Intranet Personalization that Works

Gord| December 3rd, 2009 2 Comments »

This article originally appeared on December 3, 2009 on the Dachis Group Collaboratory.

---

ThoughtFarmer is an alliance partner of ours, and provides intranet software that fosters collaboration and communication. After many deployments, they have seen many challenges to personalizing the intranets of organizations. We asked Gordon Ross, the Vice President of ThoughtFarmer, to share his thoughts and experiences with us.

---

One of the classic intranet design challenges that never seems to get any easier is the design of the homepage.  Employee Communications wants corporate messaging and CEO announcements. IT wants applications and personalization features. Employees want the cafeteria menu.

Juggling these varied interests is tough. As the intranet moves from a few-to-many internal communications tool to a hub of social activity, collaboration, and community, the homepage runs the risk of becoming even more crowded, complex, and simply irrelevant to the everyday lives of employees.

The libertarians in the organization know the answer to this dilemma and have yelled it loud and proud for years: “Personalize! Let the user choose what they want!” The paternalists know they have the answer too: “Centralize and publish! Decide for the user!” And so the tug of war continues.
 
We’ve been of the mind that you can have the best of both worlds. Through implicit personalization you can turn facts that you know about an employee into a powerful filtering tool to provide them with relevant information. But before we tackle how exactly we’ve gone about doing that, let’s dive a bit deeper into what it means to be relevant in the first place.

Signal to Noise

As good, responsible information designers, we take seriously the issue of trying to reduce the signal to noise ratio on the intranet.

We have our work cut out for us: the highest possible ratio of signal to noise is desirable. But what is relevant? What is not? Relevance is a thorny word, one that’s used and abused constantly. But what does it really mean? I’m fond of this functionally oriented definition:

Relevance: something (A) is relevant to a task (T) if it increases the 
likelihood of accomplishing the goal (G), which is implied by T.

So what’s the likelihood that you, as an intranet manager or interaction designer, are going to be able to reliably predict which pieces of information will be required for the task an employee is currently working on in their quest to achieve a particular goal?

Getting Things Done

When employees visit the intranet, they are engaged in productive inquiry – an activity where they are deliberately seeking what they need in order to do what they want to do. Said another way, it’s not inquiry in the form of general curiosity, but inquiry in the service of wanting to get things done.

Getting things done in the modern organization increasingly entails the creation of information. Productive inquiry (embodied through the act of searching the intranet, for example) begets collecting or communicating or creating information. Getting things done creates relationship between people and information. And when the action happens in the context of a social intranet, one which has the capability of storing the actions of any given user, we suddenly know a whole lot more about that person and what may be relevant to them.

We have a couple of ways on how we relate people to information in ThoughtFarmer. Our recent homepage redesign is a reflection of our thinking on this subject in action.

The Proxemics of the Intranet

To understand how we relate to information and each other on the intranet, we sought inspiration from the “real world” of material objects and physical space and the pioneering work of American anthropologist Edward T. Hall, to introduce a concept of proxemics to the intranet. Proxemics was Hall’s contribution to the study of how people relate to each other interpersonally and socially through their physical proximity to each other. Hall identified four expanding zones of relation: intimate distance, personal distance, social distance, and public distance. Each of these distances represented boundaries of physical space from centimeters (intimate) to tens of meters away (public) and represented the ability to engage in certain relationship-defining acts between people across those distances.

Hall’s expanding zones of relation represent a nice metaphor through which we can look at the relationships employees have with each other, with their company, and their content on the intranet. Intimate distance represents information all about you: your page edits, your comments, your status, etc. Personal distance represents stuff that’s been done to you or your content by others. Social distance is everything within your network, including your management relationships and group / division / regional relationships. Finally, public distance on the intranet is everyone and their activity in the organization.

We’ve adapted this approach to the homepage to create a useful filter through which we can enhance the relevance of information, all the while maintaining a middle ground between user need and organizational need.

News items can be published throughout ThoughtFarmer as a default content type within the system. Users have the ability to publish news on the homepage directly, within their departments and divisions, for their project groups or as “informal news” via a blog post. Our first step towards implicit personalization was to merge these news sources together, providing users with a mix of “global corporate” material and “local departmental” material.
 
