PayWithaTweet (hereby known as PWaT) is a service that enables web publishers to give visitors content in exchange for a Tweet or Facebook share. In this post, I will share my experience with PWaT, tweet and CTR data, and how you can use this tool to go viral and influence search engine results.
What is PayWithaTweet?
The creators of PWaT have made a great video explaining the concept and some of its uses:
The Experiment: Tweet for an Advanced Google Analytics Cheat Sheet
I created an Advanced Google Analytics Cheat Sheet during one of our Portent 'Hack Days.' I wanted to distribute the Advanced Google Analytics Cheat Sheet far and wide – and PayWithATweet was something I always wanted to experiment with.
The Set Up
After creating my Google Analytics Cheat Sheet, I uploaded it to WordPress and created the PWaT button.
When you use PWaT, you get to write the message that is tweeted when visitors download your file.
I did some quick keyword research (turns out "Cheat Sheet" is more popular than "Cheatsheet"). I also thought Portent's strong brand in the internet marketing community would entice people to click through, so including Portent in the tweet was important.
I then did hashtag research: #Measure is a very popular web analytics hashtag, and I believed the #measure community would find the cheat sheet interesting and useful. So I needed to write a message that included all of these elements in only a few characters.
I set the PWaT Tweet Text to "Get the Advanced Google Analytics Cheat Sheet from Portent - #measure - https://bit.ly/mLBPht", and created a simple landing page. (My wise and generous boss Ian later added a screencap and made it better.) Then on Friday morning, I tweeted out the link, and asked some of my colleagues to do the same.
The Results
The Numbers
Because Twitter is a real-time medium, I took measurements at various times to watch the tweets spread.
I used bit.ly tracking to compile these metrics. You can see the bit.ly numbers yourself at the link's bit.ly information page - https://bitly.com/mLBPht+. (If you see any differences between the numbers in various graphics and references in this post, it's due to the time the numbers were taken.)
As of 5 PM on Saturday, 83 people have downloaded the cheat sheet for a tweet. There are four double tweets, and five of the tweets were from colleagues at Portent I asked to Tweet. This puts the organic Tweet count at 74 Tweets. The bit.ly link received 258 clicks, so 28.7% of the people clicking on the bit.ly decided to download the cheat sheet and Pay With a Tweet.
The landing page was probably the weakest part of the experiment. (The Kanye West reference in the first paragraph was an unsuccessful attempt at writing like Groupon.) I imagine had I created a stronger landing page and a stronger message, all of these metrics would be better. I'll come back to lessons learned in the "insights" section.
The Search Engine Impact
Major search engines index Twitter, and these social shares appear to be a ranking factor in both social search and traditional search.
Jen Lopez of SEOmoz has shown tweets have a substantial effect on search in her well known "beginner's guide" case study.
I was curious if we would see the same thing, so I looked at the search engine rankings of two terms – "Google Analytics Cheat Sheet" and "Advanced Google Analytics Cheat Sheet".
In Google (signed out, personalization off, but from both office and home machines), the post ranked almost immediately for "Advanced Google Analytics Cheat Sheet." (It's apparently the only piece of content on the web optimized for that phrase.) However the SERP for "Google Analytics Cheat Sheet" was dominated by Ian's last Google Analytics Cheat Sheet and pages linking to it.
On the first day (Friday), when the bulk of the tweets were occurring in real time, I didn't seen ay modified conventional results. I did see some Tweets being pulled in through 'social circle' search functionality.
However, on Saturday afternoon, I found that the landing page was now ranking in position eight for the term "Google Analytics Cheat Sheet."
Fascinated by the result, I used the SEOmoz Competitive Analysis Tool to analyze the "Google Analytics Cheat Sheet" SERP:
By Sunday, the Cheat Sheet had received over 85 tweets. Now the page was appearing in position six for "Google Analytics Cheat Sheet."
