Want More Clicks? Tweet Less





Tweet Much? Don’t Expect a High CTR. New data I’ve been working on seems to indicate that the more frequently you Tweet links, the fewer clicks you’ll get.

I’ve been working towards a statistical model of how an individual makes a decision to ReTweet a specific Tweet and in that process, I came across an interesting problem: before someone ReTweets something, they have to notice it. If you’re anything like me, you’re only able to actually read a small percentage of the total activity in your friend’s timeline, which means that very few of the Tweets I’m technically “exposed” to ever even have the chance of being ReTweeted.

As a measure of “attention,” I started looking into click-through data. The wonderful thing about bit.ly is that it has an API that allows anyone to view the stats on any bit.ly link. I grabbed as many of the bit.ly-containing Tweets of several of the most followed and link-heavy Twitter accounts as the Twitter API allows (it imposes a limit of 3,200 total Tweets accessible per user) and the number of clicks each link had gotten. For the time of each Tweet, I also pulled the number of followers that account had and calculated a followers-to-clicks conversion rate. I’ll call this rate CTR for simplicity’s sake. I was able to get this information for about 2000 Tweets. It is important to note that ReTweets of a bit.ly containing Tweet (if the ReTweeter does not change the link) also count toward the total number of clicks, so it is possible in some cases for a link to have a CTR of over 100%.

Digging into this data, I started to notice an interesting trend: the higher the number of links an account Tweets in a given timeframe, the lower the CTR on each individual link. If you want your Tweet to get noticed and ReTweeted, you should slow down your posting rate.

First, I looked at this data hourly, by graphing the CTR of Tweets over the number of other Tweets posted in the same hour. The first graph below shows individual lines for each account measured; the second graph shows an average for all those accounts.

Then I looked at the numbers by day. The CTR fall-off in these graphs seems to be slower than those above, but the trend is still prominent.

I’ve got a bunch more stats and analysis to run on this dataset to isolate some factors that lead to increased CTR, and therefore increased attention. I’d also love your feedback on data points you’d like to see.

If you liked this post, don't forget to subscribe to my RSS feed or my email newsletter so you never miss the science.

{ 39 comments }

JayFleischman October 21, 2009 at 6:23 am

Excellent thoughts, Dan. This makes total sense – think of it as “click fatigue.” Take it offline to a conversation with a friend. The friend comes to you once with a recommendation about a book that's, “the best book ever – you've got to read it.” So you do.

Then the friend starts feeding you recommendations fast and furious. At some point your brain screams, “Enough already! I've had enough!”

In the blogosphere I've noticed the same thing. One of my blogs gets a ton of traffic in our niche, and for a time we (it's a group effort) were posting 10-15 times each day. For awhile our subscriber numbers shot up, but eventually people started unsubscribing just to get out from under the onslaught of information. It was only when we went down to 2-3 times per day that our numbers started to stabilize. I suspect that once per day would be better, but it would be difficult to accomplish for a variety of reasons.

UlysseH October 21, 2009 at 8:42 am

What's funny is that the guy from whom I got this link seems to be spending his life tweeting. It's like 70 tweets per minute. I assumed he couldn't actually read the articles he tweeted, because the more you tweet, the less you work.
Now I don't assume it, I know it.

Eric Pratum October 21, 2009 at 8:55 am

I am curious to see how phrasing, tweet length, relation or timeliness to trends, time of day, day of week, and follower demographic affects these numbers. I know that Twitter makes it difficult, if not impossible, to zero in on some of this, but any data points you're able to pull together will paint a full picture (at least for me) that can be used for planning content calendars and whatnot.

To address this content specifically, I believe it mirrors what I have seen…both for personal and professional accounts. The more links tweeted, the higher total number of clicks I tend to see, but in general, the marginal gain in total clicks diminishes dramatically for every additional link tweeted. Actually, that brings to mind an interesting question…I wonder if it would be possible to get a good idea of the ideal number of tweets per follower in order to maximize clicks/link.

