it.

Because users don't like to spend time on customization, many intranets in the study had invested in personalization, which is generally getting better over time. For large intranets in particular, simple personalization can make the navigation more manageable for average users by emphasizing the things they need the most.

Still, don't forget the simplest, most "Web 1.0"-like navigation aid: cross-reference links. On many intranets, Related Links lists were immensely helpful, particularly when they were designed consistently so users always knew where to find them and what to expect.

Timeless Design

Although easier said than done, it's important to try and envision future changes and make the IA sufficiently resilient. Doing so will help the intranet remain reasonably stable despite reorgs, corporate mergers, new company directions, new projects, new features, and simply the inevitable growth of content over time.

In our study, task-based structures often endured better than intranets organized departmentally. In our user testing of intranets, we've also found that task-based navigation tends to facilitate ease-of-learning. Thus, the benefits for IA durability are just one more argument in favor of adopting a task-based structure for your intranet.

One change that can be disastrous for intranet navigability is requests from top management to change navigation labels to reflect the latest buzzwords or corporate fashions. It might make sense to make up a fancy term to brand the CEO's latest initiative or management priorities. But such neologisms should always be restricted to narrative content. You should never use them in the intranet's main navigation system, where they'll only wreak havoc and lead users astray.

Ask management if it's worth a few hundred thousand dollars in lost staff time to put made-up terms in the intranet menus. Usually it isn't, particularly because you'll likely be asked to remove the fancy terminology next year — probably just after people have finally gotten used to it. Stick to long-term terminology for navigation labels. These labels are the most costly words in the company in terms of the number of hours employees will spend pondering their meaning.

Fat Report

We offer this extensive documentation because our research shows that intranet IA designers face a tremendous challenge: because of the very nature of intranets, designers can't see how other intranets structure their information and can't play with other intranet navigation designs. Although you still can't actually play with the 56 intranets in this study, by publishing their complete IA details, we do let you dig into them as deeply as you like and investigate how they've solved problems that might be too specific for us to comment on in our analysis, but crucial to your own work.

And don't worry. Even though the full report is rather big, we report our main conclusions and recommendations in the first 161 pages, which are a relatively quick read. So, first read through this entire overview, and then dig into the details to your heart's content.

Although there's no single solution that you can copy directly, there's much to be learned from the diversity of solutions that teams have designed to address varying IA challenges. Borrow from their strengths and avoid their weaknesses, and your solution can be even better.

The 744 screenshots also bear close scrutiny. Their primary purpose is obviously to provide a gallery of design ideas for IA projects, but there are many other interesting design details to be found. For example, we're quite enamored by the way McDonald's presents its intranet usage statistics: "27893 Visitors Served Last Week," echoing its famous signs proclaiming the billions of hamburgers served. Even though intranets are a serious business tool, and a good IA should promote employee productivity, there's no reason we can't lighten up a bit in parts of the interface, especially when it aligns the user experience with the company culture.

Learn More

Full day IA tutorialand a full-day course on navigation designat the User Experience 2008 conference in Chicago and Amsterdam.

The conference also has a 2-day tutorial on intranet usability.

Visit the site to read the full alertbox | Feed made by Bruno Torres

The contents of this feed is Copyright © 2006 by Jakob Nielsen

Ad: Hosted by Dreamhost (Sign Yearly L1 plan with 50% off using the promo code BRUNOTORRES)

It's a rare pleasure to come across a user interface on the Web that uses dialog controls correctly. Even something as simple as radio buttons and checkboxes are incorrectly used half the time. And let's not even get started on drop-down menus, which are horribly abused, or the homemade scrollbars that deface most Flash sites.

Yahoo Finance recently launched a redesigned homepage that uses tabs correctly, as shown in the following tabbed areas from two different parts of the page:

Tabbed controls from the upper left and middle right of the Yahoo Finance homepage.

