Summary: 19 new case studies of enterprise portals find slow growth in new features; the focus is on robust integration and formalizing governance.

The headline for our last intranet portal study was "Enterprise Portals Are Popping." Now, 3 years later, we revisited this space with new research and our findings would best be summarized as "Enterprise Portals Are Stabilizing." Although we saw some new features, the main push was to make existing features more robust and better managed.

For this latest research round, we looked at intranet portals in 19 organizations, spanning the globe from Australia's ANZ Bank to the British Red Cross and South African Breweries. In total, across all the research, our recommendations for portal design are now based on 67 case studies of intranet portals.

Early definitions of "portal" focused on the gateway concept, but times have changed: characterizing portals as mere doorways to other places no longer adequately describes the sophisticated role they play. Today's portals are not just about access; the best ones provide true integration of enterprise information, resources, and tools in a unified user experience. The portal is a dashboard that offers all the enterprise information and applications that employees need to do their jobs.

At the same time, the distinction between intranet and portal is diminishing as companies increasingly adopt a portal perspective for their intranets. The trend is toward "an intranet is an intranet is an intranet" and a portal is just a nicer, more functional intranet that integrates more systems.

The ROI from achieving the full portal vision is clear: saving countless, costly staff-hours that would otherwise be wasted hunting for information on various systems and learning incompatible user interfaces. Of course, the ideal vision rarely happens: for example, single sign-on remains an elusive chimera, though companies are closer and closer to achieving it every time we study this problem.

Slow Mobile Progress

The biggest finding from our new research into enterprise portals? The sad fact that portals are not adding mobile features at the expected rate. Outside the firewall, the mobile space is teeming with innovation, but inside companies, mobile progress seems to be progressing at a snail's pace. The good news: this is obviously an opportunity for one of the lagging mobile vendors to leapfrog Apple, which doesn't have a great enterprise story (to put it mildly).

Anybody wanting to launch a mobile intranet portal is advised to heed the strong finding from our research with mobile websites: Good mobile usability requires a separate design with a reduced feature set for mobile use cases, focusing on time- and location-dependent tasks. It's not enough to make an existing portal accessible through phones because the UI is optimized for desktop use.

As an example, a university in our study was considering a mobile feature that would let students use their phones to check their meal plan credits and the cafeteria lines. Although these might not be mission-critical features in the average company, helping students eat might well be a killer mobile app for a university. The key point is to start with a task analysis of users' actual mobile needs, rather than simply trying to shoehorn everything onto the tiny screens.

One of our new case studies had previously hosted a half-baked mobile user experience on its portal. Realizing this, team members identified mobile as a requirement for the new portal. To their credit, they decided to wait and do it right later. Having tried and failed before, they knew just how hard it is to do.

Most of the companies we studied saw true mobile portals as being at least a few years out. And many planned to have sales-force support as one of the initial features.

I fully agree with the idea of waiting until you can get it right; poor mobile designs are really miserable for users. And for enterprise use, you pay for every minute employees waste slugging through a bad UI. Still, I advise companies to plan to make their intranets mobile sooner rather than later. As employees increasingly see rapid improvements in their mobile user experiences on the open Internet, they'll demand it from their organizations as well.

Governance Suffering

From wrangling C-level participation to managing distributed content management, companies are struggling with governance — and, unfortunately, our new research won't fix their problems. At this point, not enough companies have intranet governance "solved" to offer insights that are broadly applicable as best practices.

The only universal conclusion actually relates to this lack of perfect governance solutions: Anybody who attempts to sell you one is hawking snake oil.

Instead, the recommendation is to take general governance solutions that have worked in our various case studies and adapt them to your organization's specific corporate culture and other circumstances.

For example, Duke Energy has 5 layers of governance:

• New Media team: oversees and governs the user experience; trains new content owners; reviews sites to ensure compliance with standards; and consults on how sites can add more value, keep their content up to date, and implement new technologies or social features.
• Site managers (throughout the business): responsible for the content on their sites.
• Internal Communication Department: manages the homepage content.
• Portal steering team (made up of VPs and directors from Communications, HR, IT, and Marketing): helps prioritize enhancements based on business value and manage the organizational and technical impacts of enhancements; meets monthly.
• Portal executive council (made of Sr. VPs from Communications, HR, and IT): approves long-term strategy, allocates funding, and advocates for portal initiatives with the senior leadership team; meets quarterly or on demand.

Do you need all this? If you're an even bigger company, you might need more, but other companies can get away with less. For example, Ohio State University Medical Center has a fairly simple governance committee. In any case, a key lesson from many of our case studies is that organizations should plan the governance structure before starting a portals project. Success doesn't come from buying a software package. It comes from running the project right, and from maintaining good governance after launch.

The need for ongoing portal maintenance and continuous quality improvement has an important implication: the portal is a job, not an ancillary project. There must be someone — or possibly an entire department in a big company — whose formal job description includes the portal.

