Sure, you can search Pinterest, but what makes the site so much fun is the exploration and element of surprise, in large part enabled by categorization of content by both the Pinterest team and the site’s users. As user trends continue to shift from search to discovery, creating the structure and process to support that discovery requires a sophisticated content strategy.

Taxonomy is about so much more than categorizing content. When crafting a content strategy, we consider the people, processes and technologies that support the content throughout its lifecycle. The same goes for creating taxonomy. As content strategists, we have to think about taxonomy from the perspective of what terms and structure will help the content perform best and support the organization’s business goals. We also have to consider the longevity and flexibility of the taxonomy.

The Taxonomy of a Pin

On Pinterest, users can explore boards in one of 32 categories—31 specific categories plus the catchall “Other”—and uncategorized boards. But when users create a board, they’re not required to assign one of the 32 taxonomy terms to their board. For someone exploring Pinterest by topic, this leaves a gap between content that exists related to that topic and content that has been categorized as that topic. This means that pins on a food-related board named “Recipes” or “Vegetarian Dishes” may not appear under “Food & Drink,” preventing people from discovering them.

Pinterest could help close this gap by requiring users to categorize their boards at the time they create and name them. Instead of requiring users to categorize each board they create, however, Pinterest’s strategy is to involve other users. When someone comes across an uncategorized board, they’re asked to help by selecting one of the 32 categories from a dropdown.

 So while Pinterest gives its user community a lot of free reign when it comes to naming and organizing content, this strategy is supported by well-placed guidance to help the community improve the quality and reliability of the content. Pinterest strikes a balance between flexibility and structure by involving users in enhancing site categorization while lowering the barrier to entry for users who would rather not spend their time categorizing.

Wondering how to bring together taxonomy and content to form a strategic user experience—and how doing so benefits your organization? Read on.

Taxonomy & Content Strategy: a Match Made in Metadata Heaven

Taxonomy can help support an organization’s content strategy, and vice versa, by:

• Helping users discover and interact with content that’s interesting and relevant to them. Taxonomy enables us to use related content to tell a story and keep users engaged. Want to increase the time people are spending on your site or the number of pages they’re viewing? Make your first impression, “This is great, now give me more of it.” Then use taxonomy to serve up related articles, photo galleries, videos, product descriptions and other content. Paired with an interaction designer, a content strategist can make recommendations for calls to action, prompts, cues and other microcopy that guides users through related content.
• Promoting older but still relevant content. Creating and promoting new content is important, but leading users to older content may also be part of your content strategy. As content strategists, we can work with information architects and developers to find ways to give prominence to content that may otherwise be buried in an archive. For example, for a series of reports usually listed in chronological order and filtered by date, a content strategist may use supporting research to recommend that users also be able to interact with this content by subject.

The Pew Internet & American Life Project does a nice job balancing the latest content with relevant information from its archives. Within each topic page, the emphasis is on the latest research, but users can also filter reports by content type (e.g., report, data set, infographic) and by year. Then, on each report page, the Pew Research Center serves up “Related Research” on the same topic, even though some of the related items might be from a few years ago. In the sidebar, a data point from the report serves as another method of encouraging visitors to explore additional content.

Each of Pew’s reports is served up with related—but not always the most recent—research

• Elevating content from deeper sections of the site. Even if you’re not a strict adherent to the three-click rule, reducing the amount of time it takes your users to discover relevant content can’t be a bad thing. Taxonomy allows us to showcase content that, on a sitemap, appears to be many levels deep on the homepage or secondary pages.
• Relating and reusing content across multiple platforms and site installations. For example, part of your content strategy is to build a stronger connection between your website and your blog, which just happen to be driven by different content management systems. Taxonomy can help. Assuming you’re using the same taxonomy terms in both systems, you can still dynamically relate content using a tool like RSS, pulling relevant blog posts into web pages that are categorized with the same terms.
• Helping our clients manage content now and over time. Content strategists, information architects and developers should work together to determine how taxonomy plays into the content management system, from creating content types to establishing workflows. Using dynamic relationships to populate content in multiple places means less work for the content administrators, who no longer have to update multiple pages or sites with the same information. And presenting content administrators with term lists instead of relying on them to enter metadata reduces human errors and inconsistency.

Keep in mind that implementing or revising a taxonomy can require change management. Choosing well-researched and tested vocabularies can support an intuitive user experience, but may also require some guidance—instructional content on the administrative interface, for example—for content authors and managers. They may be used to using the organization’s internal terms, not the terms site visitors are using when looking for information, to define content.

• Providing context for tagging. You may decide that users, not your organization, are going to define how your content is classified and labeled. Take a look at Goodreads, a social network for readers. You start with the simplest of taxonomies: three default bookshelves called “read,” “currently-reading” and “to-read.” From there, you can create and name other bookshelves, or categories, from the basic “science fiction” to the clever “it-was-earth-all-along,” and place books on more than one shelf.

This approach empowers Goodreads to support discovery in many ways. For example, on each book’s page, you can see a “Genres” callout showing how readers most often classified the book. You can also follow the “See top shelves” link for the full list of shelf names. Whether you prefer to find popular books by broad category or dig into unique, quirky lists made by other users, Goodreads provides ample opportunity to do both.

Genres and bookshelves for The Hunger Games.

Whether you call it “folksonomy” or “social tagging,” your role as a content strategist is to provide the context to empower your users to make the best decisions about tagging your content. What information do they need to tag their content in a way that supports your business strategy, but with minimal effort on their part? And what information does your organization need to maintain, promote and otherwise take advantage of the content users have indexed? Going back to our Goodreads example, the site offers a set of “Bookshelf Tips” to help users get the most out of their categories.

• Empowering designers to create more engaging interactions. Taxonomy, together with other components of information architecture and data modeling, helps designers create the interactions that drive users find the content they need and want to share. Sorting, filtering and visualizing data wouldn’t be possible without the right content structure behind the scenes.

Because taxonomy can impact everything from interface design to content management system development, the best conversations about taxonomy and content strategy usually involve diverse members of your team. The information architect and content strategist should invite designers, writers, search engine optimization specialists, CMS developers, marketers and site administrators to contribute ideas and voice concerns. With input from your stakeholders, it’s time to get started.

Getting Started: If the Taxonomy Fits…

Tags or categories? Open taxonomy or closed vocabulary? How deep should your hierarchies go? Your content strategy should help drive which type of taxonomy to use when.

If your organization’s strategy is to build a collaborative community in which engaged users are creating content, then a closed taxonomy with a limited vocabulary may send the wrong message.

If you plan on creating content about the same subjects for the foreseeable future, then relating content through taxonomy can work well. But if the subjects will change often, then relating specific pieces or types of content to each other rather than linking them via taxonomy may work better.

You may also decide to limit your use of taxonomy, for example, if your organization is highly risk-averse and leaves nothing to chance. Relying on taxonomy-driven dynamic relationships, rather than manually creating the relationships between pieces of content, may not be the right content strategy for you, since you lose control over exactly what displays where. When a database, rather than a human being, is creating content relationships, the results may be humorous or even inappropriate.

When it comes to developing the list of terms you will use, a card sort can be a good starting point. Through this usability method, you can learn how users would organize your content and what labels they would assign to each category—information commonly used to inform sitemap development but just as useful when building out a taxonomy.

What Are You Waiting For?

Gather a multidisciplinary team and look for opportunities to integrate your taxonomy and content strategy. Get up close and personal with your content management system to see how you might be able to create more dynamic relationships between content. Review your archives and dig deeper into your sitemap to see what content deserves a promotion. Figure out where you’re still making updates manually, and see if introducing a taxonomy can help reduce the time you spend administering your content across channels.

