Trust in the Little Things

By Steve Baty

Published: November 19, 2007

“We continue to see organizations implementing devices such as abbreviated link URLs that jeopardize trust relationships for little or no benefit.”

Despite nearly a decade of research into the fragile nature of trust relationships in online user experiences, we continue to see organizations implementing devices such as abbreviated link URLs that jeopardize trust relationships for little or no benefit.

What is a trust relationship? One widely accepted definition of trust comes from the classic book Foundations of Social Theory, by James S. Coleman. Coleman offers a four-part definition of trust:

  • “Placement of trust allows actions that otherwise are not possible—that is, trust allows actions to be conducted based on incomplete information on the case in hand.
  • “[If] the person in whom trust is placed (trustee) is trustworthy, then the trustor will be better off than if he or she had not trusted. Conversely, if the trustee is not trustworthy, then the trustor will be worse off than if he or she had not trusted.
  • “Trust is an action that involves the voluntary placement of resources—physical, financial, intellectual, or temporal—at the disposal of the trustee with no real commitment from the trustee.
  • “A time lag exists between the extension of trust and the result of the trusting behavior.”

The second point in Coleman’s definition has the corollary that the trustor’s past experiences with the trustee will influence the trustor’s future trust decisions with the trustee. This essentially constitutes a Bayesian decision tree in which each past experience influences the trustor’s expectation of the likelihood of success or failure resulting from trust. When the trustor does this mental equation as he assesses whether he should place his trust in the trustee, negative experiences more severely affect his decision. Put another way, once a trustee breaks trust, it is difficult to regain it.

The Essence of Online Trust Relationships

“Trust lets a visitor feel comfortable interacting with our Web site.”

In what ways are online trust relationships distinctive? Unlike physical relationships, in which we can rely on cues such as body language and speech patterns to help us assess the credibility of another person, in online environments we lack many of these helpful indicators. We are also unable to observe and verify obvious characteristics such as age and gender, instead being reliant on the statements of the other person. We cannot readily determine whether other people are misrepresenting themselves.

Because we have fewer facts on which we can rely in judging the level of trust we should place in another person, our approach to the formation of an online trust relationship is more progressive than it might be in offline interpersonal relationships.

Why is it important to establish, build, and maintain trust relationships online? Trust lets a visitor feel comfortable interacting with our Web site. On the Web, which is global in scale and lets organizations operate 24 hours a day, the availability of alternate venues where visitors could spend their time and money and give their loyalty is vast and requires minimal effort to reach. Trust enables online interactions—and the lack of it seriously undermines online ventures.

Trust Relationships and Social Networks

“Our connection and experiences with one person who links us to a social network confer some measure of trust on the other people who are connected to that person….”

Trust relationships develop in social networks as a direct consequence of building a network. Our connection and experiences with one person who links us to a social network confer some measure of trust on the other people who are connected to that person—initially, to the same extent trust exists in the primary relationship. In other words, mutual connections share a modicum of trust despite never having interacted directly with one another, because of that shared connection. This is one of the most powerful attributes of social networks.

Once we make our initial connection with people online, subsequent interactions with those people either increase or decrease their credibility and, consequently, our trust in them, based on the quality and level of satisfaction we derive from those interactions. For example, I am much more likely to view photos a shared connection on Flickr has uploaded than I am to view photos I find through some random search.

Social networks can also directly confer greater trust through recommendations from  trusted intermediaries. If a direct contact, or friend, passes along information from another of his contacts, we treat his recommendation with less suspicion than we might if we received the same information directly from someone we didn’t know.

Link URLs, Abbreviated URLs, and Trust

“When we embed a link in our content, our aim is to encourage visitors to explore beyond the bounds of the information on the current page by following the link we’ve provided.”

The Web is built on the notion of hyperlinks between pages of information that are spread across an otherwise diffuse network of server nodes. Domain names provide approachable handles for their underlying network protocol addresses—IP addresses—making them easier to enter, remember, and communicate. So, instead of requiring a URL such as http://192.168.1.23, a person can instead use http://mycompany.com.

However, beyond a site’s domain name, which represents its home page, URLs for a site’s internal pages have a tendency to become entirely too long and convoluted to remember. For example, I’d have difficulty communicating the following URL to another person in any form other than writing:

http://www.mycompany.com/Merchant2/merchant.mvc?Screen=
PROD&Product_Code=HA-PB&Category_Code=PS&Affiliate=conklink

In fact, I’d likely need to copy and paste the URL into some kind of document to ensure I didn’t make a mistake in typing it. We rely on bookmarks we’ve created, hyperlinks on sites, and search engines to locate information below the top level in a site’s structural hierarchy. And this very long URL tells potential visitors very little about the information they’re likely to see on the destination page.

Recently, we’ve witnessed the use of so-called friendly URLs in place of these computer-generated, code-centric addresses. The following example is still quite a mouthful, but starts to give me a hint about the nature of the information available on a page and, in so doing, inspires greater confidence that the information is worthwhile and relevant:

http://www.smh.com.au/news/sport/jaques-150-sees-australia-
in-command/2007/11/16/1194766916961.html

Typing this URL also requires less effort on the part of the visitor.

