Pay With a Tweet vs. Pay With an Email [Case Study with Infographic]

By | Google+ , June 27th, 2012 in A/B Testing | 49 comments

Last year I wrote an eBook and created a viral landing page to give away the PDF version of The Noob Guide to Online Marketing in exchange for a Tweet (read the case study here).

One of the things I did was to ask people – using KISSinsights – whether they’d rather get it in exchange for a email instead of the tweet. Over time, the number of people saying they’d prefer to give their email rose to 45% so we decided to run an A/B test and see which virtual payment method would turn out to be more effective.

KISSinsights results from the visitor questionaire

There are two main ways to benefit from giving away an eBook/whitepaper, and then there’s the combo method:

  1. The lead gen approach: Asking for an email in exchange for your product. This allows you to build a list of potential prospects for future email marketing.
  2. The viral approach: Asking people to tweet about your product to receive it. This method builds viral momentum as every download results in social exposure of your product.
  3. The combo approach: Running both simultaneously to benefit from both approaches. An alternative is to do both (which we did for this A/B test) so that you can continue to get exposure while building a list that you can send a discount offer for your company’s product or service etc.

The test and the results

Page A – where we ask you to tweet for the PDF
Page B – has a single email field

In the graph below, you’ll see that our champion page (the email version) consistently out performed the Tweet version. In the end gaining a 24% conversion lift.

A fake email won’t get you my eBook

Yup, some people were dumb enough to fill in a fake email in order to get the eBook. And given that it was emailed to them after submitting the form, they wouldn’t receive anything. #FAIL.

If you want to think of conversion as intent only – then we need to include those bogus emails and the conversion lift drops to 16% – still a decent result.

What can we learn from this?

There are two main takeaways here, and it really depends on your business goals as to which you’d prefer.

The Fear of Public Tweeting

As I mentioned in the original case study, the main hypothesis for our test came from the fact that just like public speaking, people are often reticent to tweet about something (usually when it’s a personal account where the subject matter may be off-topic with what they usually tweet). Also, not everyone has a Twitter account, but everyone has en email address (or 12).

Once You Tweet, You Can’t Retreat
Well technically you can delete it – but some people will still see it. Hence the fear.

Collecting Emails Diminishes the Viral Potential Significantly

While we learned a lot from this test, what it has done is slow the virality of the page quite dramatically, which impacts how often your brand name gets out into the Twittersphere (note: you’ll see the Tweet version of the page if you click that link, as I prefer the impact we see from the viral exposure and ended the A/B test after writing this post).

Which method do you prefer? Collecting leads or gaining brand exposure?

– Oli Gardner

About The Author

Photo of Oli Gardner

Co-Founder of Unbounce. Oli has seen more landing pages than anyone on the planet. He is an opinionated writer and international speaker on Conversion Centered Design. You should follow Oli on Twitter
» More blog posts by

Comments

  1. Matt Fox says:

    Do you have any tracking on how many people who paid with a tweet actually subscribed via email after getting the book? Was there a lift in general email opt-in activity because of the viral activity?

    • Oli Gardner says:

      Unfortunately there’s no way to correlate the two. People that payed with the tweet don’t provide any data and they wouldn’t come back and register for the same thing via email anyway. They may sign up for our service, but again that would depend on inserting a UTM code in the ebook and hoping they clicked through from there – but more than likely they’d just visit the site directly.

  2. Brandon D. says:

    Your results make sense Oli. Like you mentioned, by asking for your visitors to “put their name on it”, it’s understandable that CVR would drop. I can see this being acutely problematic for sensitive topics. Dating advice, for example.

    But the pay-with-tweet version does get the added benefit of traffic. I wonder what the results would’ve been like if you required the email folks to re-email the offer to X names in their address book.

  3. david says:

    interesting. yes I’d go for email but not tweet for sure.

  4. Adam says:

    Oil, why don’t you have the tweet option after someone fills in their email address? Tweet “I just downloaded Oil’s free guide”

    That way you can get both the viral boost and the email…

    • Steve Young says:

      Great idea, Adam.

      Oli, great post as always. I would think that the value of an email is greater than the value of the tweet.

      Yes the tweet causes the viral loop but an email allows you to continually market to the consumer. Any results on how sales you’re able to generate from the emails?

      • Oli Gardner says:

        No results yet – we haven’t emailed the list yet, I’ll try and remember to come back with some stats when we do :)

    • Oli Gardner says:

      That’s a great idea Adam, I’ll try it out. Thanks.