In our previous research, when we asked who owns the intranet, the answer was often some combination of IT, marketing, and HR. In this round, we found the responsibility shifting more consistently toward corporate communications, with IT playing a logical (and necessary) custodian role, providing consulting on technology decisions and portal support. Cross-department portal ownership also worked well in several of our case studies. Even when this arrangement was informal, the most successful portal projects seem to involve cross-functional business owners or steering committee members.

When it comes to aging or overburdened intranets, organizations tend to follow the old adage: "If it ain't broke don't fix it." But when the intranet does break and the solution is a new portal, that portal can transform not only the intranet, but the organization as well. One reason is that portals need a strong governance structure to support them, and instituting a portal solution can drive the need for a governance structure that once might have seemed a luxury.

At ANZ, for example, the project to field a new portal required adapting project models such as Six Sigma, user-centered design, and centralized publishing. These better ways of managing the intranet resulted in user experience improvements: staff were able to complete tasks an average of 50% faster and the number of unsuccessful attempts to find information decreased by almost 70%. Given the number of employees in such a big financial services company, these are ROI numbers you can take to the bank.

(As further discussed in our report on ROI for usability, the annualized value of a design improvement that saves t minutes in time-on-task is t × e × n × s , where e = the number of employees performing that task, n = the number of times per year a typical employee performs the task, and s = the average employee's loaded salary per minute. In a big company, a good redesign can be worth tens of millions, which helps make the t × e × n × s formula easy to remember.)

Although big organizations get the biggest ROI from intranet improvements, they might suffer under a particular type of ambition inertia : upper management has often been in the same company for a decade or more, so they've never experienced how good intranets can get in other companies. This can make them reluctant to sponsor a new portal. (In contrast, for public websites and regular software, executives have personal experience with Amazon.com, Google, and the like, making them painfully aware of their own offerings' user experience shortfalls.)

One obvious way to reignite ambitions is to study other portals and build a vision of "what can be." That's one of the main goals of our work.

Personalization Becomes Critical

In our past portal projects, personalization seemed optional. Now, this has changed, and the prevalent sense is that personalization is a critical component of a well-designed portal.

Why this change? More stuff!

Information overload might be a cliché, but it's also a truth. Even without portals, employees are drowning in information. Paradoxically, portals can become a victim of their own success as they become capable of integrating ever-more sources of information and applications. The more the portal serves up to the users, the stronger the need to curate what each person sees, or they'll truly be overwhelmed.

For example, ANZ operates in 32 countries. While it's good to integrate siloed information, Australian employees don't typically need to see New Zealand information in their search results. Personalization can achieve this.

Unfortunately, effective personalization is a tough nut to crack. Even though leading portal software platforms offer personalization features, you shouldn't underestimate how much work is needed to fully implement them.

Many portals in our study experimented with customization — either in addition to personalization or as an alternative. (With personalization, the portal automatically determines what to show, whereas customization requires users to manually choose various features.) Letting users make their own choices sounds good, but in practice, customization usually works poorly: business professionals are busy and often see the need to mess with UI preferences as an annoyance.

A frequent compromise is to offer a My Page feature. Many users are familiar with this design pattern from public websites. Initially, each user's My Page can be populated through role-based personalization, with additional customization available that lets users add and subtract content widgets and control the look and feel of their pages. (If you choose this option, be aware that many users will simply use the default My Page without customizing it.)

From Social Features to Collaboration Platform

People finders remain a killer app for intranets. Nothing beats talking to a colleague, but you've got to find their contact details first. Virtually all portals offered these tools front and center, both through traditional staff directory searches and fancier expertise locators with various filtering options.

It's important to remember the crucial importance of these old tools and to continue making them more effective. But at the same time, portals have also evolved into collaboration platforms in their own right, paralleling the growth of social features on the open Internet.

Many portals include some Web-like social features that simply help employees get to know each other, which brings its own business benefits. But goal-driven collaboration is even more important. At the law firm Goodwin Procter, collaboration was a key driver for the new intranet: the goal was to get attorneys to collaborate and reuse information so they'd be more efficient and keep billing costs down.

Are social features and collaboration features really separate on portals? In most of our case studies, companies didn't sharply distinguish between the two. The stronger distinction might be that between informal vs. formal collaboration. For example, at Cisco, formal content is officially managed, whereas informal content is left to emerge as it may. However you draw the line, the portal's transformation into a collaborative platform makes this one of the new governance issues that must be addressed. Cisco was also a leading-edge case study through its integration of multimedia and real-time formats in the collaboration platform, including its own TelePresence video technology.

Robust Enterprise Portals

The days of the intranet as the Wild West haphazardly built by maverick content providers in individual departments are over. The portal today is a more civilized place, offering an integrated platform for content, applications, and collaboration, unified under a central governance structure with buy-in from all stakeholders.

Well, that's the ideal, anyway. Most enterprises are not quite there yet, but they're making good progress.

Full Report

The full research report with case studies of intranet portal design is available for download.