Owning a share of that free time can translate into a lot of money (which the recent Facebook IPO is testament to).
Social networks aren’t sterile repositories of faces and connections.
They’re more than that.
We provide a lot of personal information, stories and pictures that reveal a lot about who we are and what we think. And like self-disclosure to a friend, we expect a certain amount of trust in return for that information.
It’s no wonder then, that users revolt at what seem to be mercurial changes to privacy policies and practices of disclosing and sharing private information. Twitter, Facebook and Google have each had their share of bad press.
Trust plays an important role on any website, not just social networks. Trust and credibility are especially important when personal information and purchasing occurs. This is why it is included as one of the essential elements of a successful website (along with usability, appearance and loyalty). These aspects are part of the SUPR-Q instrument (Standardized Universal Percentile Rank- Questionnaire).
One of the most valuable aspects of the SUPR-Q (along with its favorable psychometric properties) is that you can compare scores for websites that you manage or care about to other external benchmarks. It’s helpful to know where you fall relative to these titans of traffic and attention.
I asked between 76 and 91 current users of each social network to respond to the 13 SUPR-Q items to get standardized scores across each of the sub-scales of Usability, Trust/Credibility, Loyalty and Appearance. The data were collected between June 2011 and Jan 2012. Here are the results.
Trust: We Don’t
In general, we don’t trust our social networks. The highest scoring site was Google+ but its trust ranking puts it worse than 80% of all other websites. This is surprising considering how well known Google is as a brand. Facebook and Twitter, which have both been around longer and have a larger user base, both have trust ratings in the bottom 10% of all websites. It could be that Google is enjoying a certain innocent until proven “evil” bump.
Figure 1: Users don’t trust social networks. Percentile rank for the Trust subscale. Google+’s 20th percentile ranking is statistically higher than Facebook and Twitter. Error bars are 90% confidence intervals.
Usability: About Average
We might not trust Facebook as much but users find it relatively easier to use than Twitter and Google+. Its usability rank places it higher than 75% of all websites. Twitter and Google+ both have usability ratings about average (and not statistically distinguishable).
Figure 2: Users find Facebook most usable. Percentile rank for the Usability subscale. Facebook’s 75th percentile ranking is statistically higher than Twitter and Google+. Error bars are 90% confidence intervals.
The difference in usability ratings is a little surprising. Twitter is a rather simple website and Facebook has been making all sorts of changes which users grumble about. I suspect part of the reason for the lower Twitter usability rank is the mix of participants in the sample who were just occasional Twitter posters. In some future research I’ll explore how the usability rankings change when we look at power users. In general, users with more experience rate software and websites as more usable.
Loyalty: Google+ is still new
While Google+ is enjoying some early trust, it places third in loyalty at the 45th percentile compared to Facebook and Twitter, which both fall at the 85th percentile. Loyalty is an issue for growth. You want your users coming to your site again and recommending it to friends and Google+ has room for improvement. If you haven’t heard of Google’s Plus network then you’re not alone. This is reflected in the loyalty scores.
Figure 3: Users aren’t yet loyal to Google+. Percentile rank for the Loyalty subscale. Google+’s 45th percentile ranking is about average but statistically lower than Twitter and Facebook. Error bars are 90% confidence intervals.
Net Promoter Scores: Facebook Dominates
One of the loyalty questions in the SUPR-Q battery is the same question used to compute the popular Net Promoter Score (NPS)*: “How likely is it that you’d recommend this website to a friend or colleague.” There is both a noticeable and statistically significant spread between Facebook and Twitter/ Google+.
This is likely a function of the viral nature of Facebook and its huge number of users. Not only would Facebook users likely tell their friend, their friends are already likely on Facebook (even my 89 year old grandmother is on Facebook).
Figure 4: Users most likely to recommend Facebook. Raw Net Promoter Scores. Facebook’s NPS of 38% is statistically higher than Twitter and Google+. Error bars are 90% confidence intervals.
* Our friends at Satmetrix want us to remind you that Net Promoter, NPS, and Net Promoter Score are trademarks of Satmetrix Systems, Inc., Bain & Company, and Fred Reichheld.
Appearance: About average and the same
There isn’t much differentiation in the perception of the appearance of each website. All three are a bit better than average, with Google+’s slight edge not statistically distinguishable from Facebook and Twitter.
Figure 3: Users find the appearances all about average. Percentile rank for the Appearance Subscale. Differences are not statistically significant. Error bars are 90% confidence intervals.
While you’re unlikely to be working on a social network, there are at least three things to take away from this post:
- Trust plays a key role in website success: Features and usability are great for attracting, then retaining an audience, but if users don’t trust (or start to lose trust), then even well-intentioned new features or design changes can be looked upon with suspicion.
- Trust impacts loyalty : Low trust leads to fewer promoters (users recommending your website) and more detractors (users saying negative things.
- Benchmark your website: use questionnaires like the SUPR-Q to help you gauge how well your website scores relative to well-known websites or industries on the essential elements of trust, usability, loyalty and appearance.
Special thanks to John Romadka for contributing data for this article.