  5. Aaron says:

    I would still go for email. As email can be remarketed, meanwhile, people will delete/remove wall post after getting the comment. (speaking from experience)

    Sometimes the problem I have which make me remove the share is the quality of the content is not up to par. It may have sounded good in the copy, but what I get was is disappointing. To stop my friends from subscribing, I remove 90% of them immediately after download the ebook to make sure.

    • Oli Gardner says:

      One of the best ways to ensure someone doesn’t delete the Tweet is to provide a preview of the content, so that the person getting it knows that it’s high quality.

    • Aboutime says:

      If the product is poor in return you should tweet a bad review. Save othe people wasting their time promoting it with a tweet. That should turn the tables and hopefully improve quality of what gets offered for download…

  6. Andrew K Kik says:

    Because you’ve decided on the ‘Pay with a Tweet’ option, could you provide an additional option on the download page:
    Option 1 – Download the PDF
    Option 2 – Have the PDF emailed as an attachment.

    With option 2 you would also collect an email address.

    • Oli Gardner says:

      That’s exactly what we did for the A/B test, 50% of people got to supply an email instead of the tweet.

      Would be interesting to offer both at the same time tho if that’s what you’re suggesting.

      • I’m suggest that you’ve already made up your mind to use “pay with tweet”. Once someone hits pay with tweet, sends out a tweet with your message, they land on a “success page”.
        On that success page, could you offer two options. One, download the pdf immediate. Two, have the pdf sent directly to your inbox.

        You’re not forcing the user to provide an email address, you’re just providing it as an additional service. Would that work?

        • Oli Gardner says:

          I don’t believe there is a success (destination) page option with PayWithATweet – although it could have changed. Definitely a good idea though Andrew.

  7. Hi Oli – Thanks for clearing this up in my mind. In the face of hearing everywhere online that social media is the only way to share, it’s good to get hard evidence on the continued viability of Email sharing.

    • Oli Gardner says:

      Yeah email is certainly valuable, but only if you have a good follow up strategy. The social angle can be much more effective, but more scary for measurement driven folks as it’s harder to measure. I’ve personally seen a MASSIVE difference in favour of the social approach.

  8. Victor says:

    I guess another approach to test viral behavior is to lock the content, say a download link, with modal box with a “like” button in the middle. Once user click the “like” button only then the download link is revealed. Many such apps already exist and can be implemented in no time. Same can be done Tweeting.

    The thing with Tweeting is that you are vouching for the product even before using it and that will make anyone hesitant. While liking a product on Facebook is much milder form of vouching in the users mind and have the same viral impact as tweeting about it.

    • Oli Gardner says:

      That’s pretty much what PayWithATweet is all about – so not too sure what you mean. Unless you’re suggesting the double payment – email then tweet to get it.

      I really like Adam’s idea above of doing email capture, then suggesting the tweet share on the confirmation page.

      • Holly says:

        I think what Victor is saying that when people pay with a tweet, they have to pay (tweet) first before they’ve read the pdf to know if they really want to recommend it or not. They are in effect sending their subscribers to the page sight unseen on trust. Not everyone cares about that, but a lot of people do. I know it makes me hesitate to pay with a tweet if I don’t have a prior trust relationship with the person whose page I am tweeting.

        • Oli Gardner says:

          This is why it’s important to include a preview of a portion of the document to let people see the quality in advance.

  9. Victor says:

    I find the use of term “Pay with the tweet” a little too aggressive, specially when you are giving it away for free.

    I would have gone with appealing the user behavior of “returning the favor” in exchange of the free gift. Something on the lines of “Help us spread the word…”, you know what I am getting at.

    • Oli Gardner says:

      Yeah, good call Victor – and you can design the initial button any way you like so that would work well.

      ***** Love all the great ideas coming on this thread! ******

  10. Lee Driggers says:

    Or perhaps, an initial campaign with “Pay with an Email Address” and some sort of follow up (similar to an upsell) such as, “Enjoyed our ABC product? Pay with a Tweet to get our XYZ product.”

    Granted the # of Tweets will be significantly smaller, but I bet those tweets are way more valuable from what I would think of as a “qualified Tweet”. By that I mean a user that probably used XYZ product will add to and be more passionate about their Tweet and probably has a much larger/relevant audience to Tweet to.

    (forgot to press “Submit Comment” yesterday)

    • Oli Gardner says:

      Another great idea. I may compile these all into a post for the future.

      Thanks Lee.

    • Holly says:

      I like this one too – if the reader found the content valuable, he or she would have less hesitation to tweet a link to the next item.

  11. Liam Gooding says:

    Hey Oli, You could always do both: capture an email AND pay with a status update?

    http://virallyapp.com

    This is exactly what Virally does – sits on your landing pages and offers an easier way to trade an email and an update for an eBook.

    Disclaimer: I’m the founder of Virally

    • I’m a Virally App user and I’m very satisfied using it.