Maria Reyes-McDavis October 21, 2009 at 8:56 am

Great data Dan and like previous commenter, Jay, it makes total sense to me. The more you tweet with links, the more you slam people (Twitter and blogging as he mentioned) the less important or of value you're information seems. Tweet when you really have something of value or something to say… Love the work you're doing to shed light on Twitter!

Gerrit Eicker October 21, 2009 at 9:07 am

5 seems to be the “magic number” on the Web. ;)

Ricky Panek October 21, 2009 at 10:31 am

This is a fantastic post. One of my super fabulous SEO friends, @jennita, suggested I comment some of my discussion points on your post. I believe your statistical model could be improved by considering the weight of a tweeted link. The weight of a link could be calculated by determining the number of people followed by your followers. Some of your followers might only be following 10 people and some might be following 1000 people. The person who follows 10 people will have greater attention to your tweeted link than the person who follows 1000 people. One could apply more statistical framing for this link weight to get an average link weight or groupings of weights. Grouping the weights would be more clear when considering Eric Pratum's point of determining more demographics for your followers.

Otto October 21, 2009 at 11:02 am

I would have thought this was obvious. If all your twitter account is doing is posting links, then I'm not going to look at any of them. Why? Because as far as I'm concerned, you're spam.

Twitter is about interactivity. The way most people use twitter is simple. When somebody links to their twitter account, and I follow it, then in about 5 seconds I've either blocked that account (using the handy “block user” link) or not.

One look at your latest tweets tells me whether or not to block you. If there's no replies to anybody, if there's no interaction, if there's nothing but links all back to your site, then you're blocked.

Everybody I know who uses Twitter uses it in pretty much that exact same way. Twitter is about having a conversation, not simply advertising your own bullshit.

Eric Pratum October 21, 2009 at 11:27 am

Hmm…interesting point about weighting. Honestly, the skills to do something like that are a bit beyond me, but I would be interested to see if someone was able to put that together.

britneybennett October 21, 2009 at 11:40 am

Without question, with Twitter, less is more. People just get tired of endless tweets that really have nothing to say and after a while, they'll tune everything out.

Ricky Panek October 21, 2009 at 12:09 pm

I didn't even think of this until now but another measure would be the frequency of tweets made by your followers. If your followers tweet often, then you could deduce that those followers view their timelines about just as frequently (there really isn't a strong case for implying reading tweets before or after writing tweets, but I'm guessing they could be related). Then, apply your TOD, TOW metric to see when your followers tweet most often and that would give you a good window of opportunity to push a tweet through that contains a link for click-through effectiveness.

Paul O'Mahony (Cork) October 21, 2009 at 12:36 pm

Great analysis. Thanks v much. I take it personally and better now review my habits. “Do less mean more” is one of the sorta mantras by which I aim to live. But I've been led astray by enthusiasm for tweeting.
When I RT, I think of it as an act of adding one more flicker of paint to a convas – a tiny statement that tunes the tone of self put out there.
Think I better cut down on the commenting too. It's not always an advantage to love writing.

Eric Pratum October 21, 2009 at 12:53 pm

Wow. Yeah, it seems intuitive, but I hadn't thought to make that explicit connection before…now that you've suggested it, time for me to get on the research *mental cranks and gears spinning*

Susan October 21, 2009 at 6:10 pm

Great work, Dan! When I mentioned in a client meeting that excessive tweeting may put you in danger of being ignored by your audience there was stunned silence, disbelief, then head shaking at my” ignorance.” Thank you for putting numbers to what I suspected to be true.

Jillian Ada Burrows October 21, 2009 at 6:55 pm

A power law! As with Zipf's law, the symbols (words, links, etc.) near the knee of the plot carry the most information.

ToniAnicic October 22, 2009 at 2:48 am

Nothing unexpected but it's nice to see some actual statistics, nice work!

adimoculescu October 22, 2009 at 2:52 am

Great article Dan, so what is the magic number?

ajbarnett October 22, 2009 at 2:54 am

Interesting stuff. Well researched.

