Writing for Social Media: Usability of Corporate Content Distributed Through Facebook, Twitter & LinkedIn

by Jakob Nielsen on October 12, 2009

Summary: Users like the simplicity of messages that pass into oblivion over time, but were frequently frustrated by unscannable writing, overly frequent postings, and their inability to locate companies on social networks.

"Mega IA" tackles the problem of distributing your information across multiple outside websites and Internet services. It's no small challenge: it's hard enough to architect your own site, but when additional sites offer wider distribution, it introduces another layer of difficulty in ensuring a good user experience.

User Research

To find out how users approach corporate postings on social networks and RSS, we conducted two rounds of research:

  • Round 1 was conducted 3 years ago and focused on RSS feeds. We tested a variety of feeds with 4 different RSS readers, using two different methods:
    • Most sessions were conducted as traditional usability studies, often using an eyetracker to give us a detailed view of how people read RSS headings and blurbs.
    • We also ran several field studies, observing users in their work environments. This gave us a more naturalistic view of how people use business-oriented RSS feeds in their daily work.
  • Round 2 (the new research) included 4 different social networks — Facebook, Twitter, MySpace, and LinkedIn — and ran additional tests of RSS feeds. This round encompassed two studies:
    • Most sessions were conducted as traditional usability studies in which participants used their own RSS readers (primarily Google Reader) for the study's RSS segment. For both social networks and RSS feeds, we asked users to sign up for messages from a few pre-determined companies and organizations during the 2 weeks prior to their session. We also asked users to sign up for new companies during the test so we could observe their behavior in the moment of "following" somebody new. These lab studies gave us detailed insights into participants' browsing and reading behaviors while accessing corporate messages.
    • We also conducted diary studies in which users recorded and commented on their experiences with corporate messaging over a 4-week period using their existing social networks and RSS feed readers. This approach let us examine longer-term usage patterns.

In total, our research included 73 users, with a roughly equal number of men and women. Most of the participants were in the U.S., but we also studied users in the U.K. and Australia. Participants ranged in age from 20 to 59 and had a wide range of occupations, including bank manager, database administrator, electrician, insurance broker, lawyer, office manager, pharmacist, small business owner, and teacher.

In Round 2, we tested messages from more than 120 companies and organizations, the majority of which were tested by a single user who was receiving the messages outside of our study. To obtain more systematic usability information, we asked multiple users to sign up for messages from the following 42 companies, and then tested their usability:

Companies and Celebrities Government Agencies and Politicians
ABC News
Bruce Springsteen
Continental Airlines
EMI Music Australia
Fairfax Digital
JetBlue Airways
Microsoft Windows
Seacoast Online
STA Travel Australia
The Huffington Post
The New York Times
The Wall Street Journal
The Weather Channel Australia
WBZ NewsRadio
WHDH Boston
Australian Government
Australian Institute of Sport
City of Portsmouth, N.H.
City of Sydney
Department of Health and Ageing (Australia)
Kevin Rudd (Australian Prime Minister)
The White House
United States Consumer Product Safety Commission
United States Department of Education
United States Environmental Protection Agency
Non-Profit Organizations and Charities
American Cancer Society
Amnesty International Australia
NPR (National Public Radio)
Oxfam America
Royal Society for the Prevention of Cruelty to Animals (RSPCA)

Business vs. Personal Use

As the above list shows, we studied only business use of social networks and RSS. We didn't study the overwhelmingly dominant use of these services for purely personal use, where users keep in touch with their friends and family. We did test messages from some politicians and celebrities (such as sports stars and musicians), but even though they're individuals, they function as companies and organizations in the context of posting messages on the Internet.

When Australia's Prime Minister tweets that he's going to Tasmania, for example, it's not to notify his local buddies that he's available that evening for a beer. Rather, it's to promote the fact that he's paying attention to that state's needs. Indeed — as some of our users noted — it's questionable whether it's the PM who is posting or one of his aides.

Our research was guided by a specific goal: we wanted to discern guidelines for companies and organizations. We weren't interested in finding the best way for individuals to post their private updates.

At the same time, business messages appear in a context that's permeated by personal messages. This context sets the stage for use. Businesses that post too often crowd out the user's real friends and become unpopular (and thus risk being unfollowed). Users listed too-frequent postings as their top annoyance with following companies and organizations on social networks.

Users prefer a more casual style for business messages on social networks than what's appropriate for most corporate communications. At the same time, they expect RSS feeds to be more business-like and to cut the chit-chat. Also, for some services — such as the BBC — people preferred a highly professional tone, even on social networks.

RSS updates were viewed as more trustworthy and as more "official" sources than social messages. Users were also more likely to check RSS feeds at work, whereas they mainly accessed social networks from home.

At the time of our research, users accessed only 6% of corporate social network updates from mobile devices (and 94% from "real" computers). The percentage of mobile use might increase as mobile usability improves, but it's likely to remain small because corporate messages are rarely the type of must-have information that people need on the road.

