When we first studied social networking on intranets 4 years ago, we concluded that “across our case studies, 3–5 years seems to be a common timeline for social intranet projects.” We’ve now gone back to revisit several of the companies covered in that early report, and sure enough, we were right. Their social projects are now much more established and have moved past the experimentation stage; companies that went social early are now reaping the benefits.

In addition to re-examining these established projects, we also studied a new group of companies that were adding social features to their intranets. We now have findings from a total of 22 companies’ social projects.

During the 4-year period between our two social intranet studies, the public internet’s social networking websites have experienced turbulence. We saw uncontrolled exuberance during the Web 2.0 bubble, where many people gambled (sorry, “invested”) fortunes on Facebook’s overpriced IPO, only to see Facebook, Groupon, Zynga, and similar stocks implode. Facebook reached a billion registered users, but hundreds of millions of them went into Facebook withdrawal and either stopped using the site or severely curtailed the time spent checking updates.

Despite these ups and downs, there’s no doubt that social features are here to stay on the public internet, and that many mainstream sites are adding them. (For more on this trend, see our reports and articles on social media user experience.)

Similarly, there’s now no doubt that social features are even more useful inside the enterprise for supporting employee collaboration and knowledge management. We used the slightly hokey term “Enterprise 2.0” to summarize our early research, and the new research confirms that the real effect of social features on intranets is to change how organizations function by making communication more open.

Stable Findings

Considering how much social networking websites have changed over the 4 years between our social-intranet research rounds, it’s remarkable that almost all of our original findings continue to hold. Given this stability, it’s very likely that the following conclusions will remain true for many additional years:

  • Little training is required. Users take to social tools easily when they’re given them for the right reasons and in the right work context. It takes little training, transition time, or urging to get people on board. In general, you should design social tools that employees can easily use without special training. In addition to following usability guidelines, you can achieve this goal by emulating popular designs, such as the 5-star rating system known from Amazon and Netflix. Two other important points to bear in mind:
    • Even people who heavily use internet social tools can benefit from training on the appropriate code of conduct for using social tools in the corporate context.
    • Avoid advertising the new tools as “new tools.” Instead, simply integrate them into the existing intranet, so that users encounter them naturally. For example, you could turn an existing bookmarking (or Quicklinks) feature into a socially shared bookmarking feature without great fanfare.
  • Designing social media for an intranet gives you a major advantage over similarly tasked internet designers. You know enough about employees and their work to pre-select stuff that they’ll likely find interesting. You can, for example, pre-populate news feeds with relevant information; if you offer users a blank screen to customize, they’ll often experience the social-media equivalent of writer’s block.
  • Despite companies’ concerns about employees using social tools with impropriety, infractions remain rare. Companies with social tools often use a gentle hand in creating and enforcing terms of use. Most community managers do little more than guide wayward conversations toward more constructive paths. As long as attribution is built in and required, communities police themselves.
  • Community management is vital in social environments. Community managers are most effective when they manage with a light hand, guiding the conversation rather than controlling it. Designated community managers should serve as facilitators and moderators; they can also reignite slow areas. Ultimately, by keeping a finger on the pulse, community managers will know when it’s time to pull the plug on a topic rather than flog it beyond its time.
  • Not all users are producers. Although social software has matured, the participation–contribution ratio has remained steady. As with the open internet, there’s substantial participation inequality in enterprise communities: some employees participate a lot, while others mainly lurk. It’s therefore important to value a community based on a combination of posting and use, because those who lurk also benefit.
    • When calculating ROI, it’s important to give your Enterprise 2.0 initiative credit for the value that employees contribute to the company based on their increased knowledge and understanding.
    • Even a few active contributors can add substantial value to the entire organization. In our case studies, this was often true for tagging and rating systems, which considerably improved the quality of results prioritization for notoriously ailing intranet search tools. Traditional web methods for relevancy scoring don’t work as well on the smaller scale of most intranets. For example, counting links works only if you’re doing so across a huge base of links. But, even if only a few employees tag a page with a given keyword, it’s likely that the page will produce a good search result for that query in your organization’s context.
  • The relationship between “official” and user-generated content is a delicate dance. As it grows, user-generated intranet content can help address many employee questions. However, official content about policies and positions still has a role. You shouldn’t segregate these two types of content — the guideline to provide a single federated intranet search, for example, remains more relevant than ever. Still, your design should reflect the different statuses of different content types: label official information as such, and perhaps color-code it as well for easy scannability on SERPs (search engine results pages) and other lists.
  • Search must be integrated. In the enterprise, information retrieval is always a challenge. Social adds another layer of complexity and an exponential amount of data to parse. The social stream is swift and rich with knowledge, and search is the only realistic way of harnessing that information. In user testing of intranet search, we’ve found that it’s essential to provide a single, unified search across all intranet resources. This finding was replicated for social intranet features: they should be searched as part of the overall intranet search, rather than having individual siloed search engines for each social tool. Depending on the implementation, the need for integrated search can be a strong argument against outsourced or hosted social software, because many SaaS services don’t support federated search.