And again, according to the SEOmoz Competitive Analysis tool, the page had the lowest domain and page authority in the SERP. The page has, to my knowledge, no inbound external links. Tweets are the page's only signs of quality and relevance.
Neither domain authority nor page authority is high enough for the page to rank for that phrase. While it's hard to control for personalization, I had signed out and added the "pws=0" string to the URL. I tried it on a variety of browsers - including some I don't use - and I received the same result.
With the standard SEO disclaimer (correlation does not equal causation, what works here may not work for you, this is not science nor does it claim to be), it seems that Twitter thru PayWithaTweet can influence Google search results.
At the time of this writing, the page is not ranking in Bing for 'Advanced Google Analytics Cheat Sheet' or 'Google Analytics Cheat Sheet'. (In fact, the page isn't indexed at all in Bing.)
Insights for Future PayWithaTweet Projects
When to Use PayWithaTweet
If you have content you can give away to a Twitter or Facebook-using audience, PWaT is a great promotional option. Virtual goods, eBooks, mp3s, or coupon codes are great things to give away.
Justin Briggs had more ideas about what to do with PWaT:
Given these changes in search engine rank, it makes sense to do a PWaT promotion around high-traffic seasonal terms. For example, if you sell Mother's Day Flowers, a PWaT promotion for 5% off could help you rank for the highly competitive term "Mothers Days Flowers".
Keep in mind people can delete their tweet or share after they download the file. And in the case of a coupon code, there's nothing that would stop someone from sharing a coupon code with their friends, tweeting it themselves, or submitting it to a coupon site.
As always, you're playing with fire. Play carefully.
Start with a Big Audience
My PWaT experiment didn't really take off until Ian tweeted it to his 7,394 followers. You need to start your PWaT with a big Twitter audience.
If you don't have a big Twitter audience, you need to get someone that does to Tweet the initial link. This may be a good time to use ad.ly’s sponsored Tweets.
Include a Popular Hashtag in Your Tweet
I included the #measure hashtag in the PWaT message. The right hashtag gets your message in front of a large number of the right people. Use hashtags.org to find the most active hashstags in your niche.
Make a Great Landing Page
Landing pages are really important to the success of your PayWithaTweet initiative. In my experiment with PWaT, this is probably the thing I did the worst – I imagine had I improved the landing page, we could've gotten more than 125 tweets.
Your landing page has to entice people to download the content and share it with their friends, while not giving the piece away for free. As they say, people don't buy the cow when they can get the milk for free.
It's also important to find a way to show social proof – how many other people are grabbing your giveaway for a tweet. Using Twitter's Search Widget set to grab your PWaT tweets and showing them on the landing page is a great way to do this.
PayWithaTweet Uses an iFrame – Make Sure Your Publishing Methods Supports Embedding an iFrame
The PWaT button appears in an iFrame. There's no other way to implement it (to my knowledge). If you use a CMS that strips out iFrames (like WordPress), be prepared for this. We used the iFrame WordPress Plugin, which worked like a charm.
Use the Science of Retweets to Optimize Your Message
Social media experts (most notably Dan Zarella) have done a great deal of research on what kinds of messages get retweeted. Use this research in writing your PWaT message for maximum social reach.
- Use words and phrases like "New" "Please retweet" "How To" "Check Out", and other terms that draw more retweets
- Use Colons – People seem to like to tweets with colons over tweets with semi-colons or other punctuation
- Noon to 6 PM seem to be the golden hours for Retweeting
- Mondays and Fridays are the best days for Retweets, so launch your PWaT content on those days
Thanks for reading, and I'll answer any questions in the comments.
(I'd also like to thank my colleagues at Portent who helped with this project - Michael for editing the cheatsheet, Anna for figuring out how to embed iFrames in WordPress, Ian, Doug, Josh, and Aaron for tweeting, and of course my bosses, Ian, Tracy, and Elizabeth, for letting me do this.)
Can I say "Thanks With A Tweet"?