Tab Usability Guidelines

1. It uses tabs to alternate between views within the same context (not to navigate to different areas — a common mistake introduced by Amazon.com which has since abandoned this design).
2. It logically chunks the content behind the tabs so users can easily predict what they'll find when they select a given tab. (Card sorting is one option for researching this "mini-IA" problem. If you don't find clearly distinct groupings, then tabs are likely the wrong interface control for managing your content.)
3. Typically, users don't need to simultaneously see content from multiple tabs. If people do need to compare the info behind different tabs, then having to switch back and forth puts an added burden on their short-term memory and lowers usability compared to a design that puts everything on one big page.
4. The tabs are roughly parallel in nature, at least for U.S. users (for international users, it would be better to have a tab for "North America" or "Americas" to better parallel "Europe" and "Asia").
5. The currently selected tab is highlighted. Here, it's highlighted by its lack of color, which works fine as long as there are at least 3 tabs. (With only 2 tabs, it would be harder to tell which one was selected.)
   • In addition to highlighting, you can mark the current tab by size, a boldfaced label, an icon, or by making it appear to be in front of the other tabs.
6. The unselected tabs are clearly visible and readable, reminding the user of the additional options. If the non-highlighted tabs are faded too much into the background, there's a risk that users will never click them and never discover the many hidden features.
7. The current tab is connected to the content area, just as it would be if we were shuffling several physical index cards that had tabs stuck to them. This emphasizes which panel is being shown, and also helps tell users which tab is selected when there are only 2 tabs. Having the same color for the selected tab and the panel area reinforces the connection between the two and is a reason to support the use of reverse highlighting.
8. The labels are short and use plain language, rather than made-up terms. Tab labels should usually be 1-2 words. Short labels are more scannable; if you generally need longer labels, it's a sign that the choices are too complicated for a tab control.
9. The labels in the rightmost example above use Title-Style Capitalization: each word's first letter is uppercased. The leftmost example erroneously uses ALL CAPS, which is rarely a good idea because it's harder to read. Readability doesn't matter so much for single, short words, but — as guideline #20 for homepage usability states — you should pick one capitalization style and stick to it.
   • Microsoft's Vista User Experience guidelines recommend sentence case (in which you capitalize only the first character of the first word), whereas Apple's Human Interface Guidelines recommend title-style capitalization. In this case (pun intended), I side with Apple, but truthfully, it doesn't matter, so long as you stick with a single style. (Do avoid all uppercase or all lowercase labels, though.)
10. There's only one row of tabs. Multiple rows create jumping UI elements, which destroy spatial memory and thus make it impossible for users to remember which tabs they've already visited. Also, of course, multiple rows are a sure symptom of excessive complexity: If you need more tabs than will fit in a single row, you need to simplify your design.
11. The row of tabs is on top of the panel — not on the sides or the bottom where users would often overlook them.
12. The scope controlled by the tabs should be obvious from the visual design. Metaphorically, using tabs is like leafing through index cards in a drawer of an old-fashioned card catalog, so users must be able to tell at a glance what constitutes an "index card" (i.e., tab panel).
13. Fast response time ensures that clicking a new tab immediately brings the corresponding panel to the front. This is probably achieved through AJAX, but the programming technique is not important. What is important is to make the action fast enough (ideally <0.1 s) that people feel there's a physical connection between their mouse click and the appearance of the chosen panel.

Top navigation bar for Yahoo Finance's homepage.

Tabs should be reserved for letting users change the view while staying in the same place. Also, the little triangle beneath each label represents a pull-down menu, which is a highly non-standard widget to integrate with tabs.

Consistency

Sadly, there's a fourth tab control further down the page; here, the tabs are a different color. Mainly, though, this deviant tab control looks about the same as the other three, so it shouldn't cause major usability problems. In any case, it's so far down that page that many users won't scroll down to see it.

Consistency is important in GUI control design because it builds the user's feeling of mastery over the interface in several ways:

• Recognizability. When something always looks the same, you know what to look for and you know what it is when you find it.
• Predictability. When something always works the same way, you know what will happen when you act on it.
• Empowerment. When you can rely on your past knowledge of all the available features, you can easily compose a set of actions to achieve your goal.
• Efficiency. There's no need to spend time learning something new or worrying about the effect of inconsistent features.

AJAX Overkill

On the Yahoo Finance page, stock quotes are updated in real time and each change is announced by a color flashing behind the affected number. The constant flashing all over the screen gets tiring fast, especially for numbers like volume, which change incessantly.

