Do privacy reassurances hurt your sign-up rate? Test results...

Latest posts | Feed | By Mark Brownlow on July 22, 2008

Now there's a thought. Bill Nussey raised the issue in a post a few days ago and I decided to test it out here at the Email Marketing Reports site.

The theory


Typical advice on optimizing sign-up forms and pages suggests including explicit reassurances on privacy, rather than just a link to a privacy policy. Here's how my sign-up form does it, with the little "your privacy is guaranteed" note:

sign-up form

On my sign-up page, there are also reassurances that submitted email addresses are not sold or passed on to third-party advertisers etc., reflecting my actual privacy policy and practices.

The research cited by Nussey suggests that in some circumstances, mentioning your privacy safeguards gets people thinking about privacy issues they might otherwise never have considered. The result can be increased reluctance to submit personal data.

Could it be that removing that privacy reassurance might bring more sign-ups?

Here's the alternative sign-up form that ran for the last two weeks:

sign-up form

(I also removed the privacy reassurance from the main sign-up page.)

The "visitor to subscriber" conversion rate for the "no mention of privacy" alternative was 23% higher than for the standard form with privacy reassurance.

Let's clarify that...by removing privacy reassurances on the sign-up forms and pages, visitors were 23% more likely to submit their email address.

Stop a moment, though, before you rush off to change your own forms.

1. This was not a statistically rigorous A/B test. The standard form ran for two weeks, then the "privacy free" form for the next two weeks. It could be that some other factor improved sign-up rates, such as better qualified website traffic.

2. Though 23% is a big difference, it's still just about within the boundaries of normal fluctuations in subscriber conversion rates.

So it may be that the improvements are random and there is no direct link between removing the privacy assurance and increasing conversions.

Nevertheless, it certainly suggests you might consider testing this phenomenon yourself with your site and audience.

3. Supposing you find that mentioning privacy hurts sign-ups...this doesn't imply you should ignore privacy. Your actual policies should still be protective of the subscriber in the spirit of ethical email marketing based on explicit permission.

It should still be clear what people are getting when they sign up, so they can make an informed decision about submitting their address. Equally, your privacy policy should also be accessible and address the fate of email addresses clearly.

4. Perhaps it's not about the presence or absence of a privacy reassurance, but the wording of that reassurance.

Is "Your privacy is guaranteed" too vague and tenuous to be truly reassuring? Does it instead raise doubts? Perhaps a more specific statement would have a more positive influence, such as:

"Your email address will never be passed on to others"
"Your email address is only used to send you newsletters"


...or similar.

The critical lesson is perhaps simply to be more critical about accepted truths and simplifications. Test, test, test and test some more. Then test again later to see if anything's changed.

If anyone else has tested sign-up form language, please let us know your insights!

Tags: ,

Sign-up for the Email Marketing Reports NEWSLETTER
Twice a month, free, packed with email marketing advice and all the posts from this blog.
Email:      First Name:     
    More info and sample

6 Comments:

Great test Mark. Thanks for taking the lead on tests like these, this definitely helps open the door for others to start optimizing their email marketing.
By Blogger Adam Covati, on 23 July, 2008  
 

Thanks Adam. I'm now building another website with a more mainstream audience to (I hope) provide a venue for larger more detailed tests sometime down the road. There is just so much that could be tested!
By Anonymous Mark Brownlow, on 23 July, 2008  
 

Mark,

Thanks for sharing your results!

We've been testing privacy reassurances on our signup form for a couple months (several variations on wording as well as a "no mention of privacy" version).

Our results thus far haven't yielded any clear winner. The "no mention" version is one of the poorer-performing ones, but the difference is nomimal.

While our testing hasn't suggested any major conversion gains by changing our privacy reassurance, your example shows it's obviously something worth testing.

Thanks again.
By Anonymous Justin Premick, on 23 July, 2008  
 

Interesting - thanks Justin. More evidence of the value of testing, rather than assuming what works for one site would work for another.
By Anonymous Mark Brownlow, on 23 July, 2008  
 

Great info. Why don't you set up a Google Optimizer test so you can get a statistically significant result?
By OpenID starpointe, on 25 July, 2008  
 

Thanks for the tip starpointe: the Google tool was originally designed for testing landing pages so I'll certainly use it on the sign-up page...is it suited to testing whole websites, too?
By Anonymous Mark Brownlow, on 25 July, 2008  
 

Post a Comment