LOL
The way I understand it is that since the Linkscape index has not been updated since June 6, you would not have seen any PA (page authority) using the Competitive Analysis tool, even if there was some.
You would have to wait til the next update on the 20th of this month (https://apiwiki.seomoz.org/w/page/25141119/Linkscape-Schedule)- however the updated PA will likely be skewed now by including links from this post at least - am I wrong in this assumption?
Hi David,
You're right. The page is uncrawled by SEOMoz. (I realized this after I submitted the article for publication.)
Best,
Matt
I was just about to post the same issue. Nevertheless, its a great job Matt, and we all have to thank you for the opportunity you have given us to know more about this tool. Thx for sharing.
I've actually had a decent amount of success with this in the past. I think my numbers for one adwords campaign had me paying under a buck per tweet at one point.
That's interesting - so you were dropping adwords traffic onto a PWaT landing page?
How did the overall cost per click (if you basket the clicks from PWaT and the clicks from adwords) work out? (Feel free to keep specific names or numbers quiet, but that sounds cool and I'd love to know how it worked.)
Great post and a great service. What I like most is that the tweets and shares will happen in a much more natural way, spread out over a longer period of time and not just over a few days.
What sort of long-term ranking impacts have you seen or know of with this type of service?
Hi Bryce,
With the standard SEO disclaimer, I suspect the rankings don't stay unless the content attracts permanent links from somewhere else.
That being said, most things that go big on Twitter get a large number of links from Twitter-related services like paper.li and then get social bookmarking links, so this is part of the natural link profile of social content.
Great to see this on here, i use that button too, its sweet!
This is awesome... I have a musical friend who uses this.
Where is the trade off between PWaT and asking for email address?
With email - you build a list & can market to them later --- but with PWaT, you might gain in rankings and be found by others.
Of course, you can do a search in twitter to see who tweeted your stuff, and send them an offer via @message, but you still didn't collect their email address when you had the chance...
What do you think?
That's an interesting question.
I don't really see PayWithaTweet as being equivalent to collecting an email - obviously collecting emails (from the right people) can be a LOT more profitable over the long term. However, that will generate less referral traffic and positive search engine rankings than PayWithATweet.
I simply would use them differently. Use squeeze pages to collect email lists of customers, while use something like PayWithATweet to give content to the Twitterati and get tweets about your site.
I wouldn't push the same piece of content in a squeeze page as in PayWithATweet. For PWaT, I would want something simple and spreadable, while for a squeeze page, I want something substantial and trust-building.
Good question - eager to hear other folks' opinions. But PayWithaTweet is definitely a bank shot - maybe even a double bank - while collecting an email is a direct play.
Thanks for sharing
That's really cool that it worked for you. Well done video too. How have the results been on day 2, 3, etc. Was there a big dropoff or did you see good results day after day. Also have you seen an increase since posting this article?
Great way to get your name out there.
PWaT looks like a very useful tool. It would be interesting if you could limit downloads to users with x number of followers or with a Klout score of x.
That's a great idea.
I don't much about it, but it looks like Involver (https://mashable.com/2011/06/22/klout-gate/) does something like that for Facebook.
Nice...will have to get working on my social influence, don't want to be missing out on the good stuff!
How is PWaT not a paid linking tool cleverly disguised as a social media promotion tool?
PWaT allows people to give something of monetary value in exchange for a tweet with a link in it. Yes, it's on Twitter, but there is a clear quid pro quo: product for a link.
I happen to think that it's a wonderful tool, but I just wonder how long we'll be able to give away things on social media that brings SEO benefit without ending up being labeled as Black Hat.
Help me justify this tool as White Hat, SEOmoz friends!
I am wondering if this strategy will work on Indonesia visitors ... I will try this, and hopefuly it works in Indonesia :D
I got really excited about this and added it to my site. Then when I tested it, I found out anyone who was paying me with a Tweet was agreeing to:
This application will be able to:
Then I had a heart attack and quickly took it down. Luckily no one paid me with a Tweet before I could rectify this.