Your rewards? Engaging interactions, consistent content and happy content managers, to name just a few.

Getting outside our routine surroundings takes a little effort. You have to ask for names of people to talk to and find out if they’re willing to talk to you. Much of this can be done by recruiters or social media these days, but I swear you have to get out and talk to people. Don’t let a glass pane materialize between you and the participant (read: camera lens, laptop screen, touch screen). Technology puts us at a remove from the people we want to empathize with. Conversation puts us in their head. When someone suggests, “Set up two video cameras, one at regular speed and one on time lapse to try to find patterns in behavior, and then do some casual shadowing observations without the cameras,” it smacks of being too caught up in the researcher/analyst role. You can only guess at what is going on inside people’s minds and hearts. You must ask them and listen to their stories to find out for sure.

How do we do this comfortably? Harness your natural curiosity about the way other people think. You like making things better—but for whom? Question yourself. Usually you can make things better based on your own perspective, but to understand someone else’s perspective you need to do more than observe and interpret.

• The design language or form is language itself;
• The medium is communication;
• The system is social networking and distributed messaging;
• The content is branding, sales, and marketing;
• And the interaction is the social and conversational call to action.

Advertising: from image to talk

The old (mass) medium form of advertising is the image. It is versioned for multiple broadcast media and addressed generically to appeal to the greatest number of market segment customers as possible. Messaging will often make lifestyle references, appeals to value and utility, price and quality, aspirational imagery, celebrity endorsement, and so on. Decades of fine-tuning supported by market research and results tracking have provided marketing with a virtually scientific sense of purpose. As simplistic as many advertisements and campaigns may seem, marketing writers operate with a great deal of confidence and awareness about what they are doing.

What then of the conversational style of messaging and promotion now so important in the age of social media? Branding, marketing, and sales function differently in this medium. They eschew the image for talk. Their appeal is more personal and is often targeted. Message distribution occurs not through media buys and placement but through the distribution facilitated by customers themselves. Success is attributed to virality — but in fact communication is far more nuanced and subtle than viral transmission. The call to action is not just an appeal individual action, but is an appeal to social action (sharing, liking, forwarding, recommending, tweeting).

The matter of crafting promotional word-of-mouth messaging for effectiveness across social media is completely unlike that of broadcast messaging. For it relies on the engagement of end users, and depends on their willingness to repeat the message or share the action in their own words (even if it’s just a matter of quoting, or retweeting), to their own peers.

Here, then, is a brief overview of the social interaction design considerations of writing promotional copy.

Social media messaging is relational

The crux of broadcast advertising and marketing is the impression. Messaging seeks, through its claims, its use of image and references, and its aesthetic or style, to make an impression. Audiences are expected to tell a commercial from fact, and so the entire form of communication is bracketed: it’s understood by all that a commercial is fiction. The aim, then, is to make an impression upon audiences such that consumers identify with the message, enough so to form their own mental associations and ultimately recall or even act upon the message broadcast.

This may just involve brand awareness, but more often seeks action: a purchase. The challenge in broadcast messaging is thus the production of messages that make their appeal most effectively to the greatest number of consumers in such a manner that their response to the messaging extends company sales and awareness. The “relationship” is between the brand and the consumer.

Social media messaging is different. The medium is personal, and so messaging arrives within the context of a social tools (twitter, Google, Facebook, etc) used regularly for personal purposes. The messaging is direct, insofar as it appears as a statement received by the user in a medium used for direct communication (again, twitter, Facebook, etc). Image and brand personality references are crafted with words, and so brands may avail themselves of imagery and motion story-telling (tv ads) far less than they are used to. Creative is invested completely in the two axes of social tools: language and speech. And distribution is not controlled, but is rather given over to the audience to execute according to the the message’s appeal.

Social media messaging is appealing

This matter of the appeal is a critical distinction between mass and social media. The appeal of mass media messaging is anchored on the impression made. The appeal of social media is anchored on the user’s acceptance of the message. More specifically, the distinction is between a broadcast message invested with meaning, and a social message whose meaning is invested by the user. Power shifts from the brand to the consumer. Expression (branding) is replaced by Interpretation or Reception. The broadcast message may seek to make an impression; the social message seeks the user’s engagement. From a communication perspective, the operation of commercial promotions is reversed: brands do not supply the meaning, users do. Only then may they be expected to pass it along. Only when they accept, like, and see themselves in the messaging do they make it their own and share it.

The construction of social messaging requires a shift from objectively-framed communication to subjectively addressed communication. Broadcast messaging creates a fictionally objective claim (seeking to make an impression, an image). Social media messaging seeks to address the end user and relate directly by means of a pseudo form of speech: speech by proxy, if you will.

Addressing the (message) envelope

This means that the writer may employ objective claims, as in writing headlines and news-like statements. Or may “speak” more personally, making use of I, You, We, They, and Us. Those are linguistic elements rarely deployed in broadcast media. Most community managers and social media managers do this without thinking about it. We talk without reflecting on the constructed-ness of our speech. But that’s not to say that there’s no order or design involved in speech. All speech employs the order and construction of language, in both its form and content.

Speech is addressed, to somebody, to everybody, to a group, or even to oneself. The use of “I” is unnecessary on twitter and in Facebook updates, because the medium implies it. But “I” can still be of use, to change both tone and appeal. The linguistic effect of using “I” or “I” references (me, my) changes effect. For example: “Support relief efforts in Haiti” vs “I would really like help in supporting Haiti relief efforts.”

Similarly, use of “you” directly appeals to the reader/user. As in, “I would really like your help in supporting relief efforts.” Use of group references, such as to include by suggestion, makes use of the addressing of speech also. As in, “We should really get together to support Haiti relief efforts.”

These kinds of addressing functions of speech are used with risk in broadcast media, but are less risky in social media. The use of addressing (I, we, you, they, us, etc) in broadcast makes an implication that many in the audience may reject. There’s no personal relation in broadcast, and thus the appeal is risky. But in social, it extends an offer of relation and when appropriate (to the brand or claim it makes), can succeed precisely because it is relational — it’s personal, where broadcast can’t be.

Message content and linguistic expressions

If addressing involves the envelope of communication through social tools, then content is where references and claims are made. The message comprises of a statement or expression within which a claim is made. Here, the medium offers a greater range of differentiation than broadcast. First of all, a brand or company may craft many more messages, providing many more claims, than in broadcast. Claims can thus be made to appeal to the interests of a greater number of consumers/users. Claims based on price, utility, value, timeliness, and much more. Claims directed at personal interests, social values (causes), cultural affinities, news, and so on.

These content claims can be presented or expressed in social media messaging by means of objective statements or personally and socially addressed expressions. Here we encounter some basic linguistic types of expression. We have:

• statements of fact;
• requests;
• questions;
• answers;
• invitations;
• offers;
• announcements;
• recommendations;
• and more…

Message content and the claims of content

Any of the claims you wish to make, as the informational content of your message, may be expressed differently by use of the above types of linguistic expression. Consider:

• “We (the organization) would like your support of our Haiti relief efforts,” (appeal for help and appeal to include self in organizational efforts)
• “Our relief efforts are getting underway. Help us now!” (announcement plus direct appeal)
• “Relief efforts make a difference with the help of twitter” (factual statement, no direct appeal)
• “Join in our relief efforts today” (invitation)

In addition to the addressing and informational, or content claims, of social messaging, there is the call to action. All internet advertising and marketing works by call to action. But in social, the call to action is social — it hopes for personal response, yes, but seeks social engagement. The social call to action is unique in that it asks the user to personally repeat or distribute the message to his or her peers. The commitment hoped for is greater; net effectiveness is supposed to be higher. Because the message is distributed by means of the highest authority there is: one’s own relationships with friends and peers. Personal relationships, not media, are the backbone of this kind of marketing.

