Which Site Seals Create The Most Trust? [Original Research]

Which Site Seals Create The Most Trust? [Original Research]

When internet users share private information, they want to feel safe doing so.

One of the most popular ways to convey security on a website is by using trust badges (also referred to as “trust logos” or “site seals”).

This study, conducted by ConversionXL Institute, expands on existing research from Baymard Institute’s research in 2013 to better understand the popularity and efficacy of various trust badges online.

Results summary

  • Familiar brands like Visa, Mastercard, PayPal, Norton and Google were also the most trusted seals when paying online.
  • Generational Differences:
    • ‘Google Trusted Stores’ was trusted more by millennials (gen y) compared to older people
    • Sitelock did well with middle age (gen x) compared to millennials
    • Surprisingly Paypal was trusted by older (50+) participants relative to millennials.
  • Gender Differences
    • Males were significantly more likely than females to trust the Visa-Mastercard seal
    • Females were significantly more likely to trust the Better Business Bureau (BBB) seal
    • Females were significantly more likely to trust the Norton security seal
  • Most respondents purchased something online at least once per week (98%) with over two-fifths (42%) feeling concerned about security at least half the time.
  • Unaided awareness was very low for most trust seals, although once aided with the image of the trust seal badge, awareness increased significantly.
  • When shopping online a PayPal Trust Seal added significantly more security than any other brand.
    • BUT, those frequently concerned about online security were significantly more likely than those who were not to feel a SiteLock (20% vs. 14%) trust seal made them feel more secure when shopping online.

How do I apply this research?

If you can test, then do so with the top 5 or so results. ‘Paypal-verified’ provides the most trust, followed by Norton, Google Trusted Store, Visa-Mastercard, then BBB.

If you can’t test, then implement trust seals with the help of this study. However, as demonstrated here, a quick survey is not too difficult and doing one analogous to this with your customers could help you narrow down your choice of trust seal. Though, if in doubt, use Paypal.

If you’re primarily concerned with convincing ‘insecure’ shoppers, test a SiteLock seal.

Background

“Trust Logos” or “Site Seals” could really mean a lot of different things.

They include privacy seals, brand association badges, or SSLs (secure sockets layers) which actually create encryption between the user’s browser and the receiver’s server. Here we generally refer to them as Trust Seals, as they are all logos intended to convey trust onto the webpage, even if they differ in how they do so.

This research identifies which logos exude perceived security.  This is an important distinction to make since users don’t necessarily care about what a trust badge actually signifies (whether the seal is anti-viral, SSL, payment protection, and so forth). Instead, they simply want to see a trust seal that gives them the feeling that their information is in safe hands.

Again, our research here is a spinoff (not a reproduction) of Baymard’s trust study from 2013. In case you’re unfamiliar with Baymard’s research on trust seals, here’s a brief breakdown of the results:

  • When asked “Which badge gives you the most sense of trust when paying online?”, almost half of all participants (49%), didn’t know which badge was most trustworthy or simply had no preference.
  • 35.6% of participants voted for Norton’s antivirus seal.
  • McAfee’s SECURE seal came in second receiving 22.9% of participant preference.
  • Remaining seals tested: TRUSTe, BBB, Thawte, Trustwave, GeoTrust, and Comodo.
  • Participants showed no preference between trust seals and SSL seals but did show a preference for antivirus companies (possibly due to their familiarity with them).

This last point, the possibility that users’ preference for trust badges is determined by their familiarity with them, is the one we considered when expanding on Baymard’s study. (Disclaimer – we are not explicitly duplicating the Barymard study, rather we have different questions about trust seal perceptions).

In this research study, we hypothesize that participants will prefer the trust seals they are most familiar with.

Additionally, we took data on age and gender so we could look for any patterns of familiarity or perceived security across those variables.

Study Report

This study will be split into two parts because the methodologies were slightly different. However, the conclusions are all related.

Data Collection Methods and Operations (pt. 1):

First, here are the trust seals, which vary how or what they claim to provide….business or site trust (i.e., BBB, McAfee, TrustedShops), security (i.e., SiteLock, GeoTrust, thawte, Comodo), or simply brand associative trust (i.e., paypal, visa, google, shopify, etc).

We added the Visa-Mastercard logo not as a pure trust seal example (though it has been historically), but as a sort of benchmark test to explore the idea that familiarity doesn’t always provide security…it is our hypothesis that perceived trust won’t match the familiarity of it.

Note, we also had a “Do not know or no preference” choice in the surveys.

trust_seals

We surveyed a total of 2,100 people, all from the United States. Our study setup differed from the Baymard study in that we used SurveyGizmo combined with Amazon’s Mturk (instead of Google Consumer Surveys) so that we could randomize the order of trust seals, and show more at once.