I need to come to terms with it though, because your evidence conflicts with the philosophy of those gurus who advocate a heavy presence in order to get noticed.

To back your theory and research, I've found that I don't get as many RT's as I used to – I'd put it down to the fact that I'm quite new to Twitter and people were inquisitive about my tweets – but it may also be that I post more, AND present more links – which is what you're saying.

As an aside – I've tried to sign up to Bit.ly and it continually rejects. Has anyone else had this problem.

brucemi October 22, 2009 at 3:40 am

This is EASILY the most useful Twitter-related article (due to its concise stats) re: How to get ur Tweets read/RT ~ slow down the frequency/hr of ur Tweets to five (max).

And, the reason why I have recently Unfollowed ppl is when they Tweet more than 10 times/hour!

Wonderful & practical research…thanks so much, Dan!

bruce_mi

brucemi October 22, 2009 at 3:43 am

Dan ~

This is EASILY the most useful Twitter-related article that I've read (just passed 6 months as Twitterer)!

Wonderful & practical research! Confirmed what I've been doing lately: Unfollowing Twppl who tweet more than 10 times/hour!

Great work ~ keep 'em coming,

bruce_mi

Edin Shaba October 22, 2009 at 9:24 am

Excellent Work!
It's funny, i was trying to explain this to a client who is “in the business for 10 years” and 'knows better', that tweet fatigue is real and that quality link/content is King.

kathrynrobinson October 22, 2009 at 12:30 pm

I can personally agree with this research because out of all the links posted by those I'm following, I only click on a small few of them. Yet, I'm guilty of posting a lot of links too. Must rethink that now after reading this article. Thanks for sharing.

Camilo Olea October 23, 2009 at 2:45 pm

Thanks Dan! Information gold.

So, one thing is not clear to me: I understand people dont like accounts that “overtweet”, so, one should “slow down” his tweeting in order to avoid this. What I dont understand is, what about trying to make your tweet get noticed among the endless, constant, huge stream?

I personally do this: If I want a tweet to get noticed over the day, I program my tweets in Hoot Suite to be sent 3 TIMES. Morning, noon and afternoon. Is this considered a good practice?

Thanks again and cheers from Cancun! =)

Camilo

hrmargo October 24, 2009 at 4:19 pm

Does this count when you are tweeting while viewing your favorite show with viewers around the country? Or, after hours or on the weekend when you are having fun with your friends?

Coach Chris Berg October 28, 2009 at 12:15 am

Dan, this is AMAZING info…I dont know how you compile it and thank you for sharing!

susanbreidenbach October 27, 2009 at 7:36 pm

Are you counting ALL links, or just the ones people are using to market/sell something? For example, if I tweet a link to an interesting article I think people might like, and it's not my article and I have nothing to gain beyond building relationships by tweeting useful information, that counts the same as if I tweeted a link to a squeeze page or e-commerce site or affiliate marketing partner? Also, some people seem to tweet nothing BUT links. Does your data show those people are more likely to be ignored? Thanks!

N - Designs October 29, 2009 at 9:36 am

great and interesting article. thanks.

BetterBizIdeas October 31, 2009 at 6:31 am

This basically confirms the law of scarcity from a marketing perspective. Everyone is “marketing” their tweets (ie. trying to get opened/clicked or “see me” as Seth Godin has said). Intutively I was hoping that less tweets per hour would equal a higher clickthrough rate. It leads to better quality, UNLESS you are forwarding great info. that others have seen. The problem is, when you forward a bunch of great tweets onto your followers one tweet then competes with the next one for your Tweeps time :)

One completely opposite perspective is that of internet marketers. Pure “tactical” marketers want volume of clickthroughs so they play the game of numbers and increase their tweets. They know that the lower clickthrough rate won't offset the sheer “attention” they are garnering. It is why some of the biggest spammers were utilizing @Tweetlater to post every 6 to 20 minutes and they were asked by Twitter to remove that functionality.