Completing the message: call to (social) action

The call to action is often a link. It may be accompanied by information that provides reference and context, or not. But the call to action may be the expectation of retweeting. Or, in the case of Facebook, of liking. Brands often want users to complete the call to action based on the content of their message. But the user completes the call to action only if his or her interpretation of the message resonates and agrees with his or her interests in it. Again, the power is reversed, from expression and image-based messaging to interpreted communication. Actions are taken by users when they feel like it. When they feel like acknowledging a message; when they feel like being seen to acknowledge it; and when they wish to include others in it.

Here, then, commercial messaging wants to be as least commercial as possible. Many call this “authenticity” and “transparency,” but lets not fool ourselves. It’s commercial, and everybody knows it. So use either great creative, compelling narration, use news, or values, or discounts and package them communicatively.

Take, for example, the discount or offer. Social media are rife with discounts and offers, and companies such as Groupon, Yelp, and Foursquare (to name but a few) are heavily invested in the value proposition of social shopping and social commerce. The challenge here is to make the act of shopping a communicative act. It normally isn’t. In some cases, the challenge is in fact to make the act of window shopping a communicative act — as when companies hope that users looking for products socialize their efforts and solicit participation from friends or peers.

Transformation of this act into a social and communicative exercise demands use of the medium’s communicative possibilities, in short, the messaging variants we have described above. If I purchase flowers online, and am given a message after the purchase to share it with friends, the more effective use of messaging would focus on giving, reaching out, sending greetings, bringing a smile to somebody’s day — not the more conventional but commonplace “Save 30% on flowers today.” That doesn’t communicate beyond the value proposition, which is unlikely to be of interest to most of my friends at any given time. More likely to communicate personally would be a personal expression to which the purchase of flowers is simply a reference: “I just made somebody smile. Guess who? (link)” And so on.

Summary

In summary, the creative work of commercial social media messaging extends beyond the craft of broadcast advertising and marketing. The medium used is a form of mediated speech, and its success depends not on its image appeal but on its personal reception by users. Action is solicited on the basis of claims that engage a user’s interests and which are communicable. A number of forms of linguistic expression are available as means of “designing” the message envelope and contents. And the informational, or value propositions contained in the message may be highly differentiated for the purpose of targeting and reaching different types of audiences and their peers.

There is order, structure, and design in all things. In the facilitation of social interaction, and communication, too. Especially when it’s commercial.

Mashable’s post on the social media Sharepocalypse has caught everyone’s attention. Author Nova Spivak breaks down the issues social media users face in the sheer volume and diversity of sharing activity across our favorite social networks.  And comments on some of the resources and solutions that may be on offer if “social assistance” services can deliver effetively.

I was on a different topic recently, that of scaling and population, when I got to thinking about noise. Much of the sharepocalypse problem, I think, comes down to noise. Noise, because there are often motives behind social sharing. Motives that suggest that the act of sharing often means more than meets the eye.

This is interesting, because if sharing produces content, and if the sharepocalypse concerns an excess of content and content sharing activity, then it’s not just the volume of content that needs addressing, but the intentions of those who share. Sharing, after all, is a social act.

It’s all about sharing…

There would be no sharing if there were no friends, peers, colleagues, and fans to “consume.” And likely much less sharing if there were no measurement of sharing activities: no new followers, friend requests, comments, likes, +1s and so on.

Not to mention the meta message of sharing metrics, of which Klout is the best example. Our activity and the responsiveness of our “networks” are transformed into a meaningful number — an “influence” metric, or klout.

Point being that the act of sharing is not just an act of sharing content. It’s a social act, and social acts solicit some amount of acknowledgment and recognition. Receiving that, they can become communication (as happens when any two or more people engage in an exchange).

Content, then, is often the vehicle for a communication not yet established. It’s the opening move, if you will: the statement or expression.

The content is the vehicle

It belongs to human communication that we are able to distinguish an utterance from the thing uttered (the claim). We can tell the meaning expressed in talking from the actual sentences and expressions used. In the case of sarcasm, for example, we know that the meaning intended actually contradicts the the expression.

And this applies, to some degree, in online sharing. Knowing our friends, and less so our peers and online social connections, we’re often able to tell what a person intends when they share. The content is the vehicle, not the conversation. And in fact, content often opens up comments and exchanges permitting all involved to relate something of their own.

Content shared then is often just the ice-breaking move in social exchange. It’s the starting point, the springboard, and the context. And it’s fine, generally, if talk moves past the content itself to other things.

Noise is the problem

Which brings us to noise. Noise is the problem. Some hope it can be filtered out, say algorithmically. Algorithms may be written to anticipate the individual and personal preferences of a user. Or to collect information from aggregated activity. So individual vs a social approaches.

Noise might also be reduced by means of services that sit on top of sharing networks. This is the social assistance idea noted by Spivak.

But there’s still the matter of noise and why it is an unavoidable byproduct of social sharing. This has implications for the feasibility of noise reduction.

Social networking platforms can be viewed as social systems — a combination of mediating technologies and the practices that emerge around them. They’re self-reproducing systems: that is, it’s the constant social activity of users that keeps them going.  My thought is that if a social system reproduces itself by means of mediated interactions and communication, different types of noise are produced.

The noise of redundancy that results from distribution of activity across tightly connected social networks — a kind of noise that would not trouble situated and co-located “real world” interactions. Call this the noise of amplification. It exists because content and communication rapidly escape the site of their original production and “appear” elsewhere. (Face to face talk is governed by the physical distance in which your voice can be heard.)

The noise produced by an attention economy. This being noise resulting from the online social condition that only activity can get attention. One has to post and share in order to have presence. Here the act of sharing is what matters, less so what is shared, for the act maintains presence and creates the contexts around which others can engage.

The noise of system self reporting. This being notifications, which are system messages reporting on user activities but not authored by those users (Bill is now following you). Facebook was built on this (“Jill uploaded a photo” creates social activity by proxy, leading to more activity by those who respond to it).

The noise of bots and non-human accounts. Twitter is the most guilty of this, but wasn’t the first to allow it. (Remember Fakesters on Friendster?) This noise helps to circulate news, but results in a kind of tolerably false communication.

The noise of obligatory social etiquette. This is the noise created by adhering to online social norms and conventions, such as following back, or adding to Circles, reblogging, liking, and so on. (Social gestures — likes — have communicative purpose.) Many of these acts are simply baseline social etiquette and whether they pay off or not, are the online social equivalent of buying a lottery ticket: your chances of winning increase dramatically when you buy a ticket. A social act that has potential.

So given these different types of noise, what are the prospects for smart noise reduction? Content shared is hardly just content shared, but is almost always a form of social action. Can the social acts be separated from their contents? Should filters be designed to sift out bots? Why not then sift out users whose social media use is primarily promotional?

Or the reverse…

Or the reverse: sift out content that’s intended just to network and connect, but which has little news or information value? There could be so many further ways to tweak filtration, based on person, content, genre, timing, status, relevance, personal preferences, social preferences, recent activity, etc. It’s mind boggling.

Sharepocalypse is just the tip of the sharing iceberg. The flotsam and jetsam that drifts downstream in a medium that never stops flowing. But the currents beneath are deeply social and mean far more than meets the eye. It’s going to be hard to sort through all that noise. Because collect the empties as you will, more often than not, there’s a message in that bottle.