Business Needs, Not Tools or Fads

The most important conclusion from both research rounds is that social intranet projects must be driven by business needs — that is, the problem or pain point you’re trying to solve. The Canadian agency Klick, for example, was spending too much money on weekly status meetings. Developing an intranet collaboration solution for these reviews resulted in annual cost savings of more than $300,000. (A very nice ROI, given that the feature cost $53,900 to build.)

The reverse process is common, but deadly. Don’t start by saying, “Twitter and microblogging are cool, and I’ve heard good things about Slack.” There may be many good things about Slack, but that’s no reason to throw it onto your intranet. Maybe your business needs something completely different.

The proper sequence remains the same as it was 4 years ago: Business needs drive the required solutions; the solutions then drive the tools you should buy or build.

From Grassroots to Management Support

One major change in our second study was the level of organizational support for enterprise social features. Many of our original case studies were grassroots projects without official sponsorship. The attitude? Just do it.

Today, many companies see intranet information sharing and other social features as offering true competitive advantages. It’s not something to build because it’s fun — it’s a workday utility. Social tools are an expected part of a knowledge worker’s standard toolkit, and many executives recognize this.

Although many companies have official management support for social intranets, others are not yet ready. A company culture that values openness and communication is essential for social projects to succeed. Absent such a culture, adding deep social features is probably wasted effort as employees or managers won’t want to use them.

Widespread use of internal social media breaks down communication barriers. Although that sounds good, it can threaten people accustomed to having a monopoly on information and communication. Ironically, corporate communications departments sometimes resist the move to broader communication. They’re better served, however, in finding ways to increase new media’s value rather than trying to suppress it.

So, before implementing intranet collaboration tools, you must consider your company culture. If people are strongly committed to the “knowledge is power” tenet and don’t want to share, then sharing technologies will obviously fail.

Also, there’s still concern that, given social tools in the workplace, workers will fritter away their days and not get any work done. What we actually find — in companies with vibrant social platforms — is that employees are no more inclined to fritter away their work hours on non work-related communication with social tools than they were likely to do before these tools existed.

Silos Sprout Once Again

Many organizations with seasoned social intranets are starting to find that the social stack is yet another silo — that is, a place where enterprise technology is boxed instead of being integrated with other components. Along with this technology fragmentation comes a fragmented user experience.

The idea of social becoming another corporate silo is somewhat paradoxical. Social tools, by their very nature, are effective at breaking down silos based on role, title, and geographic distribution. However, the social stack — sitting side-by-side with the company’s existing intranet or portal — can become another silo of information that ought to be managed, ingested into search, and ideally, integrated into the greater enterprise information universe. Mostly, this isn’t yet the case.

Even companies with highly sophisticated social platforms are only beginning to tackle the hard work of integrating the social platform or toolset into the enterprise portal beyond merely creating linkages.

Other organizations recognize the challenge, but still keep the two separate and manage the consequences rather than face the complexity of marrying the two.

Social integration involves many challenges: there’s a UI challenge, and there’s an infrastructure challenge, and usually there’s a political challenge. While calling it a marriage between social and traditional intranets sounds better, the reality is that it’s a question of who eats whom. And that’s not an easy decision to make, either structurally or politically.

It’s important to integrate social features with the main intranet for many reasons, perhaps most importantly because it prevents burdening users with double work. Employees shouldn’t have to update their profiles or photos in both the traditional employee directory and a Facebook-like social-connection tool, for example, or be forced to know which directory to search in order to find the results they’re looking for. Despite the hard choices required to make this transition happen, the end game must be integration.

Full Report

The full 217-page report on social intranet features is available for download.