Just because you can do something doesn't mean that you should. Yes, it is a guideline to draw users' attention to updates within a page, but only when users need to be alerted. In this case, all users need to know is that all numbers are current and will be continually updated.

A trading system for day traders would be a different matter. In that scenario, users need to continuously monitor market trends for multiple securities, and colors flickering in their peripheral vision could alert them to short-lived trading opportunities.

On a financial portal homepage, however, users have to scan a broad set of headlines and stories. If they want to trade, they go to their broker's site, so they couldn't take advantage of anything that required trigger-fast actions anyway. Few people would sit and stare at this page to keep up with second-by-second changes; the relentless blinking is overkill and distracts users from the content they came for.

Customization Difficulties

Customization options for part of the Yahoo Finance page.

A chart accompanies the first item in the bulleted list, but this fact isn't obvious. If users want to change the default chart, they must infer that they have to change the element order on the quotes list. At minimum, the instructions should explain this. Better yet, the page should offer a set of radio buttons next to the index symbols and have the selected radio button determine which symbol is charted.

Also, investment experts know the ticker symbols for the indices they want to track, but they'll stick to their Bloomberg terminals. Most Yahoo Finance users will have no clue about what to type into this form. It's thus good that the page provides a link to the symbol lookup feature (just-in-time help). However, given that this page has few other features, it would have been better to just list the 30 or so most common symbols right on the page. Users could then select a symbol by clicking it, rather than having to type it (which is more error-prone).

(It's good that this customization interface is offered on a separate page instead of being squeezed into an AJAX-style overlay on the main page. Sometimes usability benefits from focusing users' attention on the matter at hand and allowing the necessary room for helpful features.)

I chose to customize this page because it shows the Dow index as the default. Every investment book I've ever read says that the S&P 500 is a better indicator for the U.S. stock market, so that's obviously the one I wanted to see. (In the interest of overly full disclosure, S&P has been one of our consulting clients, but that's not why I prefer their index. I prefer it because averaging 500 stocks gives a broader picture than averaging 30 stocks.)

Perhaps Yahoo should have used the S&P 500 as the default chart from the beginning, freeing me from having to enter the customization screen. However, while the S&P 500 is theoretically better, most press coverage centers on the Dow, so most novice investors likely understand it better.

Ethnographic researchcould tell us more about how people think about investing and stock indices, and it's quite possible that Yahoo did the research and had good reasons for picking the Dow. I haven't researched this particular question (our studies of investors focus on their use of the investor relations (IR) sections of corporate websites), but my guess is that Yahoo did the right thing and most users do prefer to see the Dow as their default.

While Yahoo's customization has its usability problems, the page does make good use of personalization, which is difficult to do. (The article I wrote on problems with personalization 9 years ago still holds true. The very fact that personalization improves so slowly is one reason to be skeptical about any new claims for sudden improvements in personalization technology.)

Among the page's personalized features is a list (not shown here) of current quotes for the last several stocks the user looked up. Recent interest is a good proxy for current interest, and it's very likely that users still care about the quotes for those recently checked stocks.

While the new homepage design for Yahoo Finance has its weak spots, it's mainly excellent. I'm particularly pleased to see such a correct and effective use of tabs (global navigation aside).

Learn More

The conference also has afull-day course on navigation design.

Visit the site to read the full alertbox | Feed made by Bruno Torres

The contents of this feed is Copyright © 2006 by Jakob Nielsen

Ad: Hosted by Dreamhost (Sign Yearly L1 plan with 50% off using the promo code BRUNOTORRES)

In designing any user interface, one of your key decisions concerns the tradeoff between features and simplicity. The more features, the more complicated the system inevitably becomes:

• Features have to be shown to users, so screens get busier.
• Menus get bigger and/or more numerous, making it harder for users to find the features they need.
• Features must be explained, ballooning the size of the help system and/or the manual:
  • Fatter documentation takes longer to read and makes it harder for users to extract a good conceptual model of the system.
  • More docs also make it harder for users to find the explanations they need.
• Each extra feature offers more rope for users to hang themselves: they're more likely to use the wrong feature, either as an error of intent (a mistake caused when they think the wrong feature is the one they need) or as an error of execution (that is, a slip, as when they click the wrong button in a crowded toolbar). Conversely, Steve Jobs famously defended the Mac's one-button mouse by pointing out that users would never click the wrong mouse button.
• The number of feature interactions grows by the square of the number of features: more can go wrong, and it becomes harder for users to understand why a change in one corner of the system has an effect in another corner.
• The more options users have to choose from, the more time it takes their brains to prepare for action and decide what to do. Even if a fancy feature can theoretically execute a task faster, overall system use often slows because users spend more time on the mental operations required to choose from among features than they save from the more efficient feature.