Indeed it is. Every interaction people have with your website is an opportunity for your organization to improve, hurt or confirm your credibility with them. When your content doesn’t support your business/organizational goals and provides zero value to your customers, you will end up with more unpleasant conversations and missed opportunities. You need a content strategy. So where to start? At the beginning, of course. But what is the beginning? You are in the day-to-day grind of maintaining your current site, fixing the most “urgent” issues rather than addressing the underlying problems that led to today’s real and perceived crises. The process of creating a content strategy was very rewarding, both for the Web team and the people we worked with internally. We discovered insights about how the website could support the goals of individual business units and the corporation. Stakeholders realized that there was value in creating content that people wanted to read, view or respond to. But you can’t to results without starting.

1. Realistically Assess Your Situation

Can you finish what you start?

At Cerner, we knew that there were problems with our content in the spring of 2010. We “knew” it in the sense that we got complaints about it, we noticed inconsistencies and inaccuracies in how we talked about our solutions and services, and our analytics told us visitors didn’t care to read, consume or react to what was on the site. But we really didn’t know what it would take to fix everything, or even what “everything” was. Before anything could be done to fix our site, there had to be a commitment to do so. This sounds simple, but it’s not. Is your company/organization willing to let someone (or several someones) spend the time necessary to dig into the underlying problems with the current content? Will your company let that person or persons create a plan for fixing it and then go do it? If not, is the company/organization willing to be influenced on the importance of content? If the answer is still no, then your default content strategy will be to have no strategy at all. For us, fortunately, the answer was yes. In 2010, we started a redesign of our entire site. The redesign was not limited to just the visual design and a new technical platform, although those were important considerations. We wanted to create an entirely new user experience, which would not be possible without completely re-thinking how users found our content, and what we actually chose to say (and not say) with our content. The content portion of the project was not just added on as an afterthought, but was included from the very beginning of the discussion of what we wanted the website to be.

2. Build Internal Support

Who cares about content—and can do something about it?

So great, you have permission to create a content strategy, whatever you decide that means. It’s one thing to have approval from your manager to do so, but what about the other areas of your company or organization? If people outside your team are involved in the content process (most likely creation or review/approval), you need them to support your efforts. To get their support, this group needs to care, or why they should care. We had about 30 people across the company responsible for providing or overseeing the content for their group of solutions and services. These people were not/are not writers or Web content experts, but implementing a content strategy would be impossible without their support. We targeted a small sub-set (three) of these people (we’ll call them content contributors) who we deemed as influential among their peer group. In individual meetings, we discussed the idea of putting a strategy behind our content. We didn’t use the phrase “content strategy” in the meetings; instead, we talked about their business goals, and shared ideas for how content could help them meet those goals. These people were not executives within their given organization, but rather those who would most care about content in that organization and who would be in a position to take action on it. What we discovered:

• Content contributors were desperate for guidance on how content could support their business goals;
• Governance had to be included in our content strategy. Governance by itself is not a strategy, but our content contributors needed a clearer process for governance of our content;
• This group would start to really care about their content when we gave them a reason to.

This sub-set of content owners served as advocates down the road when we shared our plan for content creation, maintenance and updating during the redesign process. By introducing the idea that content was important to an influential sub-set of this group, we faced no opposition to changing “how things had always been done” when meeting with the full group. With the internal support in place, then it’s possible to get the subsequent work done.

3. Define and Prioritize

How will you get the work done?

I listed this after building internal support, but in reality this step can and should occur in conjunction with creating internal support. While we were meeting with the sub-set of content contributors, we identified the key problems areas (both quantitative and qualitative) with our existing content. We took the time to catalog and inventory ALL of our content, which at the time numbered about 7,500 pages, including our nine global sites. Taking inventory of your content is time-consuming and hard work. But you have to do it in order to define the problems with your content and prioritize how to fix them. There are numerous examples on the Web of what your audit should include, from a spreadsheet to more sophisticated software tools. Devote some time looking at them (hint, a spreadsheet will probably work for you). Ask yourself how that approach would work for your website. Ask people who work on content strategy how they handled a content audit. We decided to capture the following information in our audit:

• Page Title;
• URL;
• Metadata keywords;
• Last date page was updated;
• Who updated it?
• Did the page include an image? Video? Downloadable PDF?
• Was it any good?
• Is it written in appropriate tone and voice?
• Does it contradict content in other locations on our site?
• Is it just a regurgitation of a flyer or product spec sheet?
• Is it written to inform a visitor, or to please a product manager?

The last item (was it any good?) took the most time to complete on our audit. But you aren’t doing an audit without answering that question. Otherwise, you are just filling out a spreadsheet. Yes, quantitative data was essential to understanding our overall content issues, but ultimately you need to make judgments about your content. Our content issues immediately surfaced once we completed the audit. And we now knew exactly how bad the problem was (or wasn’t). For us, these issues were are biggest challenges:

• Lack of tone and voice – For example, we might talk about our radiology solutions in an informal, conversational way, with the content focused on how it might benefit the end-user or health care organization. But on our page for our pharmacy solutions, we would list 15 bullets of feature functionality, with no consideration of how this solution benefits the end-user or organization. So are we a company that wants to talk only about how great we are, or are we a company that is focused on helping our clients improve how they care for patients? Can visitors to our site believe what we say about our solutions? Are we credible? It was hard to tell;
• Bad metadata and information architecture – Incorrect metadata of course has the consequence of making your content harder to find by search engines. Beyond that, if you are sloppy with your metadata, page titles and naming of sections on your site, it can give the user (even subconsciously) the idea that you don’t take content seriously. If you won’t take the time to make sure that pages are named consistently, then should visitors to your site take seriously what you have to say? For many of you, this might be stating the obvious, but don’t farm out the metadata and naming conventions to your technical team (unless they are content champions). Take the time to understand how your content management system works and how to fix metadata issues;
• No standards – A user would have no reasonable expectation of what they would encounter on types of pages within our site. It became obvious that we needed a standard for what type of information a user should expect to see on a solution page vs. an event page vs. an executive biography. This includes the words, images and video, as well as the font types, sizes and color used on each of these sections. These things matter. Take the time to consider them and stick to them;
• Governance confusion – Some groups were allowed (more or less) to publish content on the site with little oversight, while others went through a semi-review process before publication. This issue was an underlying reason for all of the previous three issues.

There were other issues, of course, but these – if corrected – would most improve our content, support the website and business goals. Now it was time to find out if our plan would work.

4. Demonstrate Success

What results did you get?

To make content strategy an integral part of your company/organization, you’ll need to show it’s worth the time and effort. Here are a few examples of success we attribute to the content strategy work.

• Users more easily find our content – Prior to the redesign of our website (including a strategy for content), less than 20 percent of our visitors came to the site via search engines. Within a few months of redesigning the site, that number increased to 30 percent of our visitors, and today we are approaching 40 percent of visitors to the site from search engines. This represents new groups of people finding our content and interacting with us;
• We increased Cerner’s credibility online – Our content strategy included the creation of a corporate blog. We wanted to use the blog for our own industry experts, as well as selected clients, to talk about key issues in health care. The blog does not overtly tout our solutions, but focuses on sharing our perspective on key health care issues. Recently, Investors.com (part of Investor’s Business Daily) referenced a Cerner blog post in a story on ICD-10, a change to how diseases are classified by health care organizations. Providing credible, relevant content resulted in a third-party recognizing us as a source for information;

Figure 1: Investor’s Business Daily cites a Cerner blog post

Figure 2: The quality content of this Cerner blog post by Lisa Franz led to its citation by Investor’s Business Daily

• Our content resonates with users – Six months after we redesigned our website and implemented our content strategy, we provided feedback to our content contributors on what types of content resonated with people. Product pages with video (and especially those with client testimonial videos) received more visitors than those that didn’t. And visitors – on average – spent significantly more time on the product pages with videos than those without videos. (See Figure 3) For us, more traffic + more time spent consuming content = win. For those of you who rely on other groups in your organization to supply content, you must give them guidance on what works. At the time I shared the video results with our content contributors, we had just 13 product pages with video. Today, that number is more than 30.