We did two surveys, both asked respondent age (in 10yr intervals) and gender. One asked “Which badge gives you the most sense of trust when paying online?” (1,054 respondents) & the other asked “When shopping online, which badge are you most familiar with?” (1,046 respondents).

Both surveys combined had approximately 51% male and 49% female.

Here is the breakdown for age:

Percent of respondents according to age groups
Percent of respondents according to age groups

Findings (pt 1)

Here is the distribution of results for both questions:

Distribution of results from both trust seal surveys

Takeaway – Familiar brands like Visa-Mastercard, PayPal, Norton and Google were also the most trusted when paying online.

Here’s another way to look at it, and more directly see the relationship between familiarity and perceived security:

Familiarity versus perceived security of common trust seals
Familiarity versus perceived security of common trust seals

The Most Familiar vs. Most Secure chart plots familiarity (x-axis) against security (y-axis). The red dashed line shows where respondent perception of familiarity and security would meet if they were equal.

The hypothesis that familiarity accurately predicts a populations’ perception of security is generally supported. For the most part, greater familiarity also meant greater feeling of security. The exception, as hypothesized, is Visa/Mastercard, which was the most chosen badge by the familiarity group.

The familiarity group and the security group chose the trust seals in roughly the same frequency and most badges stick close to the dashed red line. PayPal was selected as both more familiar and more secure than Norton, which was similarly chosen more than Google Trusted Store, BBB, McAfee and so on.

This general pattern held for both males and females, but there were some interesting discoveries in the demographic data:

Takeaways – ‘Google Trusted Stores’ was trusted more by millennials (Gen Y) compared to older people, Sitelock did well with middle age (Gen X) compared to millennials, and surprisingly Paypal was trusted by older (50+) participants relative to millennials.

One in four respondents (25%) say PayPal gave them the best sense of security. Across all of the age brackets, PayPal was the most chosen, or tied for most chosen.

For the Google Trusted Store, 20 to 30 year olds are statistically more likely  (19%) than the total (15%) to say that the Google badge gives them the best sense of security. An interesting trend for Google and security is older respondents are progressively less likely to say Google gives them the best sense of security. Among those over the age of 50, only 3% say it gave them the best sense of security, a statistically lower percent than the total (15%) of those who selected Google.

Data Collection Methods and Operations (pt. 2):

In the second part of this study, we take a step back and in addition to asking about participant age and gender, we ask additional questions to get a better feel for how people feel about trust seals in relation to how they feel about online security in general. We ask 1,037 people, all from the USA:

Demographics

  • Age?
  • Gender?

Online shopping behavior and security concerns

  • How often do you make purchases online?
  • How often do you have security concerns shopping online?
  • What are types of things you buy online?
  • Types of things you avoid buying online due to security concerns?
  • List your security concerns for shopping online

Trust Seals

  • Write in any website elements or any specific ‘trust seals’ you look for when wanting security while making online purchases (Unaided question – so before we show them any trust seals)
  • Then we showed them a long set of trust seal images and ask them to sort the images into 5 categories.
    1. Never seen of it or heard of it
    2. Seen it, but doesn’t add to my sense of security paying online
    3. Makes me feel a bit more secure paying online
    4. Adds a lot to my sense of security paying online
    5. Makes me feel 100% secure paying online

Demographics

Proportion of male and female survey participants
Proportion of male and female survey participants
distribution of respondents by age
Distribution of respondents by age

Online shopping behavior and security concerns

Distribution of survey responses for how often people make purchases online
Distribution of survey responses for how often people make purchases online
Distribution of survey responses for how often people have security concerns online
Distribution of survey responses for how often people have security concerns online
  • What are types of things you buy online?
    • Responses included – Books, clothes, electronics, household goods (‘toiletries’), pet supplies.
  • Types of things you avoid buying online due to security concerns?
    • Many people responded with ‘nothing’ or ‘can’t think of anything’. Other responses included – things from ‘unfamiliar websites’, things ‘out of the country’ or ‘overseas’, ‘big ticket’ or ‘expensive’ items,  jewelry, cars.
  • List your security concerns for shopping online
    • the vast majority of responses here were related to credit card information theft and fraud. Secondarily, responses mentioned identity theft. So the terms ‘personal information’ or ‘credit card’ was used for just about every response. Otherwise, ‘unknown websites’ was a common concern, and less commonly reported was viruses, spyware, phishing, hacked, and bank accounts.