Right-Click: An Extra Feature That Works

Two weeks ago, I observed dozens of average-skilled business users as they attempted common business tasks with two high-end applications. Even though these people were neither geeks nor experts in the software we tested, most of them frequently used right-click shortcuts. The contextual pop-up menu was often their operation of choice, allowing for great efficiency in many tasks.

(Before you redesign your user interface around right-click pop-ups, be warned: less skilled users rarely use these menus.)

Right-click helps medium-skilled users because it's a consistent interaction technique that works the same everywhere. (Indeed, high-skilled users are often disappointed when an application doesn't support right-click -- for example, if it's implemented in Flash and brings up the Flash player menu instead of contextually-appropriate application commands.)

Right-click also works because business professionals and other mid-level users typically depend on their PCs and are willing to learn a few techniques to use it better.

User Engagement Levels

Mostly, though, users have a low engagement level with user interfaces and just want them to get out of the way. People don't want to spend time learning, they want to spend time doing -- a well-documented effect called the paradox of the active user. (It's a paradox because people might save time in the long run if they spent more time learning about powerful features. But, empirically, users almost never want to do this, and you should design for how people actually behave, not how you wish they behaved.)

An example of a high-engagement design is the seventh and final book in the Harry Potter series. The first 100 pages of Harry Potter and the Deathly Hallows make no sense unless you're intimately familiar with the first 6 volumes in the series and can remember every detail about a profusion of magical objects and earlier plot twists. I found it a bit difficult to follow the story in Potter #7, but on balance, I think the author made the correct design decision in targeting customers who were very strongly engaged in the series: People aren't likely to buy #7 if they haven't read the earlier books, and Harry Potter fans tend to be on the fanatical side of fandom.

Traditionally, the design of a book series caters to lower-engagement readers by adding exposition that brings them up to speed. But -- at 759 pages -- The Deathly Hallows is already fat enough that adding, say, 50 pages of background material would have pushed it over the edge and made it less enjoyable for most readers.

Three Levels of Photoshop Complexity

1. Photoshop CS (list price $650), targeted at professional graphics artists and photographers.
2. Photoshop Elements (list price $100), targeted at photography enthusiasts and people who want to do basic image manipulation, such as cropping screenshots.
3. Photoshop Album Starter Edition (free), targeted at average consumers who've just bought a digital camera and might want to fix a redeye or brighten a dark image.

However, the very success of such training products proves that users will gladly pay more than the price of the software to learn how to use it. For professional users, Photoshop creations are their work products, and being able to make a picture look better is worth almost any amount of training and user interface complexity.

To its credit, Adobe recognized that Photoshop CS is much too complicated for people who just want to clean up their personal photos. Most users don't need that many powerful features, and they certainly don't want to read beefy manuals and extra books just to get a better-looking vacation snapshot.

The Photoshop Elements manual is both thinner and easier to read than the CS manual. And the documentation for Photoshop Album is only 20 pages long.

Each of the three versions is appropriately targeted at a particular level of user engagement, from people who care passionately about image manipulation to those who aren't particularly interested in graphics software.

Shallow Engagement with Websites

The user engagement level with websites is incredibly low, as dictated by information foraging: people don't commit easily to any individual site, because it's so easy to get to other sites. Skimming the cream from each site is usually the superior browsing strategy.

As studies in my recent book document, users visiting a new site spend an average of 30 seconds on the homepage and less than 2 minutes on the entire site before deciding to abandon it. (They spend a bit more time if they decide to stay on a site, but still only 4 minutes on average.)