      I’ve already tried paying with a tweet, but the results are very poor. Anyway, what really has value for me is getting the email for further lead nurturing.

      Virally has the best of both worlds: getting not only the email, but many more information from the lead’s social profile, and promoting my service thru the lead’s timeline, all in a very simple way: a download button;

    • Oli Gardner says:

      Thanks Liam – really good to know about a new app that covers both these bases. I’ll be trying it out.

      Cheers
      Oli

  12. lisa says:

    total pants – couldn’t get the download, but you got my tweet ;o((( You don’t seem to support mac users. Is that the same with the rest of your product?

    • Oli Gardner says:

      I prefer #TotesPants Lisa.
      If you didn’t get it, it might have been an P.W.A.T. bug. Email me at oli@unbounce.com and I’ll send you the direct link. (note I might be a little slow – currently on holiday).
      Cheers
      Oli

      p.s. We’re all Mac’s @Unbounce

  13. Jason says:

    While conversion is a great metric to measure the success of this test, shouldn’t the real measuring stick be the number of downloads each option produces?

    Although it’s initial conversion is lower, I wonder if the 4% conversion difference from the Pay with Tweet option is made up by the viral effect you get from followers of the person tweeting your link coming to your site and repeating the process themselves.

    As you alluded yourself, the next step should be to test using both at the same time. In fact, it would probably be best to add Facebook and Google+ options to the mix or even test them against the email option to see which converts the best.

    My initial concern with offering 4 options (email, Twitter, Facebook, Google+) would be a drop in conversion because visitors would be overwhelmed by the choices, but it would still be interesting to see if the viral effect would make up for any initial drop in conversion.

    Jason

  14. […] de um formulário. Assim a barreira para o download fica bem menor (estudo de caso sobre isso aqui) e, principalmente, você consegue coletar informações de contato e gerar oportunidades de […]

  15. […] de um formulário. Assim a barreira para o download fica bem menor (estudo de caso sobre isso aqui) e, principalmente, você consegue coletar informações de contato e gerar oportunidades de […]

  16. Hisocial says:

    You can also use “pay with a Facebook like”. From what we’seen at Hisocial, tweet/ like campaigns work better than “paying with an email” campaigns. This applies not only to white papers but to coupons and giveaways

  17. […] of the most fun and surprising A/B tests we ran was to offer up an ebook in exchange for either an email address or a tweet (on a 50/50 traffic split). It was a dual purpose test: by splitting it up this way, we got leads […]

  18. […] of the most fun and surprising A/B tests we ran was to offer up an ebook in exchange for either an email address or a tweet (on a 50/50 traffic split). It was a dual purpose test: by splitting it up this way, we got leads […]

  19. Johan says:

    In both cases you still need to market the book. The tweet will ‘remarket’ (spread) it to others. So my question then is: how did you market the ebook?

    • Oli Gardner says:

      There were a couple of approaches (that we’ve refined recently with newer ebooks). For this one we mainly did it via social.

      For our most recent ones, we used our new increasing lists (from all of our ebooks) to market via email – which has been very successful, and also via banners at the end of blog posts.

      For the email campaigns we got a 74% conversion rate on the landing page we sent people to, and 43% from the banner. This was for our “Ultimate Guide to Landing Page Optimization” – http://get.unbounce.com/lpo-guide-email/

      And so with the social viral potential together with the large list, it was a really successful campaign.

      Hope that helps.

      And grab the ebook from that link :)

      • Johan says:

        Thanks for the reply. The issue is that, in order to do an email campaign, you need email addresses. An email campaign usually goes 1 step deep (the receiver). A tweet starts… where?… and it ‘should’ have more than 1 step deep (re-tweets). So how did you measure this? Isn’t that comparing apples to…. toasters? Just curious..

  20. deborah says:

    where does the pay with a tweet button appear? on your twitter page?

    • Oli Gardner says:

      You can place it on any web page. In this example it’s on our landing page. (You just need their script to place it with their button – or create your own). Go to PayWithATweet.com for more details.

      • deborah says:

        so can i have a button on my website that says ” extended trial” and clicking on that takes my users to a page that says tweet for us and get an extended trial???

        • deborah says:

          inbetween thanks for the reply..:)

        • Oli Gardner says:

          You can make a button that says anything, but it would be a little misleading I think to ask someone to tweet your message – although if you make it very clear what will happen when you click the button it might be okay. You’ll need to make sure that the person on the page can get the same deal for themselves.

  21. […] executed well, both approaches can produce healthy conversion rates. But as demonstrated by some studies done by the blog ubounce (described with infographics), some sort of combination of the two is likely the best […]