Instead of just broadcasting them public news (the corporation, its initiatives, news from HQ) let’s throw in a mix of social, personal, and intimate content as well. City newspapers have known this model works for years to capture the attention of their audience. International, national, regional, and local current events all wind up sharing space on the front page of your typical daily city broadsheet.

The second design intervention was the redesign of Recent Activity or our “workstream” feature. The signal to noise ratio of this feature was previously way too noisy. On installs of significant size and activity, there was simply too much content to comprehend for the majority of users and the filtering wasn’t intuitive.
 
In version 3.6 of ThoughtFarmer, we’ve organized activity into Your Groups, Status Updates, and All Site Activity. Your Groups aggregates activity that is happening in the groups you belong to. That might be a formal project or an informal community on the intranet (i.e.: Digital Photographers). Users have further control (explicit personalization) by being able to filter which groups they want to show/hide on the homepage.

Status Updates are Twitter or Facebook like user updates, answering the question, “What are you doing?” This feature is now a familiar design pattern within Enterprise 2.0 software suites, enabling phatic communication throughout the organization.
 
And finally, should you want to drink from the proverbial information firehose, you can browse All Site Activity, an aggregation of high priority activities merged from every user across the site. We’ve recently spoken with a few intranet administrators and managers that find this feed quite useful and fascinating. They observe it flow by in real-time throughout the day, occasionally intervening or helping users out in different areas throughout the site where activity is occurring.

The Power of Defaults

The debate about personalization vs. segmentation on the intranet has been much discussed and researched by many pioneering intranet designers and consultants. As keen observers of user behaviour in the real world, we believe that well chosen default options are a sound design strategy. Adoption rates of personalization features are low, driven by a lack of understanding of the business benefit from the user and the inertia of human nature to simply be lazy and accept defaults. By placing the user at the centre of the information universe and using their relationships to information and each other as the default filter, we can provide them with an intuitive view of their world, making significant progress towards our goal of a more relevant and valuable intranet.

Cognitive Friction and System Adoption: Inversely Related

Chris| December 1st, 2009 7 Comments »

Alan Cooper defines cognitive friction as “the resistance encountered by human intellect when it engages with a complex system of rules that change as the problem permutes.”

Cooper was talking about bad interaction design on computer interfaces.

Gentry Underwood of IDEO elegantly describes the impact of UI friction in this chart from his presentation, “How To Build Collaborative Software That People Will Actually Use“.

The message is clear: as user interface “friction” decreases, system adoption increases.

How much UI friction do you see on your intranet?

Intranet Survey Results from Webcom Montreal: How 2.0 is Your Intranet?

Chris| June 1st, 2009 Comments Off

At the Webcom Conference in Montreal last month I presented on Intranet 2.0 in 10 Not-So-Easy Steps. Throughout the day, Allie & I conducted a survey with conference attendees to measure how 2.0 their intranets really are.

Some of the questions we asked were:

• How often does your CEO participate in your intranet?
• How transparent is your intranet — can employees see who is adding the information?
• Can the average employee contribute to your intranet?

The results from the survey indicate that there is still plenty of opportunity for corporate intranets to be more inclusive, collaborative, and open:

• 61% said employees can edit nothing or very little on their intranet

It appears that intranets are fully transparent at about a quarter of companies:

• 26% said employees can almost always see who is adding or editing content on their intranet

Signals, a key piece of an Enterprise 2.0 intranet (read about Signals in this post), are used at about half of organizations:

• 42% of employees do not receive signals about changes to their intranet such as e-mail notifications or RSS feeds

An ideal Intranet 2.0 will see active participation from the senior team or CEO (step 9 of our 10 steps to Intranet 2.0 is “lead by example”). About half of CEOs participate in their intranet, but 45% never do:

The results from this edition of our Intranet 2.0 survey are quite similar to what we found last year in Boston at the Enterprise 2.0 conference. Compare last year’s survey results.

Video demo: Tree view navigation

Chris| April 29th, 2009 Comments Off

Easy navigation is a key area where ThoughtFarmer outshines all other intranet software. It’s not only easy to use, but it’s easy to build: because ThoughtFarmer is hierarchical, all navigation can be generated automatically.