Figure 3: Adding quality video to our product pages led to more visits and to users spending more time on those pages (image credits: John O’Nelio)

Get to Work

I came across a quote from Thomas Edison during this project.

The first requisite for success is to develop the ability to focus and apply your mental and physical energies to the problem at hand – without growing weary. Because such thinking is often difficult, there seems to be no limit to which some people will go to avoid the effort and labor that is associated with it.

Applying a content strategy to your website requires work. As you begin the process by building internal support and evaluating your content, the path toward a content strategy that will work for your organization will become clear. It’s likely you’ll realize additional benefits, too. In our case, the relationship between the Web team and content contributors – and the business units they represent – is much stronger due to the successes achieved. Discussions about content occur (more often, not always) during the planning stages of projects rather than the middle or end. This approach is one that worked for us. But none of it would have been possible without getting started.

A Few Resources That Helped…

• Kristina Halvorson, The Discipline of Content Strategy – A List Apart, Dec. 16, 2008;
• Colleen Jones, Content Analysis: A Practical Approach – UX Matters, Aug. 3, 2009;
• Erin Kissane, Content Templates to the Rescue  – A List Apart, July 7, 2009;
• Kevin Nichols, Examples of Content Audit and Inventory Tools;
• Content Strategy Knol (Google)

I’ve been watching two trends recently in the realm of digital product development. First is the incorporation of gaming concepts into products that seemingly have nothing to do with gaming. Second, the importance of designing products that are not only easy to use but a pleasure to use.

To be sure, these trends aren’t new. My point is not to shed yet more light on what we already know. Rather, the potential impact of these trends as they go mainstream is significant for UX designers– our skills are becoming strategic rather than tactical.

Let me explain. A wireframe is a tactical output that (hopefully) partially fulfills a strategic direction for a system. But working with a product manager to figure out how to incorporate gaming concepts into a product moves us, the UX designers, in a strategic direction. This changes the opportunities in front of us as designers. The term I use to encapsulate these opportunities is “digital product strategy.”

What is digital product strategy?

Product strategy binds business strategy to product management. Marty Cagan put it nicely in a April 2009 blog post: “Think of it this way. The business strategy and business portfolio planning provides a budget and a set of business metrics. The product organization then lives within that budget to pursue as aggressively as possible the best ways to hit those business metrics.”

Product strategy (let alone digital product strategy) is a relatively unused term – no Wikipedia article exists as of yet, and it ranks fairly low as a competitive keyword (at least as I write this). As such, there’s not a lot of consensus as to what it encompasses. So, I’ll provide my thoughts with an emphasis on products that are digital by design – they make heavy use of software as part of their interaction model or delivery mechanism.

To me, a good digital product strategy brings together seven areas of expertise:

1. Market/industry expertise: A deep understanding of the domain you are engaged with;
2. User expertise: Engagement with actual or potential users of the product;
3. Competitive expertise: Commitment to finding “sustainable differentiation” – the “secret sauce” that cannot be easily copied;
4. Related-Industry expertise: Engagement with other industries or markets that you can learn from to create a better product for your industry
5. Design expertise: knowing how to make a product easy and fun to use with the latest design techniques for many different devices
6. Technology expertise: Knowing what is technically possible today and in the future and the devices that make sense
7. Business expertise: Knowing how the product will fit into the operational realities and capabilities of the business

Here’s an example, I have been working with a customer over the last few years to help them introduce a direct business-to-business channel alongside their traditional distributor-based model. The main channel became an e-commerce website, and our “product strategy” was about achieving parity with two main competitors. From a digital product strategy perspective, the website became the primary delivery mechanism for a tangible product, and thus a huge part of the product UX.

As we emerged from the parity phase, we consciously moved to an innovation phase. What once appeared as “solutions” in the first phase – an e-commerce website – looked like a limitation when we focused on the market and the users from this new perspective. Customers were using the website during lunch hours, and we knew that they were walking from the storage cabinet to the PC with written notes about what they need to restock. These two things pointed to a fundamental inconvenience in usability. This inconvenience couldn’t be fixed by a more usable website, however. It was a great opportunity for a mobile application with a bar code reader for replenishing inventory.

Had our product strategy remained focused on the website, we would have run repeated usability tests to fine tune the features, and we would have continued to focus on competitors to keep up with new features. The idea of a mobile app that really addresses that fundamental inconvenience wasn’t possible until we shifted our perspective. By combining our knowledge of the market, the users, existing technology capabilities, and design expertise, an innovation became imaginable. A new product is conceived that can move into the more traditional product management processes.

Product Strategy and User Experience Design

The example above points out how UX design is a strategic skill. As the realization of business strategies become more dependent on the development of digital products (or products that make heavy use of digital technologies), the UX designer offers the unique combination of:

1. How to understand the real life of users;
2. The capabilities of technologies and devices;
3. How to make something easy and fun (or at least really convenient) to use—three of the seven areas of expertise I described above.

This moves us closer to business strategy and simultaneously requires a change in our deliverables. In an earlier article, I discussed how the requirement is a “somewhat strange and antiquated way to capture what a software system is supposed to do.” We have to develop new deliverables to replace the requirement as the first and best way to express the system we want to design. Conceptual wireframes, sketches, storyboards, and user models (like the famous Flickr model) are more appropriate deliverables for product strategy work. Companies that consciously do product strategy as a discipline know this, but there’s plenty of opportunity left in the mainstream projects many of us work on each day.

As such, we supplement the work of the CTO, whose job is to set a technical direction. Martha Heller describes this role well: “… the digital product groups hire a CTO, who designs and executes against the digital product roadmap.” In other words, the CTO is the technical expertise part of digital product strategy, while UX design is the easy-and-fun-to-use part and the knowledge-of-real-users part.

UX Design, Product Strategy and Gaming

A new opportunity exists for the UX designer as gaming concepts become part of product strategy. Who else is better equipped than the UX designer to bring this discipline to the table?
To decide that a product is going to be structured as a game rather than, for instance, a document sharing system is a strategic product decision, not a tactical one. When we start thinking about incorporating gaming concepts into our products to increase engagement, we’re making fundamental decisions about our products.

A lot of people are talking about gamification of digital systems. I could choose any number of people to quote about the fundamental structures of good games and how they can be applied to digital products. I like the succinctness that Jane McGonigal provides, so I’ll use her definition: “When you strip away the genre differences and the technological complexities, all games share four defining traits: a goal, rules, a feedback system, and voluntary participation.”

There are two conclusions to draw from this for UX design:

1. These are not simply features we add into our digital products; they invite us to think about our products in a fundamentally new way;
2. The UX designer is the best equipped discipline to bring the full force of these concepts to the product strategy conversation.

Blending these traits into an engaging and compelling UX – that is fundamental to the product itself – is really what the UX designer is equipped to do. That companies are now getting on board with the engaging power of gamification in formerly utilitarian software systems yields lots of opportunities for our once tactical discipline to become strategic.

For a long time, using a computer was a merely cognitive thing to do – a dialogue. Those were the times in which computers were sitting at desks, and were merely operated through buttons and mice, and displayed information either as text or graphics. This paradigm has already fallen, now that computers are ubiquitous in our everyday lives, and other paradigms are perhaps soon to follow.

