Let's get connected:
Last week I set out to discover how trust logo recognition effects online purchase behavior. It is widely accepted that trust logos can have a positive effect on conversion rates when purchasing online, but is that enough? What about which trust logo? As someone who works in the online industry I consider myself relatively knowledgable of the different supplies of SSL certificates, but what about Joe Consumer?
Disclaimer: This is not a scientific study, rather an ad-hoc attempt to gaining some actual insights.
Using several great tools to set up a guerrilla research study I took it upon myself to get some sense of the recognition of trust logos among consumers and its effect on online purchasing. See it more as a conversation starter for you and your colleagues or you and management.
First off, I started with a visual test using Usabilla which I followed up with a short Wufoo survey. In the Usabilla test, I wanted participants to answer 4 questions:
As far as recognition goes, many brands seem to do well. McAfee (79%) leads the way with Verisign (76%) coming in a close second and Paypal (72%) third. After that, the recognition levels drop significantly with BBB (37%) and TRUSTe (28%) picking up the trail in fourth and fifth place respectively.
On the topic of trust, there is only a slight difference between the first three brands, but we do see that recognition does not equal trust. Paypal (66%) leads the pack ahead of Verisign (63%) and McAfee (62%).
This is where the test got interesting. When asked which single brand gave the user the best sense of trust between the top 3 and the rest shifted from being as wide as the North Sea to being as wide as the Atlantic with the percentual difference between Paypal and BBB being a staggering 22%. Paypal got 29% of the ‘votes’. Verisign (25%) and McAfee (23%) followed closely behind. BBB and TRUSTe scored 7% and 3% respectively.
So the last question, well, what can I say. Of course there are many brands out there, even with this question the top 3 brands from the questions 1 thru 3 (tongue twister alert) scored so ‘non-love’ points. SynergyDetox and Comodo scored the worst. For those of you familiar with the brands in the test, you will see that Comodo is present with 3 different logos. Comodo had a combined score of 26% (13, 11, 2) on the last question.
With Wufoo I set out to ask a few additional questions after the Usabilla part of the test was complete. Now, I would like to re-iterate that this is in no way a scientifically approach research project. The results of the following questions do give some food for thought though. One questions that arises is… how much potential income are you missing out on by using less recognizable trust logos?
Ok, so it is clear that trust logos have an effect… but then again, we knew this already. Just checking…
Interesting. According to 2/3 of the participants trust logo recognition has an added affect to their sense of trust.
Ok, again another verification of the affects of the presence of trust logos. 61% of participants said that they have at one time NOT completed a purchase because there were no trust logos present.
Ouch, this one does have to hurt, especially if it can be scientifically proven. 75% of the participants stated they have at some time NOT purchased a product or completed an online purchase process because they did not recognize the trust logo. Even though a trust logo was present, it seems that recognition precedes presence.
This question was in reference to this screenshot. It seems that a significant amount of users do pay attention to the https status in the URL address bar in browsers. This is great news!
The browser URL address bar seems to be a hot item according to the open end feedback we got in the survey:
So, the next time you need to review your SSL license, I would suggest digging a little deeper into who you are doing business with and how this can potentially affect your bottom line. In this case with trust logos it seems that recognition precedes presence.
For those of you interested in where all the 150+ participants came from, here is a small break down:
 |
 |
I have always wanted to do this test, but never on such a scale. I’m not surprised McAfee ranked well, but very Surprised Paypal did. Thank you for this.
Were you able to make the order in which trust icons appeared to the respondents randomized? If everyone saw the logos in the same order, then your stats are off. With survey questions where you have more than 3 responses, you nearly always want to randomize for accuracy because people react differently to stuff depending where on a list or screen it appears…. but you already knew that, right?
Hi Anne,
Thanks for your reply and exquisite attention to detail
The tool I used for the click heatmap, Usabilla, does allow for some form of randomness, but like the disclaimer said… this is not a scientific study. I could put a lot more work into it, but I guess I tend to go for the shock and awe effect before taking it another step forward.
In a follow up study randomness would play an important part (you weren’t the only one who mentioned this) as I am sure I can dump the logos into a testing tool that will show different grids of the logos together or even display them one at a time (this to validate… once again). My main goal was to get a conversation started that we should look beyond the basic facts of effects on conversion online. I think I reached that goal
Trust logos, or as the British call them trustmarks, undoubtedly play a significant role in persuading a visitor in the security of the website and that any personal information will be dealt with securely. But like other things in life, I am more inclined to get onboard a Boeing 777 aircraft than a Russian built Ilyushin il-96 because I am more comfortable with the Boeing brand flying me safely from A to B than Ilyushin, based on personal experience. So I don’t mind paying more for a ticket for an airline company that uses equipment that I trust. (I am not great with metaphores, so cut me a little slack… haha)
Thanks again for you feedback, I will definitely keep it top of mind when I retest. By the way, were there any brands ‘missing’ from the selection?
Take care,
Matthew
I agree with your disclaimer – it’s not a particularly scientific study, however I think you’ve done a good job of creating interest around a topic is regularly the subject of conversion testing.
Recognition of trust symbols is an interesting one; a good follow up test would be taking one of the poor performing symbols from your test and running a straight A/B test with an add to cart stage that doesn’t have any symbols at all. My point is, that it’s all very well and good asking people to state their level of trust between two items (one of which is well known), but what really counts is what people actually do.
As we know – what people say in response to a quant survey doesn’t always match behaviour.
Good article though Matthew!