Great article Matt, though I just noticed it :) Has any changes to the PwaT (and other social channels) concept in the eyes of search engines come up since you wrote this article?
I was wondering if this solution would be a suitable alternative to freemium tools, i.e. sign up for the free package by paying socially? It will surely take a hit on the sign up conversion rates, but would be interesting if others have tried this, or hearing your thoughts Matt.
This was a good article. I hadn't thought about the SEO benefits of something like this.
Great post! Pay with a tweet is a great tool and can really help leverage the viral spread of your followers. Need to leverage Twitter now before they start learning how to make money themselves :)
Great resources and tips.
Dynamite Post! Especially with the hashtag research, it can be almost as important as SEO keyword research, but in the context of Twitter of course.
Very Nice, I never knew tweets can be that beneficial ;)
Really good article. You provided excellent information and insight about the project.
Bloody excellent service, I wish I'd heard of paywithatweet earlier. Thanks for the enlightening article.
I just had a discussion regarding this same concept and there were opinions on both sides of the fence. We were debating whether we should require people to either sign up with their email, share on Facebook, or tweet in order to access our white papers. The point that came up though was whether people would be willing to tweet/share content with their friends before they were able to read it and confirm it was quality stuff. How well do you think this would work?
In my experience, people don't buy the cow when they can get the milk for free.
If you want to use PayWithaTweet, make sure there's enough of the content on the landing page for the visitor to understand what they're sharing and getting.
Beyond that, it's about knowing your audience. A whitepaper on online marketing or metrics might be great for PayWithaTweet, but there's lots of categories (like a piece of content on health, or finances, or finding a new job) where people just won't feel comfortable with that.
When in doubt, do a test, and see what you get.
Not bad nice concept to make it viral your product with twitter or facebook button, and thanks for sharing the knowledge becuase this button is new for me.
Thank you for describing each part of the campaign so thoroughly. Makes it so much easier to convince self + others to use it :)
The SERP impact is likely less after Google lost access to Twitter, but it should bring some quality leads and engagement nonetheless.
Thanks for sharing.
Great post Matt! You have giving out some great ideas here!
Good stuff! Now for PWaT to add PWaP (pay with a Plus.) Or maybe that should be PWa+?
Seriously though, many commercial pages like-gate/fan-gate content on FB, and it would be good to see more implementation of "paying with a share" versus just goosing the numbers with some fairweather fans.
Good point with the hashtag research, it can be almost as important as SEO keyword research, but in the context of Twitter of course.
duplicate comment, please delete.. sorry!
Pay With A Tweet is a pretty interesting tool, it's definitely a very helpful and measureable action... and it can be used in dozens of interesting ways.
I had a chance to play with it when it was still pretty new, and the Facebook Status was kind of buggy. I haven't used it in like a year, anyone else use it recently? If so, have they fixed the FB issue?
I was a little skeptical of your article and the rankings, especially since the screen shot shows the site in purple which means you are using the same browser with the browser's web history turned on.
But, I just did the search, incognito, and you are maintaining position #6. What will be interesting now is to watch your analytics for that search term and the number of clicks your site gets and then compare any ranking boosts based on click throughs on that term.
Oh... and thanks for sharing the PWaT site.
Carry on.
Thanks Thos,
I was worried about result personalization too - so I used the SEOMoz kw difficulty tool.
However, given the sheer number of factors present in today's SERP (geography, personalizaiton, etc.), I generally wonder about the validity of rank checking as a whole. How do we know are SERPs aren't geographically different, even if we have everything turned off?
If you have any suggestions on future measurement strategies, I'd love to hear them.
Thanks for the kind words,
Matt
Great Article! I love the detailed analysis.