Long URLs are very unwieldy and are especially unsuited to platforms such as mobile devices, which let people send messages via SMS—where character limits are in effect—and use social networking services such as Twitter. In response to this problem, a number of services have been developed that translate a real URL—such as those I’ve shown previously—to an abbreviated URL people can use in its place. One such service, TinyURL, would convert the example URL I gave for The Sydney Morning Herald into something like http://tinyurl.com/7tjy—a much more convenient form of the URL, but not without issues of its own.

When we embed a link in our content, our aim is to encourage visitors to explore beyond the bounds of the information on the current page by following the link we’ve provided. By doing so, we provide both a richer experience for visitors and greater value to them through their relationship with us.

However, abbreviated URLs do have some disadvantages:

  • They obfuscate the real destination.
  • They eliminate some of the contextual information that gives us confidence we’ll benefit by following the link.
  • They require an even greater level of trust—blind trust.
  • They introduce a third party into the technical architecture—and potentially, an additional point of failure.

Social Networking and Tiny URLs

“In Twitter’s social networking context, … we are probably willing to overlook the lack of a direct trust relationship, instead accepting our direct contact’s trust assessment of the destination page.”

In Twitter’s social networking context, shown in Figure 1, we might receive an abbreviated URL that a trusted source or friend has passed along as a recommendation. In this circumstance, we are probably willing to overlook the lack of a direct trust relationship, instead accepting our direct contact’s trust assessment of the destination page.

Figure 1—Twitter

Twitter

However, in other environments such as public message boards like Yahoo! Message Boards, which is shown in Figure 2, the use of abbreviated URLs would immediately elevate our inherent level of suspicion. We lack any personal history with a poster providing a link through which we could assess the URL’s associated risk.

Figure 2—Yahoo! Message Boards

Yahoo! Message Boards

Blogs operate more like trusted sources—like the blog Schneier on Security shown in Figure 3—although we might not have had prior exposure to a particular blogger. Our repeated visits to a blog let us assess an individual’s credibility, on which we can then base a decision on whether to follow an obfuscated URL.

Figure 3—A link in a comment on Schneier on Security

Link in a comment

For commercial sites, there is a more complicated trust relationship at issue. On one hand is the visitor’s assessment of our credibility as a purveyor of information. More prominent news sources such as the Daily Telegraph in the UK, shown in Figure 4, engender a greater degree of initial trust through brand awareness and their resulting credibility.

Figure 4Daily Telegraph

Daily Telegraph

However, site operators must remember the fragile nature of online trust relationships and show respect for the trust and faith each visitor places in them. The use of abbreviated URLs—and the lack of transparency they entail—does not demonstrate such respect. We risk diminishing visitors’ trust in our site and our brand if we use abbreviated URLs.

Some guidelines for the use of abbreviated URLs include the following:

  • Always provide descriptions of such links.
  • Where possible, use meaningful links.
  • Don’t assume visitors can understand or recognize abbreviated URLs. Add some explanatory information about such links.

Most importantly, remember that your brand image is in the hands of your visitors and customers. Take care to engender trust through all the interactions visitors have with your brand—both large and small. Respect their trust, and they will reward you by helping you build a positive brand image. Abuse their trust—even in small ways—and you run the risk of damaging a customer relationship to the point where a formerly loyal customer becomes a vocal opponent. Trust matters!

4 Comments

Steve, I like how you clarify the concept of trust and how it relates to the online medium. People build trust from different factors, and especially on the Web, they build trust from user interfaces that work, expecting that all subsequent sites they visit will work the same way.

I think we should clarify the reason when it is necessary to display the entire URL on a Web page. After all, we have the tag to display link text that masks a specific URL. Following the Daily Telegraph example, they should have displayed text as a link, in relation to the context of the link—Game Web.2.0 Entrepreneur—instead of displaying a TinyURL.

A TinyURL is beneficial when you cannot use HTML elements—for example, in email lists, text-only emails, and comments sections—that is, in blogs—where you cannot embed HTML.

Users’ expectation is that, if they see a link, the link will take them to the URL embedded in it, which will be in relation to the context of the link. Why change that expectation by displaying the entire URL on a Web page?

Juan is exactly right: URL abbreviation is appropriate when linked text can’t be created or relied on. Linked text to an abbreviated URL seems so ridiculous. By obfuscating the actual URL, you also break the browser’s status bar and “Copy link location” right-click menu command—maybe not their outright functionality, but their expected usability.

Relying on URL rewriting services is a fast way to break the Web. However, for links within emails and so on, it’s pretty handy. As for trust on the Web, you make some good points.

I am surprised that a browser extension doesn’t exist that handles tinyurl links, so you could see the final destination by simply hovering over the link.

Even better would be an accepted method of encoding or compressing URLs that is built into the browser, so huge query strings and/or path information could be shortened and expanded by the browser, but still maintain the original destination domain.

Thank you all for the comments.

Juan: Your example for the Daily Telegraph is pretty much spot on. That style of presentation would give people a much stronger sense that a link was taking them somewhere safe.

Chris: I’ve since come across Decent URL—a similar service from a New Zealand-based developer. It has some nice features that really help resolve some of the issues outlined above.

Steve

Join the Discussion

Asterisks (*) indicate required information.