There are several auto-generated navigation views, but the new addition in ThoughtFarmer 3.5 is Tree View Left Nav.

Video: Tree view left navigation

As shown in the video, the main features are:

1. Current page is highlighted
2. Child pages are identified
3. Sibling pages are listed
4. An expand/collapse icon appears next to each page with children
5. Click the page name to go directly to a page
6. Click the expand icon to browse the pages in a section before navigating

Read about other new features in ThoughtFarmer 3.5:

• Email distribution list management & archiving
• New people directory

The New Laws of Intranet ROI

bryan| April 27th, 2009 5 Comments »

How do you define the value of your intranet? The easy return on investment (ROI) of first generation intranets has been realized. Next generation intranets should take a cue from newer models of network value, such as Beckstrom’s Law, where network value is created by producing value  for employees.

Those early intranets had it easy.

Don’t get me wrong. I won’t belittle the blood, sweat, and tears that went into building them, not for a second. But that late 1990’s – early 2000’s dream was so easy to explain, so easy to justify, so easy to quantify. That low hanging fruit was so sweet and succulent. Those early intranets dripped value and oozed Return on Investment (ROI).

Going from no intranet or from a hodgepodge of skunk works internal websites to a single corporate intranet brought tons of value to organizations during that era:

• You mean we can post a single “interactive” (remember that word?) copy of the company directory where it can be used by all?
• You mean we can web-enable any flavour of internal application from timesheets to procurement to the payday 50/50 draw?
• You mean we can create e-learning modules that reduce the need for travel?
• You mean everything can link together from one spot?

You could hear the din of CIOs everywhere: “Sign me up!”

Achieving the bullets above were huge victories that delivered real benefits off the backs of TCP/IP networks, HTML, and web browsers. Prescient Digital Media’s Finding ROI White Paper is chock a block full of examples of the terrific value and ROI that was realized during that time. The value was there, even when so many of those early intranets accomplished tasks with atrocious usability and suspect reliability. It didn’t matter that they were only valuable because they did a new trick, not because they did it well – examples of what Allan Cooper (of The Inmates are Running the Asylum fame) calls “dancing bear ware”.

Undeniably, there was a lot of value in linking employees together in that one spot — particularly, it was thought, when those employees were in physically different locations. What was really convenient was the value of connecting all those people together was actually quantifiable using Metcalfe’s Law. It believes that the value of a network lies in the number of possible connections that could be made. Metcalfe’s Law dates back to 1980 but was popularized in 1993 at the dawn of the World Wide Web.

As stated in the Wikipedia definition:

The law has often been illustrated using the example of fax machines: a single fax machine is useless, but the value of every fax machine increases with the total number of fax machines in the network, because the total number of people with whom each user may send and receive documents increases.

(Please don’t get spooked by the math below. Stay with me.)

The formula says that for n number of network users, the number of connections is

n(n-1)/2

The closer n gets to infinity, the more the formula behaves as simply n².

Later, in 2001, David P. Reed argued that Metcalfe’s Law dramatically understated the value of networks. He believed that the value of a network was fundamentally in its ability to produce sub-groups, particularly in social networks.

Reed’s Law therefore believes that increasing the number of participants produces value much more rapidly than Metcalfe’s Law. For Reed, the utility of a network is expressed as:

2ⁿ – n-1

The closer n gets to infinity, the more the formula behaves as simply 2ⁿ

In their eloquent piece titled Metcalfe’s Law is Wrong, Bob Briscoe, Andrew Odlyzko, and Benjamin Tilly take aim at Metcalfe and Reed, arguing that the euphoria of these calculations contributed to the “build it and they will come” and growth before profits logic that characterized the dot com bubble. They fear that while that bubble is long burst, the thinking that led to it lies just under the surface:

Remarkably enough, though the quaint nostrums of the dot-com era are gone, Metcalfe’s Law remains, adding a touch of scientific respectability to a new wave of investment that is being contemplated, the Bubble 2.0, which appears to be inspired by the success of Google. That’s dangerous because, as we will demonstrate, the law is wrong. If there is to be a new, broadband-inspired period of telecommunications growth, it is essential that the mistakes of the 1990s not be reprised.