@Brian – I’m glad you noticed the disclaimer
A/B testing would be great to check out. Variants such as:
- Original brand logo
- No brand logo
- Better known brand logo or Lesser known brand logo (whichever is applicable in reference to original brand logo)
Back to your statement about matching behavior with attitude is the holy grail as far as I am concerned. I have been looking into an approach like this combing a voice-of-the-customer tool with a customer-experience tool… in English, integrating iPerceptions’ webValidator survey tool with Tealeaf. With an integration like that you can scan survey results for specific attitudes, then replay the sessions in Tealeaf and potentially match the two together. I think it is possible and I honestly think that companies need to start looking into the potential of such solutions, maybe not directly on a monetary value, but pure on visitor insights.
Glad you liked the post and I hope the data will help you start some interesting discussions. If you have any test ideas, let me know. I’ve got access to a plethora of tools and various high traffic volume websites.
Hi Matthew,
I must declare an interest in this topic first, as our company has just produced a new type of trust seal so my views may be biased but I am doing my best to be objective!
It is really interesting stuff and just goes to prove something I have thought for a long time. People seem to place a lot of trust in “The Brand” rather than the information being presented that underlies that brand. If people took the time to look under the hood and see how or what those brands were doing to help websites establish trust, I think the public’s view might change.
As a consumer I want to know the information that is being relied on to tell me to trust someone. Where is it coming from? Who controls it? Can it be manipulated? So for me, site seals need to move a step forward so that instead of providing a blanket statement with no supporting detail, I want to see the detail, with a recommendation and then let me make up my own mind as to whether I believe or trust the information or not.
Our new Trust Passport product (trustpassport.com) has tried to do this. We aggregate information from government sources such as the companies registers and the tax authorities, from commercial sources such as map information on where offices are actually located, active malware and blacklist scanning, reputational information from web users as to their experiences and some help and advice on when things go wrong. All this is presented in a easy to view graphical format that can be checked out in 10 seconds. The sources are all clearly stated and most importantly, the fact that the company has been physically confirmed as having been contacted in the real world at their stated registered address. This ties real identity to virtual identity and will help elimate fakers. Finally, the information being presented is collated dynamically each time the seal is viewed.. The seal can be removed remotely if the website or company blots its copy book. We think it is a new way to help someone establish trust without having to rely on a Brand to do it for you.
We would very much appreciate it you could add our product to your review list the next time you conduct a survey of this kind.
Well done on an excellent piece of work! Great stuff!
[...] study shows the vast differences in recognition between leading marks. Take a look at the article: Trust Logo Recognition Precedes Presence | Actual Insights __________________ Ted S Find me on twitter: @teds027 Latest Post: Are you attracting a Fan or [...]
Matthew, we’d be happy to assist with a follow up study with randomized logo’s. We could set up for example 4 different cases (each in different, random order) and combine the results.
[...] Insights ran a set of tests to see which security logos’s people recognised and trusted and the results are very interesting and perhaps worrying if you’re showing a [...]
[...] If you want to discover how trust logo recognition affects online purchase behaviour follow the link below. It is widely accepted that trust logos can have a positive effect on conversion rates when purchasing online, but is that enough? What about which trust logo? http://www.actualinsights.com/2011/trust-logo-recognition-precedes-presence/ [...]
[...] http://www.actualinsights.com/2011/trust-logo-recognition-precedes-presence/ [...]
[...] Hi, This makes some good reading: Trust Logo Recognition Precedes Presence | Actual Insights Alicia __________________ > website certification from [...]
[...] keurmerken Uit cijfers van Actual Insights blijkt dat er in het buitenland maar een handjevol keurmerken is dat daadwerkelijk herkend wordt [...]
[...] did an amazing remote test on trust and logo’s which is also [...]
[...] interesting “Heat Mapping Tests” with Logos: http://www.actualinsights.com/2011/trust-logo-recognition-precedes-presence/ Share [...]
This is really cool. As you might expect, Verisign has become a joke. You may want to look at your OG tag settings. I tried to post to Facebook but didn’t get an image and it was using your home page URL and title.
[...] online, but does it matter to them who says it’s secure? A recent post at Actual Insights, Trust Logo Recognition Precedes Presence, indicated that it does matter, but we wanted to find out for [...]
[...] from Actual Insights suggest that just a handful of trustmark logos are actually recognised by [...]
[...] Process: Booking seemed easier than with Airbnb. No list of questions. Trustlogo’s seemed a bit small and few. Location of the Complete Booking button too far to the right… move [...]
[...] Process: Booking seemed easier than with Airbnb. No list of questions. Trustlogo’s seemed a bit small and few. Location of the Complete Booking button too far to the right… move [...]
[...] by Actual Insights showed that 76 percent of consumers did not purchase something from a website without seeing a [...]
[...] the blog actualinsights, the founder and CEO of TrustCloud, Xin Chung, read about Usabilla Live. His hopes were that this [...]
Thanks so much for this article. We’ve been debating this subject for the better part of year and it was very helpful. Mainly, we’ve been discussing joining the BBB, but it after reading this it doesn’t seem as effective as we though it might be. It’s surprising to see that Paypal did so well, being that they just give that away to any verified user.
Has anyone done a follow up on this test? I’d love to see it.
[...] to Actual Insights, we seem to trust certain logos more than others. Take the test and see how you [...]
[...] website’s security badges. But what security badges should you use? According to a survey from Actual Insights, 76% of respondents said the reason they didn’t purchase something online is because they [...]
[...] I am building a panel of fantastic folk like yourselves (am I going a little overboard with the smooching? hehe) who I can invite to participate in online research projects. Research projects like the one I did with Trust Logos. [...]