• For an example of user fickleness and lack of engagement with each individual site, see the transcript of a B2B user session.

Intranets typically sustain mid-level user engagement because a company usually has only one intranet. That's one key reason that we have hundreds of separate intranet design guidelines. (The other reason is that employees' intranet tasks differ from the tasks that customers perform with websites.)

Be aware that not all applications can expect deep user engagement levels. Particularly for ephemeral applications embedded within websites (say, the configurator on a car site), users often have close to zero commitment: if an applet's purpose isn't immediately obvious or if features are too complex, people will leave a Web-based application just as readily as they'll leave a content page.

When we tested the investor relations sections of corporate websites, we found that most individual investors were confused by advanced charting tools or other features that went beyond very simple interaction techniques. Even though investors have thousands of dollars at stake, they still prefer websites to be simple because they have to deal with multiple sites.

Loyal Users: Higher Engagement Levels?

A famous example is Amazon.com's one-click shopping feature. It's complicated to understand, and pretty scary to boot. Still, one-click shopping helps some users. Amazon's usability rules differ from those for most other sites because it's big enough and established enough that many users have established a close relationship with the site and are thus willing to engage at some level of depth.

But even Amazon makes it easy to shop in the traditional way by using the shopping cart, which is a design patternthat everybody knows by now. Indeed, I strongly advise the vast majority of websites to scale back their features and dramatically simplify the user experience for initial use. After all, to progress to the deeper engagement levels, prospective customers must first successfully pass through the initial use phase.

Typically, when new prospects first visit your site, you're simply one of ten sites on the SERP. The only way they'll shortlist your site is if you can convince them in two minutes.

Thus, websites should have almost no features: focus on the words.

To determine how much complexity you can afford in a user interface, you must analyze user engagement levels: Do they care deeply, or do they just want to get something done as quickly as possible? Typically, users care less than you think! You're not important to them. This is one of the main reasons companies need systematic usability studies: to make explicit the fact that outside customers don't find your design as important as you do (because you work on it all year).

Visit the site to read the full alertbox | Feed made by Bruno Torres

The contents of this feed is Copyright © 2006 by Jakob Nielsen

Ad: Hosted by Dreamhost (Sign Yearly L1 plan with 50% off using the promo code BRUNOTORRES)

Not all design decisions are a matter of website survival. Of course, it's important to get the big things right, or you won't have any users. But getting the small things right enhances usability and fosters user comfort. A perfect example here is the breadcrumb trail.

Breadcrumbs won't help a site answer users' questions or fix a hopelessly confused information architecture. All that breadcrumbs do is make it easier for users to move around the site, assuming its content and overall structure make sense. That's sufficient contribution for something that takes up only one line in the design.

Breadcrumbs have always been a secondary navigation aid. They share this humble status with site maps. To navigate, site visitors mainly use the primary menus and the search box, which are certainly more important for usability. But from time to time, people do turn to the site map or the breadcrumbs, particularly when the main navigation doesn't quite meet their needs.

Despite their secondary status, I've recommended breadcrumbs since 1995 for a few simple reasons:

• Breadcrumbs show people their current location relative to higher-level concepts, helping them understand where they are in relation to the rest of the site.
• Breadcrumbs afford one-click access to higher site levels and thus rescue users who parachute into very specific but inappropriate destinations through search or deep links.
• Breadcrumbs never cause problems in user testing: people might overlook this small design element, but they never misinterpret breadcrumb trails or have trouble operating them.
• Breadcrumbs take up very little space on the page.

The main argument against breadcrumbs is that many users overlook them. So, why do something that only benefits a minority?

As I've long argued, breadcrumbs are different than most other little-used design elements for the simple reason that they don't hurt users who ignore them.

Growing Popularity

In testing an e-commerce site last month, for example, one user complained: "This is missing a feature to go back to the previous page."

I found this apparent request for a Back button puzzling, since the button was featured prominently in the browser and the person had easily used it earlier in the test session. Also, for six years, it's been an established guideline to avoid duplicating browser functionality in the page design.

It quickly became clear, however, that the user wasn't asking for a duplicate Back button. Elaborating on the previous complaint, she pointed to the place on the page where breadcrumbs typically appear and said she wanted the list of links to higher-level pages.

In other words, the user wanted breadcrumbs. She'd seen them before, but didn't know what they were called, so she asked for them using words that -- if taken literally -- would have been easily misinterpreted.

This is a great example of the hard-won lesson of usability: don't comply with user requests. Give more attention to what study participants do than to what they say.

Consistency Breeds Familiarity

Why are people now using breadcrumbs to the extent that they actually miss them when sites don't offer them?

It's exactly because of breadcrumbs' modest nature that people are becoming accustomed to them. There aren't too many ways to mess up breadcrumbs in a design. No fancy stuff, just a line of textual links.

Breadcrumbs are almost always implemented the same way, with a horizontal line that

• progresses from the highest level to the lowest, one step at a time;
• starts with the homepage and ends with the current page;
• has a simple text link for each level (except for the current page, because you should never have a link that does nothing); and
• has a simple, one-character separator between the levels (usually ">").

Breadcrumbs are also useful for intranets: 80% of this year's award-winning intranets use breadcrumbs. [See separate article on intranet information architecture (IA)].

Hierarchy or History?

In user interface design, it's often dangerous to take metaphors too far, and breadcrumbs are again the perfect example. Offering users a Hansel-and-Gretel-style history trail is basically useless, because it simply duplicates functionality offered by the Back button, which is the Web's second-most-used feature.

A history trail can also be confusing: users often wander in circles or go to the wrong site sections. Having each point in a confused progression at the top of the current page doesn't offer much help.

Finally, a history trail is useless for users who arrive directly at a page deep within the site. This scenario is when breadcrumbs show their greatest usability benefit, but only if you implement them correctly -- as a way to visualize the current page's location in the site's information architecture.

• Breadcrumbs should show the site hierarchy, not the user's history.

For non-hierarchical sites, breadcrumbs are useful only if you can find a way to show the current page's relation to more abstract or general concepts. For example, if you allow users to winnow a large product database by specifying attributes (of relevance to users, of course), the breadcrumb trail can list the attributes that have been selected so far. A toy site might have breadcrumbs like these: Home > Girls > 5-6 years > Outdoor play (note that the links should be colored as well as underlined, but I don't do this here to avoid confusion with these dummy links).

Looking ahead, people will use breadcrumbs even more because they're an important navigation tool in Windows Vista. Most users don't distinguish clearly between the operating system, applications, and content or websites. Users will thus transfer their understanding of Vista's interaction techniques to your website.

If you don't have breadcrumbs, it's time to start planning for them. They'll improve your usability a bit now, meet increasing user expectations in the future, and -- most importantly -- they won't hurt.

Learn More

User testing and the correct way to analyze different kinds of user behaviors and comments are covered in the conference's full-day tutorial on correct user testing methods.

Visit the site to read the full alertbox | Feed made by Bruno Torres

The contents of this feed is Copyright © 2006 by Jakob Nielsen

Ad: Hosted by Dreamhost (Sign Yearly L1 plan with 50% off using the promo code BRUNOTORRES)

So far this year, we've watched users shop on about 50 e-commerce sites. All but one of the sites violated a documented guideline for checkout design: they required users to manipulate a drop-down menu to enter their state abbreviations, rather than simply let them type in the two characters.

The exception was Amazon.com, which offered the faster and more pleasant typing option. Amazon thus confirmed that even though the average e-commerce site should not copy its overall design it continues to be the leader in complying with usability guidelines for individual design elements.

Knowing a better design exists made it painful to sit, day after day, and watch users fight with the mouse to scroll through the huge menu. Sometimes users selected the wrong menu option and then had to waste even more time with the drop-down. And, in this study, we mainly tested young, able-bodied users; the situation is even worse for elderly users, who have more difficulty with extensive, fine-tuned mouse manipulations. And it's worse yet for users with disabilities.

We observed the same problem again earlier this month when watching Chinese users shop on international sites: users suffered a lot of needless interaction overhead when trying to select "Hong Kong" from immense drop-downs containing hundreds of countries and territories.