Recent developments in the research field of Human-Computer Interaction point to emerging styles of interaction that make use of our very abilities as human beings, putting us directly in touch with the digital world.

Embodied interaction

If we look at the ways in which we interact with computers across the last three decades, a number of major changes become evident: In the 1980’s, we have interacted with computers in textual ways. In the 1990’s, we have changed to graphical interaction – and in the most recent decade, again, these ways are changing.

One theory, as proposed by Paul Dourish in 2001 in his book ‘Where the Action is’, to conceptualize these recent changes, is ‘Embodied Interaction’. Dourish points out that computing is moving into the social and the physical space, and he proposed the term ‘Embodied Interaction’. Advantageously, these new styles of computing draw upon skills that we already have, skills that we, as human beings, embody.

In this article, we will look at three series of prototypes that illustrate what ‘Embodied Interaction’ is – in the different physical and social spaces that we live in.

How can we make digital content graspable?

Nowadays, our lives take place in two worlds: on the one side in the digital world, and, on the other side, in the physical world. While many things happen in the digital world, and while it is in many ways an influential place, things in the digital world are not tangible for us. Many people do not understand the digital – it is hard to grasp.

As humans, we are very skilled in engaging with the physical world. So the question is: How can we use our everyday, real-world skills to interact with digital contents? How can we move things from the digital world into the physical world? If we look at emerging interaction techniques, it is obvious that this is already the trend: the iPhone’s touch and the Wii’s bodily activity make clear that interacting with the digital is getting more and more physical. The question is: what’s next?

In our research, we have developed a series of three prototypes that provide a tangible glimpse on how the digital could be made graspable.

The first prototype involves weight as a representation for digital content. It is a mobile phone-shaped box that, on its inside, features a motorized weight. This weight can be moved, and thereby the device’s center of gravity can be moved. This allows for a variety of applications:

Firstly, the device’s ergonomics can be changed. If it would usually fall out of the user’s hand, the center of gravity can be moved to balance the device automatically. Secondly, content on the device’s inside can be made feelable – especially the distribution of content ‘in the device’ can be represented, e.g. if the majority of content is on the device’s right side (for instance, in a list of songs), the device would also be heavier on the right side. Thirdly, a shifting weight can be used to represent contents that are external to the device, in a certain direction. This could be useful for navigation, in which the device would physically point, by shifting its center of gravity, into a direction that would remain the same when the user turns: as a haptic compass.

The second prototype draws upon shape change as a means of displaying digital contents. It is equipped with a set of motors that allow for the actuation of the device’s geometry. In doing so, they allow for making the device thin in the pocket, and thick, ergonomically shaped when held in hand. Furthermore, they allow for the physical display of content amount or direction, similar to the previously described weight-based variant. More content can simply be thicker, as in an e-book that has, when read from the beginning, all of its pages on the right side (resulting in a feelable thickness on that side), slowly moving over to the left side while progressing through the book. Also directional information can be conveyed, shifting the device’s thickest point into the desired direction.

The third prototype of this series draws on our human ability of empathy – as social beings, we are well able to feel how other people or beings are. Inspired by this ability, the ‘living mobile’ provides information about missed calls or text messages through breath and pulse. The device vibrates in a heartbeat-like manner, and has a motorized chest that moves in a breathing-like movement. In cased of no missed calls or events, the phone will behave calmly, while it will utter excitement in its movement if it needs the user’s attention.

How can we make mobile phone calls more polite?

The blurring of the digital and the physical can be observed not only in interacting with content – it is also visible when it comes to interacting with other people through digital channels. Here, these channels may interfere with the ways we normally communicate, and thereby result in impoliteness.

A particular social problem of mobile phones is the issue of incoming calls in busy situations. Even though users are able to tell who is calling, they generally do now know whether the matter at hand is important or not. Similarly, users may find it inappropriate to call just for a chat, as they are unable to express that their call is not important.

Users often find themselves in a conflict when noticing an incoming call – should they interrupt what they are doing, and take the call? Or should the reject the call? The latter is often considered impolite, given the fact that the caller not even had a chance to express what the call is about.

To overcome this issue, we have built a prototype that employs a pressure-sensitive dial button. A more important call can be placed by pressing the button stronger, while also gentle calls can be placed – by pressing the dial button only gently. The prototype also has a filter, allowing to change vibration, ringing and also voicemail behaviour depending on the urgency of the incoming call.

How can we make phone calls more emotional?

Giving users a feeling for digital content in their device shows that there is great potential in making the digital physical. But in mobile phones, the even more relevant field of research could be another one: giving two users a feeling for each other.

In their current form, mobile phones are well-suited for one of the major reasons of telecommunication: to exchange information. For the other major reason of telecommunication – the whish for nearness – speech, video and text might be not all we can do: sometimes, we just want to be in touch. We developed a series of prototypes that explore this field.

The first prototype, the ‘grasping mobile’, telecommunicates pressure and potentially allows for an experience of holding hands over a distance. It is equipped with a strap on its backside, into which the user’s hand is placed. The pressure exerted by the telecommunication partner’s hand on their phone is then telecommunicated through a motor that pulls the strap tighter, and vice versa.

The second prototype, the ‘kissing phone’ telecommunicates moisture. While there are many people one may not want any kind of ‘moisture-enabled telecommunication’ with, there may be some exceptions worthwhile exploring. The technology used in the prototype is a semi-permeable membrane that lets liquids out, but not in, and a motorized sponge that is wettened and pressed against the membrane.

The third prototype, the ‘whispering phone’, telecommunicates airstream. In normal speech, airstream is generally involved, but only felt in close-by conversations, like whispering or sighing. The prototype involves a set o air jets on the phone’s surface, allowing for different styles of airstreams.

These prototypes render a certain future vision of telecommunication tangible – and in doing so, they provoke questions: how much nearness do we want? What kinds of privacy protection will we find ourselves needing in the future? They also demonstrate the value of prototyping – in making a future vision experienceable today they provide concreteness to an otherwisely abstract thought, and thereby allow for discussion.

A new world of interactions

The works described here are only a small section of the current developments in interaction design. What is obvious, though, is that the new omnipresence of computation brings along a whole new world of interactivity. The ways in which we manipulate and experience the digital have undergone radical changes in the last decades, and it will be exciting to see how these ways will change in the future. In the end, it’s probably not humans that should get more technical.

It’s technology that should get more human.

Interaction 12

Fabian Hemmert will be one of the keynote speakers at Interaction 12. It is the fifth annual conference hosted by the Interaction Design Association (IxDA). Each year, IxDA aims to gather the interaction design community to connect, educate, and inspire each other. This year it is held in Dublin, Ireland.

The key to designing social media well lies in designing it for a user’s social interests. Conventional software addresses the user’s task-oriented needs and objectives. But social media succeed when they engage the user’s social interests.

Social interests involve two psychological insights: that users are interested in others generally (social activities, or what’s going on); and users are interested in others particularly (another user).

Each of these is doubled up by the self-reflexivity of social action: users are interested in how they themselves appear to others in general (one’s self image, impressions made, the stuff of “self-presentation” common in social media); and another particular user’s relationship to him or her (e.g. their interest in us).

From this we can quickly see that social media are not a matter of straightforward goal-oriented interaction design. As users, we are aware (if not consciously) of what and how social activities proceed. We become interested in ourselves, in how we are perceived, and in the relation others take up to us.