This analysis clearly illustrates that people seeing to become “authority” figures in their respective niches would be better served to tweet less often :)

Nice work Dan!
Dan Ross
@BetterBizIdeas

Robert Phillips November 10, 2009 at 7:30 am

Till now, I was doing exactly opposite to what you have mentioned here. That is, I was giving more tweets and getting few clicks. You opened my eyes by posting this amazing article about twitter. I am very thankful to you for letting me to know about this most important topic. You must have done some good research to write this blog post. Thanks once again.

Danijel Maricic November 23, 2009 at 6:46 am

hey Jay,
today i searched about digital media on twitter and i saw that almost evry tweet contains LINK.
how can i decide which link or tweet is important for me?
There is bunch of links on twitter, most of them are just commercials of brands and products.
then i think of your line “Enough already! I've had enough!” !!!

Danijel Maricic November 23, 2009 at 8:19 am

are you tweeting 3 times the same msg?
i think that it's not good thing. If you want to get noticed, carefully choose ppl that you want to follow. I think it's better to have less followers/following with same interests then thousands of followers/following that are not interested in what you want to say.

that's my opinion, maybe i'm wrong :)

kierann November 27, 2009 at 3:57 am

Your business strategy is very good, and I think changes in Twitter will provide a well concept to the modern society. I like it and must try to know more about it.
… Thanks for your great contribution….

elainefogel November 28, 2009 at 2:37 pm

So glad to discover that my lack of frequent tweeting is a GOOD thing. :)

elainefogel November 28, 2009 at 10:37 pm

So glad to discover that my lack of frequent tweeting is a GOOD thing. :)

vrosenman January 4, 2010 at 2:26 pm

Dan, this is partially true. But cumulatively the optimum is somewhere around 5-10 posts per day. Assume you send a single post and get 30%. Good your CTR is 0.3. Now if you send 5 posts your average CTR is 0.45 (15% of 5). This rate seems to stay more or less the same till 10 posts: 0.45 – 0.5 CTR. After then it declines.

Christopher F Clark January 25, 2010 at 7:05 pm

I know this is a very late comment, but some of your stuff holds true, but some of it needs more analysis. True: on days I tweet alot I get less CTR. However, on days after I tweet a lot I get more CTR. If I cut my total tweeting down, I get better CTR on each day, but tweeting more than 5 times per day seems to be necessary to get people to see me. Worth noting by tweeting, I am actually talking about RT'g, as almost all of what I tweet is RTs of stuff that I am passing on.

Nurul Chowdhury February 12, 2010 at 10:52 pm

Dan, thanks for this useful information. When I started using twitter, I was told that it was a good place to promote your business. As my followers started getting bigger, I started doing some tracking myself and found that there it really does not work anymore.

I have stopped now and do it once in a while when there is a big occassion to tell my twitter friends. And the action is much better than random tweeting. I prefer bringing them to my blog for informative information and then they can wander around on my blog.

Yes I am a humble follower of your knowledge sharing.

chyanne April 27, 2010 at 2:37 am

thank you! let's spread the word! this fact alone has been my huge reluctance to join twitter – i don't want to have to sort through people's garbage. if you tweet more than 5 times a day i'm going to stop following you because you clutter.

Guest123 August 18, 2010 at 3:39 am

I disagree. Including a link in a tweet is no more spam than a tweet without a link. In fact, I consider tweets without links often less worth reading… if you’re going to make a claim I’d prefer you back it up. Also, since Twitter is sort of a stream of info rather than something you expect everyone will see, I am not sure CTR the way you describe it is the most appropriate way to guage ‘success’. I’d argue total clicks per day would be more important (keeping in mind the follow/unfollows as well).

DigitalReasoning September 3, 2010 at 3:04 pm

Dan,
Great info but a couple of questions:
1) does this mean that just the “existence” of a link impacts the results or do you mean when someone simply puts a link with no additional text/context for the reader?
2) would it be fair to extrapolate this info to all Twitter messages regardless of the existence of a link? i.e. if you chat so much that you get a good Klout score you might have a large percentage of our messages effectively missed.

again, thanks for the insight.

{ 25 trackbacks }