Sites offer drop-downs for state abbreviations under the theory that doing so prevents input errors. But that's not true: menus are more error prone than typing because the mouse scroll wheel often makes users inadvertently change the state field's content after they've moved their gaze elsewhere on the screen. In contrast, everybody knows how to type their own state's two letters, and it's always faster to enter this information through the keyboard than the mouse.

(Regarding input errors: whatever the input method, sites should validate that the ZIP code/postal code corresponds to the state, province, or other locality entered by the user. Because postal codes are more error prone, you must use validation code on the backend, regardless of whether or not you use a drop-down for the state.)

One Annoyance? Irritating, But Bearable

The drop-down menu is unpleasant, and we sometimes hear users sigh when they encounter it. That said, they know how to beat it into submission, because other sites have similarly annoyed them before.

Therefore, it's unlikely that many sales are lost due to this user experience degradation. The drop-down does cost sites money: given the scroll wheel's revenge, it's inevitable that companies will ship packages to the wrong state or — more commonly — that they'll have to call users to resolve state/ZIP discrepancies.

So, why go on about a design mistake if it doesn't cost a company sales? It's certainly an error we'd classify as a low-priority issue in a client report. But it's still a usability problem, because — as nearly every test session confirmed again this year — state drop-downs annoy customers.

Many Annoyances? Disruptive

A site that has many user-experience annoyances:

• appears sloppy and unprofessional,
• demands more user time to complete tasks than competing sites that are less annoying, and
• feels somewhat jarring and unpleasant to use, because each annoyance disrupts the user's flow.

In many ways, repeated annoyances are like slow response times: one slow page, and you'll stick with a site; many slow pages, and it's toast.

Fixing annoyances won't double your business metrics the way fixing usability catastrophes will. But, eliminating annoyances increases customer satisfaction and user loyalty, and thus improves the long-term business value of the site.

Learn More

The conference also has a full-day seminar on which GUI widgets to use when for forms and applications.

Visit the site to read the full alertbox | Feed made by Bruno Torres

The contents of this feed is Copyright © 2006 by Jakob Nielsen

Ad: Hosted by Dreamhost (Sign Yearly L1 plan with 50% off using the promo code BRUNOTORRES)

The 10 best-designed intranets for 2007 are:

• American Electric Power (AEP), United States
• Comcast, United States
• DaimlerChrysler AG, Germany
• The Dow Chemical Company, United States
• Infosys Technologies Limited, India
• JPMorgan Chase & Co., United States
• Microsoft Corporation, United States
• National Geographic Society, United States
• The Royal Society for the Protection of Birds (RSPB), United Kingdom
• Volvo Group, Sweden

Contrary to last year, when most of the winners hailed from outside the US, this year 6 of 10 winners are American. Of the four winners from other countries, three come from countries that have generated many past winners: Germany, Sweden, and the UK. Given the size of Germany and the UK, it's not surprising that they've had so many winners. Sweden's continued top placement, however, is striking. Maybe there's something to the claim that Scandinavians emphasize good design. (Then again, other Scandinavian countries have never produced a winner, so maybe Sweden rules in intranet design.)

Most of this year's winners are from countries that have fostered previous winners. However, we also have a new country represented this year: India. Although the World Bank Group used an Indian design agency when it won in 2002, the bank itself is a multinational organization headquartered in the US, so we counted its intranet as US-based. Thus, Infosys is the first winner truly based in India. Having India join the ranks of winning countries is a clear symbol of its growing might as a software superpower.

Rise of Usability in Manufacturing

Compared to past years, we have many more manufacturing companies in this year's top 10. Viewing "manufacturing" broadly, this sector has four winners: AEP, DaimlerChrysler, Dow, and Volvo. Last year, we noted that "manufacturing companies have historically focused on physical concerns and thus have less experience in creating good screen-based designs." Of course, in the modern world, manufacturing is highly intertwined with computers because it's a highly knowledge-intensive business. We can hope that the prominence of manufacturing companies among this year's winners symbolizes this important sector's embrace of the value of usability, and that the previous poor showing of manufacturing intranets is truly a thing of the past.

While most winners are big companies, we also have two winners from the non-profit sector: National Geographic Society and The Royal Society for the Protection of Birds. As this shows, you don't have to be a large organization or a traditional company to benefit from intranet usability. Well-designed intranets support employees and volunteers in achieving non-profit missions just as well as they improve productivity — and thus profitability — in for-profit businesses.