Thus the interest captivated by social media is twofold: it’s a self-interest and an Other-interest. And the habits that engage users with social media engage users are not just the interaction between a user and the site, but between the user and other users. In the course of using social tools, reciprocity by others, and our mutual recognition of each other, deepens our interests and interactions.

the interest captivated by social media is twofold: it’s a self-interest and an Other-interest

Because social tools use a medium that works by representing our identities and activities, representations themselves become interesting. Klout is an example of meta data used to create social reputation that becomes motivating in and of itself. Many other representations that have become meaningful (for better or worse) include follower numbers on twitter, being listed, circled, commented on; or being retweeted, cited, tagged, and badged.

Activities that would normally pass by unnoticed in the daily course of work and life accrue different meanings when they are captured and represented online. We become extended. These extensions of ourselves (our social media presences) reflect on us. In turn, we become interested in an externalized and represented “version” of ourselves.

This is possible, as a motivation of action and habit, only because we’re able to form and sustain the ideas involved in extended presence. The idea of friendship, the idea of relationships, the idea of popularity, of importance and attention are all motivating and interesting. Social media can seem to make friends count for more than friendship. In some cases this is positive. In others, it is undermining.

To the social interaction designer, this doesn’t matter. All mediated activities that users may take an interest in become motives and those motives become habit — the ingredient, if you will, of successful social tool design and adoption.

The task of social interaction design is to capture and sustain user interest, even if it’s an interest in the abstraction and idea of accumulating friendships, getting noticed, becoming popular, and so on. Doing that requires successfully generating and feeding interests.

the task of social interaction design is to capture and sustain user interest

To the extent that these might produce meaningful and valuable information in the form of commerce, viral communication, social marketing or meta data, human interests are critical factors of social interaction design. A site or system that fails to captivate these basic social interests will wither on the vine.

The user may become interested in any of the following. Note that in each case we are talking about the perceived status of an interest and relation. Social realities are all subjective, interpreted, and can only be validated to the extent that communication provides truthful and sincere verification. Social media require neither to be successful.

Social and interpersonal interests grow from Self to include an Other in person; Others as friends, peers, groups; Others in general (an audience); and online social activities and pastimes.

The user’s interests develop around:

• his or her own self image as represented
• his or her image and presentation as a reflection of acknowledgment by others
• a particular person the interest that person has in oneself
• a scene or social activitysocial position, or who’s who
• an audience or community
• news and social facts, as circulated by known people

These personal and social interests become habits of use. Habits form not around needs and goals, but again, around the deeper motives that structure individual personality and sense of self. Habits are supported and extended by the tools themselves, and are ever evolving with change in the industry and technologies. Social technologies are simply the functional application of individual and social techniques, applied to identity, relating, interacting, and communicating.

User’s activities can include:

• collecting socially relevant items (including friends)
• accumulating socially relevant distinction
• sself promotion, brand promotion, site promotion, profile promotion (social capital)
• appealing to others through requests, posts (bog, video, audio), and comments, etc
• participating in collaboration (wiki, lists, tagging)
• extending daily activities such as shopping, bookmarking, keeping in touch
• avoiding risks, embarrassment, social faux pas and failures (real or imagined)
• work and work-related successes (admittedly more or less interesting)
• social games, including socialized games, and gamified social
• small habits, from instagramming to music sharing
• influence monitoring, projection of persona and reputation

To conclude, then, social tools can never be grasped from a technical or functional perspective alone. Granted, they are designed, architected, built and extended by means of current industry technologies and standards. But their use, and use is the central orientation of any user experience or interaction designer, is explained not on the basis of what tools do, but why and how they are used. The uses of social tools are not utilitarian — comprising of tasks, needs, or goals. Rather, they are intrinsically psychological and social. And as such, comprise of the relational interests people take in their own self and relations to others as represented and communicated online.

In a world of unlimited access to information and infinite choices, it can be hard to make decisions. We’ve created tools to help us sort the humongous mass of information we’ve created; tools to help us find what we’re looking for. We’ve grown so used to using these tools that we rarely notice their downsides and limitations. This article aims to highlight some problems related to findability and discoverability and encourage you to find alternative solutions to the existing paradigms.

What is serendipity and why is it important?

Serendipity denotes the property of making fortunate discoveries while looking for something unrelated, or the occurrence of such a discovery during such a search.

The experience of browsing items in a physical space or online catalogue can differ substantially. For example when you’re browsing records in a store you often come across items you weren’t actively looking for but which you instantly recognize as desirable. Online stores offer some mechanisms for discovery but they’re highly limited in scope when compared to physical environments. They may offer a much higher number of items on sale than a physical store, but because screen space is scarce catalogues have to resort to categorisation trees, so users are only exposed to a small subset of the full range of possibilities. Online, users have less peripheral vision and a limited awareness.

“First we shape our tools, thereafter they shape us” – Marshall McLuhan

Existing Paradigms

So far we’ve had three paradigms for enabling findability: search, categorisation and social recommendation.

Search – the mirror

Search is great when users have a precise idea of what they’re looking for, but lacks the ability to propose related content that could bring new insights into the subject the user is researching. It acts as a mirror, you only get what you put into it. The problem is compounded by predictive text, which actively influences a user’s decision process while she’s typing a search query. The problem is the algorithms behind the search and autocompletion mechanisms are based on statistical data culled from the collective behavior of all users. This leads to the progressive erosion of the less used words from the suggestions mechanism and can have a huge impact on a user’s perception of a given subject. This can have terrible implications, for example consider how someone’s perception of a political event can be shaped or influenced by the keywords being suggested to them while they’re conducting searches. Would you be happy to have your worldview influenced by Google’s auto-complete?

Add to the equation that Google is a private company whose profits depend on advertising and sponsored results, it is a no-brainer to imagine that predictive search will soon start featuring sponsored suggestions too.

What is popular becomes exponentially popular, and everything else gets buried with no chance of surfacing. The price we pay for Google’s efficiency is having our perception of the rich complexity and variety of the world filtered and reduced to its average sum.

Category drill-down – a box inside a box inside a box

As a user traverses a navigation tree, her choices exclude a high number of paths that could provide unexpected insights. Of course users can trawl up and down the tree, and cross-referencing mitigates this limitation, but this becomes unpractical for very large collections. Cross-referencing is often limited in scope, what usually happens is you have a group of objects sharing some common theme pointing to each other but this forms a closed loop, so if you follow the cross-reference links you end up going full circle and sooner or later you find yourself looking at the initial item again.

Social recommendation – The echo chamber

Recommendation engines are the prevalent method for adding serendipity to navigation systems, however they operate in a rather linear fashion: they analyze a user’s past behavior and that of his peers, and use that information to extrapolate recommendations. The intended purpose of recommendation engines is to suggest to users objects or information which might interest them. But because the recommendations are based on a user’s past behavior, it tends to only offer back to the user items within their existing range of interests. So it reinforces a user’s tendency towards a certain behavior and never triggers alternative responses. Instead of offering something truly new to a user, recommendation creates a self-referential loop where a user’s body of knowledge is limited by her existing frame of references.

The notion of “Echo chamber” describes this process: “ [...] participants may find their own opinions constantly echoed back to them, and in doing so reinforce a certain sense of truth that resonates with individual belief systems.  This can create some significant challenges to critical discourse within an online medium.

the echo chamber

The echo-chamber effect may also impact a lack of recognition to large demographic changes in language and culture on the Internet if individuals only create, experience and navigate those online spaces that reinforce their “preferred” world view.  Another emerging term used to describe this “echoing” and homogenizing effect on the Internet within social communities is “cultural tribalism” [...] “ – Wikipedia (On a related note but from a journalism point of view Eli Pariser talks about what he calls the “filter bubble”, worth checking out.)