Seeing as the Pay with a Tweet concept certainly does have potential to influence rankings (the page was #6 for me as well) I'm wondering if it will ever be considered "Paid Linking" or some other form of unsavory manipulation in the eyes of the anti-spam team. Personally, I think it is a social transaction where value was traded for value- and if enough people really do care about the service and find value in it enough to share with their friends it is probably high enough quality to deserve to rank higer. If massive twitter botnets start impacting search results for spammy products, the party will likely end quickly... but for now it seems to be a valid SEO tactic, and the social traffic/exposure sure doesn't hurt either : )
I think they can use authority scores and geography to deal with some of the spam potential.
As for the search engine spam team, I see two ideas in play:
1) You're paying for something that affects rankings directly. Spam.
2) Lots of people are tweeting and sharing an excellent resource - surely they are doing this because it's excellent and that's a signal we should use?
Additionally, these are nofollow'ed links. If there's a way I can nofollow a tweet or facebook share to stay compliant in the eyes of search engines, I would love to use it. But right now I don't know how to do that.
It's pretty common to have a promo that involves tweeting something, and the life-span of a tweet is pretty short. I doubt Google will throw it in the same category as "link buying." These are real people tweeting it rather than fake twitter accounts with robots.
Yeah it seems to be within the 'spirit of the rules', but you never know. It would be great to get a ruling from the Spam team here.
I think the idea of spambots and virtual users is a great one! Anyone know who (if anyone) is working on it?
https://matthewhunter.snappages.com/blog/2011/07/14/the-future-of-seo-is-buying-people-not-links
it's really interesting, i didn't know PWaT. The best part is the analytic one, now we need some nice content to try
I'll do an experiment myself with the hashtags trick. That's something I missed.
Thank you for sharing this. I will try this out and do my own experiment.
Fascinating insight here. Your right, correlation does not equal causation, however, logically it seems like a solid strategy to follow.
My first impression was that this was going to be an 'ad' article. Turns out I was pleasantly incorrect. The idea isn't original (PWaT), but the implementation so simple that it makes the difference.
As I'm sure this was written a while ago, I wonder if the effect would be notably different without the Twitter firehose.
I'd be interested to see the change without the twitter firehose as well.
Thanks for reading.
The experiment was conducted June 24th-June 26th, and report was written immediately after. (The Twitter firehose deal expired between the time when I wrote the post and the time it was published.)
Matt, if you or anyone else wants to do a repeat experiment now that the firehose is off, we'd be happy to publish an update.
That's a damn fast YouMoz publishing timeline!
Thanks for sharing your results - PWaT is something I've been thinking of using too. I'll be honest, I had thought it might be a good way to increase Twitter followers, so I'm really glad I read this first!
Great idea, thanks for sharing this.
thank you, I got the free cheatsheet & tweeted. I forward this article to my friends as well.
so u guys should be getting some action on twitter! :-)
Would be interesting to run the same test now after the Twitter/Google deal has ended.
I think we may see some very different results.
Kind Regards,
James Norquay.
Hi James,
I think I'm going to look into doing another experiment - the market demand seems to be present.
What sort of measurements would you need to see to make you feel it was valid?
thanks
matt
First of all, Good article Matt! I would love to see a legit Twitter test, and I'd love to help promote it if you get it going (I'm @Drosenhaus). If you do though, I think it would be great to see how Tweets help a page rank for a semi-competitive keyword and not have the request originate from a huge source like SEOmoz. Obviously number of Tweets and ranking before and after and then may be even a month after Tweeting dies down would all be great things to see.
Great article.
'In fact, the page isn't indexed at all in Bing.' - I had a similar experience recently, it took Bing a while (a month or so) to index a new site I set up, and even the homepage took several weeks to be indexed even though Google picked up on it within a couple of days. - Jenni
This concept is exactly why we created GoPromoIt - a product that is quiet similar in concept but since many products are not as simple we set it up to be an incentive based system instead. So, users may tweet about a product and save a percentage on the purchase. It works more like a twitter badge or facebook like button. Please check it out for ecommerce it has been quiet successful. Users can also customize the message, making it more valuable...
(edited to remove link)