Multimedia, News, and Ratings

Many intranets have long offered news feeds, but this year's winners have taken extra steps to make their news offerings more relevant to employees, both for internal news and for industry-related external news. Labeling and categorization are more extensive than before, and several intranets let users rate and comment on stories.

Star ratings and user comments have long been found on public websites — from Amazon.com to weblogs — but they become much more useful on intranets, where they're not degraded by the Bozo effect. Employees of the same company have shared goals and interests, they have passed the quality filter of getting hired, and they have their reputations to protect. For all these reasons, ratings and comments from colleagues are likely to be much more useful than those of random blog readers.

In addition to providing news on its intranet, Microsoft offers its employees a range of email newsletters and the ability to get stories through a newsfeed (RSS). Email newsletters are a simple way to reach beyond the intranet to give employees news on their mobile devices. This only works, however, because the messages are formatted for mobile devices, which is rare.

AEP doesn't use an automated feed for outside news. Instead, an editor reviews the available stories and posts only those that will be of most use to the company's employees. Such extra work is amply rewarded in employee productivity by saving people from long lists of irrelevant news. For example, at JPMorgan Chase, the intranet homepage is viewed 620,000 times per day, so even one superfluous headline that required one second to scan would cost the company the equivalent of 22 full-time employees in lost productivity. The JPMorgan Chase intranet team is equally selective, displaying only the most important news on the homepage.

Of course, in some cases, you just have to provide the news, even when it's not particularly work-related. For example, DaimlerChrysler provided the latest scores during the soccer World Cup in Germany. If they hadn't, employees would surely have spent much more time following matches on external sites.

In terms of respecting employees' time, perhaps the ultimate design feature comes from the Volvo Group. The company's 5 Minutes Only area gives people the most important information they need that can be consumed in five minutes.

Multinational Intranets

For example, DaimlerChrysler offers global content in German or English, automatically setting the initial language based on the language preference settings in the user's browser. It's somewhat rare to see browser language preferences used correctly on public websites, so it's great to see this feature making inroads on intranets.

Dow uses English for most global content, but translates the most important content into six other languages (Dutch, German, French, Italian, Portuguese, and Spanish). It also translates selected content into Chinese, Greek, Japanese, and Thai.

Although translation is important, it's not sufficient for a true multinational user experience. Dow's intranet shows how to further achieve this with its employee recognition application. Users can nominate employees in other countries for a recognition award, and if the award is approved, recipients are notified in their local language. Even better, the awards are appropriate to the recipient's culture, and can be redeemed locally. Without this true internationalization of the underlying features, consider how difficult it would be for, say, a German manager of a cross-functional team to give an award to a deserving employee in Brazil. Such a manager is not only unlikely to know Portuguese, he or she is unlikely to know what a Brazilian might appreciate receiving as an award or which local vendors might offer redeemable certificates. At Dow, the intranet comes to the rescue, thus encouraging more cross-cultural employee recognition.

Several intranets featured a simple, but highly useful design element to help users work with overseas colleagues: a world clock. Calculating time zones and understanding the International Date Line's effects are difficult for humans but trivial for computers. Let them do it, and you'll never again call clients or colleagues at 4 a.m. on a Sunday, thinking it's their Monday afternoon. As an added benefit, a prominently placed world clock serves as a tangible reminder of an organization's worldwide status.

Technology Platforms

The most-used products were: Windows Server, Google Search Appliance or Google Mini, SharePoint, SQL Server, Google Maps, Omniture, and Vignette.

Some people might claim that it's "unfair" to include Microsoft products on this list, given that Microsoft's own intranet was one of the winners this year. Obviously, Microsoft tends to use Microsoft products, but many other winners did so as well. Also, IBM won last year's competition, and many other technology companies have won throughout the years. In each case, we gave the awards for the quality of user experience on the intranets, not for the product lines. The profile of Microsoft's intranet serves as a valuable case study in how to design a great intranet while building on Microsoft products — just as last year's IBM intranet profile is useful to the many companies that employ IBM products.

Standard UI, No Standard CMS