Social Proof Tool Boosts Landing Page Conversion 32.4%

Inbox Ideas | Email Marketing Tips by AWeber

Posted by Justin Premick

The recent release of a new subscriber count chicklet met a mixed response.

We’ve read plenty of positive comments on the blogs who covered this feature.

But there were definitely others (both in the release post’s comments and on other blogs) who doubted the usefulness of such a feature.

Of course, this is one of those cases where you should apply the best lesson in marketing — Test!

Fitness expert and AWeber user Carl Juneau did just that, setting up a split test on his landing page to see if the presence of the subscriber count chicklet affected opt-in rates.

He swung by our blog the other day to share his results.

Adding The Chicklet Increased His Opt-In Rate By 32.4%

In a comment, Carl shared his results:

Justin, Mark, other readers:

I implemented the the readers widget on my landing page the day it came out in an A/B split using Google Optimizer (services.google.com/websiteoptimizer).

Here are the results:

OVERALL
Visitors: 552
Conversions: 140
Conversion rate: 25.36%

PAGE A: NO WIDGET
59 conversions / 271 visitors
21.8% ± 3.5% conversion rate

PAGE B: WITH WIDGET
81 conversions / 281 visitors
28.8% ± 3.7% conversion rate

Page B chance to beat Page A: 97.2%
Improvement: 32.4%

Results impressed me and are significant at the a=5% level. I only have 2300 readers, so even (relatively) low numbers benefit conversion.

Key Takeaways

You don’t have to have a huge list for this to work.

I think a lot of people believe social proof is only effective when you’re talking about huge numbers of people, like the McDonald’s “Billions and Billions Served” signs.

Carl doesn’t have billions of subscribers. He has just over two thousand. And yet, the chicklet still made a big impact on his landing page conversions. Why?

Here’s what I think: the more homogeneous your audience is, the more powerful social proof is.

If McDonald’s said “2300 Served,” well… who cares? There are millions (billions?) of people who eat hamburgers. 2300 isn’t that many in the grand scheme of things.

But narrow your audience down to people actively interested in getting six-pack abs, and suddenly 2300 readers is a helpful cue that Carl knows what he’s talking about.¹

Test. Then, test some more.

Carl’s not haphazardly adding and removing stuff from his page.

He’s using Google’s Website Optimizer to test and track, and he’s making sure to get statistically significant results.

As we’ve discussed before, split testing is a crucial element of a successful marketing campaign.

So start today! To help, here are some split test ideas you can use for your email campaigns.

More Split Test Results To Come Soon

I’m in touch with Carl directly to discuss some other possible tests we can do to raise his opt-in rates further (and share the results with you, of course!).

Want To Get The Results In Your Inbox?

Join our blog newsletter and we’ll email them to you when they come out.

We’ll also periodically send you other email marketing tips we publish here. The newsletter goes out 1-2 times per week.

Naturally, as a permission-based email marketing company, we respect your privacy.

1. For more on social proof, I strongly recommend Robert Cialdini’s Influence: The Psychology of Persuasion and his newer book Yes! 50 Scientifically Proven Ways to Be Persuasive. Both give numerous examples of how social proof affects decision-making.

Tags: Case Studies, list building, testing, web forms

Print This Post

20 Responses

Jancice
July 25th, 2008 at 10:12 am

I would like to give this a try, however I’m not interested in the entire
optin form. I would only be interested in the actual Reader Count.

We have a very specific Option form and would like to place Reader Count below "our" Customized optin form.

Is this possible at this time??

Also can the term "Readers" be changed to something like "subscribers"
"followers" "fans"
Ray Randall
July 25th, 2008 at 1:55 pm

The insidious facet of internet marketing is wondering about results. "What will happen if I do this?" is seldom asked. Most of us just do something with hopes for the best.

Testing seems tedious, yet necessary. Learning how to test, testing, and adjusting is necessary. There’s nothing static about serving customers or interested readers.