Mob Rulz

Because people spend so much time on social network platforms, it is safe to assume this will have an impact on culture. It impacts how we relate to those who don’t share our worldview, creating invisible walls between people rather than connecting them. The reverse side of connecting with some people on social networks is we are simultaneously isolating ourselves from all the other people we are not connected with. Users are implicitly creating two groups, those who belong to the circle and those who don’t. We create safe bubbles for ourselves where only the echo of our own preferences can be heard, and opposing views never get a chance to be heard. It’s like a child who is given the option to only ever eat what they please, and exclude anything they don’t know or like, as on Willy Wonka and the Chocolate factory. Seducing if you’re 5 years old perhaps, but would you really enjoy only eating snacks and candy for the rest of your life?

Alternatives

So what can we do to increase the serendipity of the systems we design? Being aware of the issues is already a good first step, I suspect what happens is designers specify a recommendation system but never get down to thinking about the details of how the algorithm behind actually works. So it’s up to the developers to decide and this usually defaults to the “top averages” approach. In most cases no-one even notices these funneling effects taking place.

Search – revisited

In the case of search there is no easy answer, it’s supposed to be functional and work as a mirror. What can help in some contexts is to offer alternative search terms to be combined with the initial query, or showing loosely related results clearly identified as such. Google has changed their results page while this article was being written and now it sometimes displays a “Something different” category as part of the faceted navigation for filtering results. For example if you type “Johnny” it assumes you’re looking for Johnny Depp and under “Something different” it suggests “george clooney, orlando bloom, matt damon, leonardo dicaprio, brad pitt”

It’s going through the most popular results, doing a semantic analysis that identifies “Johnny Depp” as a popular film star and suggesting other popular film stars. This is the El Dorado of semantic search, and it is great to a certain extent, but is not without its problems, for example Johnny Holland or Johnny Hallyday won’t get a chance of being suggested because they don’t fit the semantic class of “hollywood film star”. Funneling towards the most popular expressions is still firmly in place.

Spezify for example does something interesting, it displays related search terms next to the initial search term, so when you click on them the corresponding search is executed. I must admit I am puzzled as to how they determine which terms to display, as sometimes the relation is really unclear. Regardless, the result is you can certainly enjoy a chase down the proverbial rabbit-hole by surfing the suggested search terms. A very successful serendipity tactic, albeit overly random at times.

Categorisation – revisited

A few years ago while working on the information architecture for a VOD catalogue of independent cinema I noticed their films didn’t fit the traditional genres (drama, action, comedy,…) so devised a different system based on assigning multiple tags to each film and proposing a navigation system that would let users switch from tag to film to tag. This enabled a high degree of serendipity without the unwanted randomness. This wasn’t implemented but it’s certainly a viable model.

I built a prototype to test the idea and after some experimentation came up with a second model similar to Spezify’s but with a distinct difference: As you search for a keyword, you receive a set of results, each one with its related set of tags. These tags become the suggested tags. Clicking on a suggested tag adds it to the search query, and filters the results. So it’s like progressive filtering, the suggested keywords get added to the mix as your search query grows in size. This model provides a high degree of serendipity without the unwanted side-effect of making suggestions appear random, there is a logical progression as the user creates a growing query. Here is a screenshot of the prototype:

Social recommendation – What else?

Other possible tactics are…

Following strangers

Instead of presenting recommendations based on a user’s peers, show instead recommendations based on the preferences of non-peers, and identify them as such.

Parsing multiple sources of information

During UX Lisbon 2011 Chris Fahey suggested having an app collecting a user’s information from multiple platforms (eg facebook+twitter+last.fm+amazon+netflix) and using the information from multiple sources to make more intelligent suggestions. I suspect Google is about to pull this sort of trick in the near future, if your Google account becomes your identifier across several products (eg. google+, picasa, mail,youtube, maps), it would be easy to combine it all and make more accurate predictions. The prospect will send a chill down the spine of privacy advocates…

Friend of a friend of a friend

Considering the 6 degrees of separation theory, which has some flaws but is nevertheless interesting for our purposes, a system could make recommendations based on indirect relations, for example, showing what films a friend of a friend likes to watch.

Seeing the world through someone else’s eyes

Twitter has recently introduced a nice little feature which lets you see Twitter the way someone else does, and jump between the people you follow. What if this was extended to the people you’ve never heard of? Google+ is doing that really well, you can merrily jump from profile to profile in a highly serendipitious way.

Polysemy

Because images can have multiple meanings and are more open to interpretation than textual information, they can be used as powerful suggestion mechanisms. If combined or sequenced in certain ways they encourage free association of ideas and discovery.

Get our more

It’s good to let go of digital environments and let real life surprise us. Encourage your users to go out more and place themselves in situations they’re not usually comfortable with, nudge them to venture beyond their comfort zones. Location-based mobile apps are the key here.

There’s a lot to be done in this area and I encourage you to take these issues in consideration when you’re next creating your design solutions, and remember how algorithms can limit user’s choices in a detrimental way.

One of the most interesting aspects of Google+ are the Circles. What could be the idea behind this? What’s the social function? I am trying to find out.

Twitter’s follow/follow back

Of all social tools still going strong today, Twitter’s use of the follow/follow back as a means of launching and gaining traction has been the most copied. I can’t think of a faster way to populate a new social service than to connect new members by means of following/following back. And it’s genuinely useful: users don’t have to think of who to follow — they are shown who they follow already, and asked to confirm or ignore.

The follow works so well because it is gestural. It places no obligation on the user followed to reciprocate, but is rewarding if reciprocation follows. It’s a social solution to a bit of technical awkwardness: how to initiate, invite, solicit, and communicate a connection request without doing so verbally or explicitly.

Google+ introduces Circles

It’s interesting to see, then, Circles in action these past few weeks. Circles are ostensibly a means of organizing friends and colleagues into groups that make a bit more sense of the social graph. Given that the social graph is already in many ways an imperfect and inaccurate representation of one’s social connections. (The social graph is flat. Social relationships are lumpy.)

But Google+ notifies Circle activity. What then might have been kept private becomes social. My act of adding people to circles notifies them of the fact, and the system notification by Google+ to those people in effect becomes a standardized follow notification. This works well for Google+ insofar as it quickly ramps up not just the user base, but also the activity of circling, and the connectedness of members.

Member connectedness is essential to any feed-based system. For connectedness is the filter on feeds. It’s what initiates the subscription to member activity (posts).

Ambiguity

What is perhaps unintended, however, in Google+ Circles notifications is the follower phenomenon, as well as ambiguity about the transparency of Circles. The follower phenomenon suggests to me that Google+ aims to make use of social capital, influence, popularity, and other social effects of a user base differentiated by quantity (number of followers/connections). The ambiguity around Circles utility stems from the invisibility of Circles to anyone but their author: notifications do not state what Circle I have been added to by somebody; nor do members of a Circle know about each other.

Google+ may have opted instead to preserve the personal social utility of friend grouping that seems the most obvious benefit of Circles. In which case, Circle notifications are already introducing the popularity bias that’s intrinsic to a public social follower model.

Google+ may also have intended to make visible shared Circles available, in effect offering groups. In which case, it will be interesting to see how well this works with the openness of the present feed model.

Flat social differences

Social technologies flatten social differences, providing access to people unencumbered by social boundaries and distances. To wit, Zuckerberg is Google+’s most followed user. Circles seems to have been designed to increase utility in a social networking world of easy access and flattened social hierarchy. But the reciprocity and mutuality of following/back that acts as a soft social norm in follower models commodifies relationships in the service of social capital, or popularity. So it will be interesting to see how the team navigates feature and design evolution, now that the floodgates are open on some social practices that to me, at least, seem possibly at cross-purposes.