Managing Streams

All the media forms in our study had a single stream (i.e., timeline or "wall") of postings, sorted in reverse chronological order with the most recent on top. Although some RSS users sorted the messages by source, the time-ordered stream was the predominant user experience.

People appreciated this user interface's utter simplicity: no special effort or commands were required beyond looking down the list and maybe scrolling a bit. Users didn't seek out past postings that they might have missed; they were content to read only the newest information.

So, once your message drops off a user's main page, it might as well not exist. Users who continue browsing messages on the second page are almost unheard of.

This contrasts with email newsletters and other email notifications that users have to manually delete. Social network updates float down the timeline and eventually dissipate on their own, requiring nothing of users. Although our participants appreciated this fact, it does make stream-based media less powerful than email newsletters in terms of maintaining a customer relationship.

Posting Frequency

Most users visited Facebook and Twitter at least daily, and MySpace and LinkedIn less frequently. In the future, other services will no doubt become popular, but the basic finding will likely remain the same: some services lend themselves to frequent use and highly timely updates, whereas others live at a more relaxed pace. You should adjust your corporate postings accordingly.

If you post too rarely, your material will drift out of users' active timestreams before they visit again. But, if you post too much, you'll crowd out other messages.

The three great motivators are fear, greed, and exclusivity, and social network postings can address the latter two. Users were particularly interested in getting deals (greed). Yet, while users recognize that corporate postings are commercial — rather than friendship-driven — they do resist overly aggressive selling. Finding the proper balance is crucial.

Users want postings to be current. One user, for example, said the information she received on social networks made her feel like she was "the first to know." Such feelings give followers a sense of exclusivity.

In some cases, companies had established a presence that they didn't bother to update. These graveyard sites gave users a very negative impression when they were looking into companies' social features. Even more irksome were cases in which friend requests weren't promptly answered. Start using a social networking service only if you have the budget to support reasonably frequent postings. And, if you later find out that you don't, close it down gracefully rather than letting it get overgrown by cobwebs.

Finding Companies to Follow

It's rare for users to actively seek out companies and organizations on social networking sites. Typically, the impetus to follow a company came via a prompt of some sort — such as a recommendation from a friend, an email (newsletter or confirmation) from the company, or a link from the company's website.

Unfortunately, once users decided to follow a company, it wasn't always easy to find it. Users often visited a company's own site to find subscription info because current social networking sites offer poor search and navigation. Sadly, even a company's own site sometimes failed to help users find that company's social services. At a minimum, make sure your own search engine coughs up the appropriate pages as a "best bet" when people search for query terms like "Twitter" or "Facebook."

Changes in RSS Feed Use

Because we studied RSS use in two rounds that were 3 years apart, we can track changes in users' approach to this format. The main finding? Not much has changed. All 15 usability guidelines from the first research round were confirmed in Round 2. (However, we did discover several new guidelines, for a total of 24 usability guidelines for RSS feeds.)

One of the main findings from Round 1 was that non-technical users didn't typically understand what the term "RSS" meant. This remains true today, and we still recommend using a phrase like "RSS feeds" to supplement the acronym with one or more explanatory words.

The biggest change from our earlier research is that RSS is now being used more and by a broader audience. Previously, RSS use was fairly experimental and mainstream users weren't sure how to best use the features. Now, people are more accustomed to RSS and select their feeds carefully.

Subjective Satisfaction

In Round 2, we asked people to rate their satisfaction with 292 corporate messages on various attributes using a 1–7 scale (with 7 being the best).

Message usefulness scored the lowest, with an average rating of 4.3 . This is lower than the satisfaction ratings in most usability studies. Clearly, companies have yet to discover how to send customers the postings they really want.

The messages that received the highest scores had three things in common: they contained something of substance, were timely, and provided the kind of information that users expected from the source company or organization.

Although content usefulness is a problem, company trustworthiness scores were generally high, with an average rating of 5.7. The companies with lower trustworthiness scores were mainly those that included advertising in their messages.

Social Messaging and RSS Usability

As the satisfaction ratings indicate, we have a long way to go to improve the usability of social network messaging and RSS feeds.

The problems start with something as simple as the choice of username. For example, the United States Department of Education's Twitter ID was "usedgov," which sounded to users like "used government" and was off-putting. Logos were often bad as well, particularly in the small rendering that some services offer. Users depend on the ability to scan down a stream to pick out logos and user names, but this basic need was often thwarted.

The shorter the message, the more important the writing. Don't simply repurpose the first N characters of a longer piece of content. Too many corporate feeds didn't bother writing for the medium and suffered accordingly, as users didn't know whether to click the links (and therefore didn't).

The good news is that we can only go up. Users do want these messages. In moderation. If they're good.

(For more information, please see the full research report on social media usability, which is available for free download.)

Share this article: Twitter | LinkedIn | Google+ | Email