"Test. Then, test some more."
Mike Herberts
July 25th, 2008 at 5:35 pm

A great feature and I will start an a/b test immediately.
I DO however agree with Jancice. Please give us the option of using the term ‘readers’ subscribers’ ‘fans’ or similar. I send out videos rather than a newsletter and ‘readers’ just wouldn’t make any sense to my site visitors. And hey guys….keep at it…you’re doing a great job.
ted
July 25th, 2008 at 6:22 pm

If social proof is so great then why doesn’t Aweber use it in the sign up form above?
Gobala Krishnan
July 26th, 2008 at 10:57 am

It’s definitely a useful feature. However if you’re starting at zero subscribers you may want to exclude the chicklet till you have 100 or so subscribers. Social proof can also work negatively - ie others see you have no subscribers at all (or too little to even impress them), they may think twice about signing-up.
Cassandra
July 26th, 2008 at 11:29 am

So what is this widget that people are putting on their website that is getting people to opt-in at higher rates?
Steven Vrancken
July 27th, 2008 at 2:55 pm

It would be nice if it is possible to change `readers` to `downloads` also. For on my opt-in pages I offer a free download, I do not offer something to read.

Will this option be available soon?
June
July 29th, 2008 at 11:18 am

I have to agree with Gobala. With a brand new site and no list, hence no clicks on the chicklet, it might prevent people from wanting to sign up.
Jancice
July 30th, 2008 at 9:26 am

If you don’t have any subscribers yet, just use the regular optin without the chicklet.

When you do get a list, then use the chicklet.
Alex Price
July 31st, 2008 at 3:15 am

What is your overall marketing method
to drive traffic to your site, when using
this method?
Justin Premick
July 31st, 2008 at 7:59 am

Alex,

I’ll have to check with Carl to see if he wants to address that here.

Is there a traffic generation method for which you would expect the subscriber chicklet to work particularly well (or not well)? I’m curious about what you think the relationship between traffic source and the chicklet’s effectiveness might be.
Mike Herberts
July 31st, 2008 at 9:09 am

I think any subscriber method that requires the website owner/promoter to be able to prove that they have something that other people find of interest, will see a benefit.

I think there is a danger in over analysing the psychology here . . .It’s kind of…’Why would I want in, if nobody else does’ (or the reverse of that)
Alex Price
July 31st, 2008 at 1:28 pm

I was just wondering the
method Carl was using to
attract visitors to his site
so that he could see the
effectiveness of the chicklet.

As Gobala and Janice said
it wuold make no sense to
use it if you do not have a
list of over 100 subscribers
yet. As I am just starting out.
Carl Juneau
August 1st, 2008 at 3:31 am

Hi guys,

Traffic is 80% from PCC. Mostly Google, a little Yahoo.
Want To Double Your Blog Newsletter Readers? Here’s The First of 4 Tactics That Helped Us - Email Marketing Tips by AWeber
August 4th, 2008 at 2:54 pm

[…] placed at the end of posts for a while now (example: our recent post on increasing opt-in rates via social proof – click the image at left to see that signup […]
Greebo
August 5th, 2008 at 7:43 pm

I agree with Gobala - advertising how many subscribers you have it fine if it’s a decent number, but if there’s not very many surely it will just put people off. Who wants to subscribe to a newsletter that’s only read by the author and his mum?
Moorman
August 7th, 2008 at 1:55 pm

Great information. I will start to use split testing on all my campaigns. I wonder can I add a sign up form to my wordpress blog?

I’m going to start selling a resource manual for indie musicians. I’m going to create a three landing pages and split test them all.

Do you think it would be a good idea to implement a couple of different ways you can join.

For example, you could have a opt-in form with name and email.
One with just email opt-in and one with a couple of questions to answer. How the peron answers the questions determines what opt-in form see.

I don’t know. I’m just trying to play around with the concept.
Justin Premick
August 7th, 2008 at 2:42 pm

Moorman,

We have some information in our knowledge base on how to add a signup form to your WordPress blog.
adam libman
August 10th, 2008 at 1:52 pm

this is very interesting stuff. I’ll try in on my site!
chris
August 13th, 2008 at 12:09 pm

I was split testing with a pop-in box. In this case, it was the same content but different times periods in which it popped up. I did NOT have the chicklet at first. My results were about the same for all time periods…and none were acceptable rates of conversion.

I added both the chicklet and a link to a past newsletter so the visitor would see how many others subscribe and the type of content to expect. My subscription rate sky-rocketed!

I just checked to see how many people viewed the sample newsletter (pdf format) and it was a surprisingly low number. It looks like that chicklet did the trick. And my existing subscription count at the time was below 40 so even that low of a number was still high enough in the eyes of the visitor!

Thanks, Aweber!