Trust Seals

First, here are the trust seals, which vary how or what they claim to provide….business or site trust (i.e., BBB, McAfee, TrustedShops), security (i.e., SiteLock, GeoTrust, Thawte, Comodo), or simply brand associative trust (i.e., PayPal, Visa, Google, Shopify, etc). Note, we added a couple new trust seals to this study that were pointed out as missing from our Part 1 study, these are the Verisign and TrustE seals.

Common Trust Seals Used Online
Trust seals used for this study

Here’s what we found when asking the unaided question: “Write in any website elements or any specific ‘trust seals’ you look for when wanting security while making online purchases”

chart4

The unaided awareness of trust seals supported the Trust Seal Part 1 study, with Paypal (17%), Norton (6%), BBB (7%), & Google Trust Store (4%) each being mentioned at least 4% of the time.

Generally, unaided awareness was very low for most trust seals, although once aided with the image of the trust seal badge awareness increases significantly.

chart5

When shopping online a PayPal Trust Seal added significantly more security than any other brand

Demographic differences

femaletrustfigFemales were significantly more likely than males to feel a Norton (53% vs. 44%), BBB (43% vs. 35%) or Shopify (18% vs. 13%) trust seal made them feel more secure when shopping online

genytrustGen Y (aged under 30) were significantly more likely than Baby Boomers (aged 50+) to feel a Google (54% vs. 31%), Norton (54% vs. 45%), TRUSTe (31% vs. 20%) or Trusted Shops (9% vs. 3%) trust seal made them feel more secure when shopping online.

sitelockThose frequently concerned about online security were significantly more likely than those who were not to feel a SiteLock (20% vs. 14%) trust seal made them feel more secure when shopping online.

Females were significantly more likely than males to have security concerns when shopping online at least half the time (46% vs. 37%)

Limitations

  • How people say they’d behave and how they would actually behave are not synonymous–just because some seals are more familiar does not mean that they will increase conversions
  • In other words – brand recognition does not lead to user behavior
  • Does not shed light on combinations of seals
  • The most trusted seals might also be the most visually simplistic/clear, and are preferred for this reason.

There are some additional ways we could have improved the study (we might follow this up if it generates enough interest), and this would be to expand the questions to get a better sense of who is answering the questions and why they respond they way they do. This would include additional questions like:

  • Unaided awareness (write in all brands you are aware of)
  • Additional aided awareness (select all you have heard of or seen before)
  • Contingent trust (of those aware of how much do you trust each one when transacting online)
  • Concerns when transacting online
  • Frequency of online transactions
  • Types of things they buy online
  • Types of things they avoid buying online due to security concerns

Conclusion

Generally, the more familiar a trust symbol is, the higher it is rated in terms of perceived security as well. There were also gender and age differences in perceived security, so make sure to optimize for your specific audience.

cxli-logo

Join the Conversation Add Your Comment

  1. Great stuff? Do you have these results split across demographics like age and gender?

    Reply
    1. Ben Labay

      Hey Joe, yup, that’s in the study. Check out the result summary up top for generational and gender differences. Cheers, Ben

  2. You have PayPal verified in this list but this is no longer a viable seal supported by PayPal. It has since been replaced ( https://www.paypal.com/webapps/mpp/logo-center ) I’d be interested to know how one of their new seals would rate such as “Secured by PayPal” versus the older verified seal.

    Reply
    1. Ben Labay

      hey Rob, thanks for the heads up, or initial study setup (a while ago now) just focused on a commonly used version. Would be interesting to compare versions, we’ll keep that in mind. Thanks for the comment, cheers!

  3. This is interesting stuff!

    I wonder how this transfers over when it’s applied to A/B testing. I’ve personally done extensive testing on the McAfee Secured trust mark.

    I was actually their partner for some time, when a client wanted proof of the ROI of having that badge on their site. That McAfee trust mark in particular produced a lift between 7% and 20% relative improvement to primary goals, when I worked with it.

    You guys should do a follow up where you A/B test these badges!

    Reply
    1. Ben Labay

      Hey Danny, want to share your experience with McAfee as a case study?! Let me know and we can set something up. Cheers, Ben

  4. We absplit PayPal on a purchase screen and it resulted in fewer conversions. We wondered if PayPal has associations with amateur sites and this could lead to mistrust. We did remove some information about availibility to make room for it though, so who knows?

    Reply
    1. Ben Labay

      Hey Alex, thanks for sharing. Cheers, Ben

  5. This is awesome to see how people trust on website by a mere logo of it. Still as peoples mindset varies and maybe depending upon the websites people visit the logo would have different impact. For e commerce pay-pal should work and for movie and software downloads security software’s…

    Reply
  6. Amazing study! This has really helped out with building my website. Appreciate the good R&D behind this article.

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *