Judgement Day for Snap Preview

02.1.07

Design UX

In the age of plug and play widgets and turn-key publishing, bloggers are empowered to add new functionality, behavior and information to their websites with a push of a button. One only has to look as far as MySpace to see how individual expression manifests itself in the form of plug and play additions and customizations. Let me assure you: there is nothing wrong with accessorizing one’s piece of digital property.

However, if you’re an individual publisher adding widgets with the intention of improving the user experience, a great deal of restraint is in order. With these kind of goals, you have left the realm of self expression and entered the world of design. Which brings us to Snap Preview Anywhere™ - a tool that allows you to preview a website with a small thumbnail before committing to a click on a link.

Allow Snap’s marketing team to introduce the Preview widget:

Previews give site visitors the ability to ‘look before they leap’ when determining whether or not to click on a link.

Previews help you, the site owner, keep the user on your site instead of losing them to the site behind the hyperlink (increases relevant, on-site page views).

While it supposes a somewhat miserly approach to links, beneath the veneer lies a pretty compelling value proposition that promises to improve the user experience on a website.

Still, bloggers implementing this functionality on websites have garnered mixed reviews from their readers. Nick Wilson felt really passionate in his assessment, in a post titled: 3 Reasons Why Snap Preview is Ruining Your Blog, and Hurting Your Readership. The first gripe he mentions, really resonated with me, as the accidental triggering of Snap Preview forced me to consciously change my typical browsing behavior, and left me constantly aware of the daunting possibility that another Snap Preview might appear.

Snap Preview links leave my mouse scrolling the content gutterWhen I view a website, I leave my mouse stationary and use the scroll wheel. Each link that has Snap Preview makes the vertical area above and below it a hazard for accidental activation. Forcing me to find a gutter in the content to place my mouse cursor.

What problem does it solve?

Overusing any tool without first addressing what problem it tries to solve will leave you with frustrating results. The Preview has the capacity to provide extra information on a destination link that the user wouldn’t be privvy to unless they visited the website, such as:

  • Is the destination website a blog?
  • Does this link point to a wikipedia entry?
  • Is this the same website the author linked to earlier?

The use of a visual thumbnail preview to qualify a link is not exclusive to Snap, and can also be seen in Ask.com’s search results, among other sites. This reaffirms that there is some viability to the approach, as more than one group of UX designers is convinced of its potential benefit.

Ask using the preview pattern

The challenge

While a Snap Preview seems perfectly capable of addressing some unanswered questions before clicking on a link, this problem is not universal to every visitor. The challenge is to provide the extra information for those who need it, when they need it, and abstract it for those who don’t. Additionally, not every link should qualify to have a preview.

A prime example of problems Snap Preview can cause would be a large photo that links to Flickr: when used, the Snap Preview pops up and ends up obscuring the image, only to show a smaller thumbnail of the image that was just obstructed. Here is an analogous example that might be accompanied by a tear in the universe (thanks to yizzle for the inspiration):

A recursive Snap Preview

I wanted to show some compelling examples of the widget in action, but many of the sites that have used it have simply removed it by now. Either way, when I was using sites that had the Snap Preview enabled, I found myself leaving the site and, out of habit, hovering over links on sites that didn’t have the functionality, expecting a thumbnail preview. This might suggest that there is some more value left in the concept than the recent backlash suggests.

I will argue that no content publisher needs such functionality. Still: can this widget be redeemed and turned from a nuisance to a complimentary feature? How does one decide whether Snap Preview is right for their site, or how does a blogger still have fun with it in a way that doesn’t burden their readers? What are some good examples? Feel free to post your personal experiences with the widget in the comments section. I’ll summarize people’s sentiments and offer some solutions in my next post.

Get Your Internet On

« Older Heuristics for Choosing a DOCTYPE in 2006 (First Stab)
Newer » Snap Preview Extreme Make-out

9 Comments

1Erik

Overusing any tool without first addressing what problem it tries to solve will leave you with frustrating results. The Preview has the capacity to provide extra information on a destination link that the user wouldn’t be privvy to unless they visited the website, such as:

Is the destination website a blog?
Does this link point to a wikipedia entry?
Is this the same website the author linked to earlier?

This information is unobtrusively visible to the user in the status bar.

2Ephram Zerb

@Erik. You’re absolutely right, for most users, the status bar will help them answer questions like the wikipedia one more efficiently (it certainly does for me). Knowing you personally, I know you’re an engineer, and a URL communicates much more to you than it would for a typical website visitor.

As for checking if a destination website is a blog: this example was derived from a conversation I had with someone who cited this as the reason they like Snap Preview.

Admittedly, “Is this the same website the author linked to earlier?” is not a clear example. What I was trying to illustrate is that a visitor’s representation of a website in memory definitely includes visual information. If I visit a website for the first time, I will more likely remember what it looks like before I remember what URL it was at.

Regardless, a successful use of Snap Preview would have to be a considerate improvement on what the status bar. Thanks for your input.

3Erik

For some reason the list item tags were stripped out of my previous comment, so please forgive the literal list enumeration below. (You might want to change that behavior or at least warn users about it.)

A hyperlinked thumbnail is suitable for high-resolution photographs because:
1. a photograph is a static, well-defined entity
2. a scaled-down representation of a photograph provides an impression of the whole
3. it offers significant bandwidth savings

Web pages, on the other hand:
1. encapsulate dynamic content, feeds, hyperlinks, etc.
2. lose meaning when text is scaled down to the point of illegibility
3. are designed to load quickly

Unlike a photograph, the content of a web page is not easily summarized in image form.

It is the responsibility of the author to summarize it between the opening and closing of an anchor tag. Google’s PageRank algorithm works as well as it does, in part, because authors tend to do this as well as they do. When an anchor’s reference is ambiguous, a user can hover over it and look in the status bar to see the referent. Beyond that, the best way for a user to evaluate the quality of a link is to click on it.

4Erik Wingren

My name is Erik Wingren and I head up UX Research for Snap — the company behind the Snap Preview Anywhere (SPA) service.

@ Ephram: I am extremely impressed with your analysis of the user experience of our product. I can’t go into details on everything we are doing but I wanted to share some of the near-term updates, aimed at solving some of the usability issues described here and elsewhere:

- This afternoon we released an update to the SPA code that disables the preview functionality while the mouse-wheel is in use. Since users hardly ever use the mouse to scroll and point at the same time this enhancement *should* eliminate a large number of the “accidental triggering” instances. Please let me know if this in fact solves the problem for you.

- We are exploring different solutions to the linked image problem you point out. This one is a little tricky — some users see this as a feature while others, like yourself, see it as broken functionality.

- We are currently in the final stages of testing a new control that will let the site owners manage readers’ expectations by visually denoting which links are SPA-enabled and which ones are not, and thereby in effect eliminate the “surprise element” of rolling over links on sites that uses SPA.

@ Erik: I believe you represent the very definition of user that we are currently failing to serve.

I believe that unless we design the technology to meet the needs/wants of at least 80 percent of the user base we are not doing our job right. As of right now you are in the “other” percentile… and I would love to understand what it would take to get you over on our side ;)

If you have the time and energy, I would be very interested in what type of at-a-glance-information would help you navigate with greater speed and accuracy?

Again, to the both of you: Thank you for your articulate and insightful feedback. Your viewpoints are informing the ongoing development of this product. And I sincerely hope you will continue to share your thoughts as this product evolves.

P.S. 2 x Erik with a “K” on the same thread is a first for me! D.S.

Best Regards,

Erik Wingren
Snap UX Research
erik[@]snap[.]com

5Jonathan Nicol

I admit that prior to reading this post I haven’t given much thought to Snap Preview Anywhere, but when I had encountered SPA links in the past, my user experience was a negative one. Here is an attempt to qualify my negative reaction:

1. SPA previews look like a popup, which is intrinsically associated with advertising. The presence of a company logo on the ‘popup’ did little to disillusion me of this notion.

2. I fail to see how a thumbnail representation of a website will help me make a decision whether to click a link. It might allow me to see the masthead of the site that the link points at (a site I may, or may not recognize by its logo), but it doesn’t allow me to evaluate the content potential of the site. In many respects, the anchor text can perform this task more adequately. A link such as “… recent [statistics show] that …” informs me perfectly adequately that I’m about to see some statistics proving the author’s point. I’m not sure how a Snap preview adds value to the proposition. As Erik #1 has already mentioned, the humble status bar will allow me to ascertain the URL of the target website. Lets not forget the ‘title’ attribute either, which allows the author to specify an unobtrusive hover that explains the link in more depth. For example: title=”Read survey results on the National Statics Bureau website” does a great job of telling me what will happen when I click a link, and where it will take me.

3. It is distracting to wait a second for a preview image to load, look at and absorb the image information. In other words, it slows me down.

4. Previews can be misleading. On the SPA website, there is an SPA enabled link to the SPA blog. In the preview I can make out the headline “Snap Image Search Adds Filtering Options”. Of course, when I navigate to the SPA blog that is not the current headline.

6Haarball

This is interesting, but as many of you have summarised, it is at times useless, maybe even obtrusive.

I see it as more of a niche product. I can only speak for myself, but I often browse around the web in search of design inspiration. I look for certain aspects of sites I like, certain types of layouts, etc. and in this respect I can see how Snap would be useful in speeding up my browsing. It can be useful in regular web browsing as well, but not more often than it would be obtrusive and slow the user down.

I’d say it’s a feature that would have to be carefully considered *with each link* in relation to the nature of the given link and whether or not Snap would improve or worsen the usability — a huge obstacle towards making it a successful feature.

Due to the inherent nature of web browsing, given the existence of the status bar and the fact that a visual overview of a site won’t necessarily give the user any relevant information, I can hardly see this feature becoming widespread and widely successful among regular web users. UX might be better off focusing on people whose web experience is more likely to be enhanced by a visual overview - Web Designers, for example - and thus alter their goals as far as a user base goes. I don’t know what they regard as their user base — nevertheless meeting the needs of 80 percent seems over-ambitious at best.

Best of luck with the project, but I fear this is one feature that has lost before it’s even started due to the very nature of browsing the web — that is if the regular web user is considered as part of the demographic.

7Ephram Zerb » Snap Preview Extreme Make-out

[...] About a month ago, Snap Preview was dodging a backlash to its suddenly-pervasive preview tool. Along with dozens of bloggers, I offered my assessment and promised to post a follow up with some solutions and approaches to using Snap Preview effectively. [...]

8subcorpus

yes its good when it works fine …
but on slow connection and pages …
it actually becomes a disruption to the browsing experience …
it hides what i am trying to read …
may be it’ll develop into something less intrusive …
its kewl …

9alisty

snap has improved the preview service a lot by letting site owners to decide which link (blocks) they are willing to enable… i think the preview idea is not bad, if they can make it more informative and customizable.

Leave a Reply

required

required (will not be published)


required