Briscoe, Odlyzko, and Tilly propose a much more sober formula than Metcalfe and Reed. They draw on Zipf’s law, commonly associated with the long tail concept and surmise that the second most important person in your network will have half as much correspondence with you as the most important person, the third most important person will have one third the correspondence, and so on. Extrapolated out their value of the network formula is:

n log (n)

In 2008, Rod Beckstrom, former Chairman of open source wiki supporter TWIKI.NET, co-author of The Starfish and the Spider, and former Director of the National Cyber Security Center (NCSC), added his own approach to the heap.

His formula appears complex, but is actually quite simple. It states:

The value of a network equals the net value added to each user’s transactions conducted through that network, valued from the perspective of each user, and summed for all. (Wikipedia)

In his presentation available on Slideshare, he starts with the value for a single user

V = ΣB – ΣC

Value = the sum of all benefit value of transactions – the sum of all transaction costs.

He gives the example of a book purchase:

V = $26 – $16

V = $10

If one could purchase a $26 book online for $16 including all costs, the network value provided is $10.

If we are able to add up the user value of all transactions for all users for a given time frame, we will have quantified the network value according to Beckstrom’s Law. The final formula looks like this:

Vi,j = net present value of all transactions of k = 1 through n to individual i with respect to network j
i = one user of the network
j = identifies one network or network system
Bi,k = the benefit value of transaction k to individual i
Ci,l = the cost of transaction l to individual
rk and rl = the discount rate of interest to the time of transaction k or l
tk or tl = the elapsed time in years to transaction k or 1

(Wikipedia)

Just to sum up, comparing all of the formulas, we might get a table like this:

	Formula n= number of participants	Value of 10 participants	Value of 20 participants
Metcalfe’s Law (c. 1980, popularized 1993)	Value is in the number of possible connections n(n-1)/2	45	190
Reeds Law (1999)	Value is created by the ability to form groups. 2n – n-1	1013	1,048,555
n (log n) Rule [Bob Briscoe, Andrew Odlyzko, and Benjamin Tilly] (2006)	To an individual, the value of each ranked participant is ~ 1/rank # n log (n)	10	~26.02
Beckstrom’s Law (2008)	Network value is the sum of benefits – costs for each individual in the network	Depends on the total value derived by each participant added together

So, why am I off in these weeds? What does any of this have to do with the value of the intranet?

While the difference in values found by the formulas above is somewhat interesting, what’s FASCINATING is the change in philosophy that the two newer models represent. Almost uncannily, they parallel the push towards Enterprise 2.0 capabilities.

The n log (n) rule and Beckstrom’s Law:

• Look at the network from the edge rather than the centre
• Calculate value from the user’s perspective, rather than the enterprise’s
• Quantify actual value rather than potential value

For many companies, Intranet 1.0, based on that 1990’s vision, has been built. Success has been achieved perhaps by making it available to all within the company or perhaps by migrating it to a stable CMS platform. From a Metcalfe perspective, mission accomplished. All that low-hanging fruit has been harvested.

From a Beckstrom perspective, however, your intranet may still be a primitive wasteland. Usage of the intranet may no longer provide utility. We are finding that the “why do you use the intranet?” question that we like to ask as part of intranet visioning research is eliciting more and more incredulous responses. Use of the intranet, contrary to how it is commonly positioned, is no longer an optional activity. In today’s corporate environment, the intranet is now often the ONLY available way to perform certain tasks, such as use the corporate directory. Our interviewees tell us that they use the intranet because they have to.

Indeed, recent Watson Wyatt survey results shared on Prescient Digital Media’s blog, reflect a similar sentiment:

• 80% believe intranet navigation needs improvement
• 50% don’t actually use their intranet on a daily basis
• 50% find search ineffective

They are using their intranet because they have no other choice.

Put simply, if employees can’t find what they are looking for or can find no reason to use it, your intranet is worthless. If your organization has yet to shift your intranet’s focus on to your employees, on building value for them, on helping them perform true collaboration within its confines, on helping them build the connections that they never knew they needed, then it is time.

Whether you blow up your intranet and call ThoughtFarmer, or work on improving the intranet you have, you must step into the shoes of your employees and see value through their eyes. The challenge will be great, because following the new laws of intranet ROI requires a significant change in thinking.