Time Is On Your Side

Did you see those silly cats on Tumblr, that breaking news on Twitter, and those photos of your friend’s kids on Facebook? Different social networks have their own distinct personalities. Bitly links are shared across all social networking services, giving us a unique viewpoint on how these networks differ.

We track metrics like the main type of content being shared on a network, the geographic locations of the people sharing and viewing the content, and how the popularity of the network has risen and fallen compared to other networks. Studying the differences between these networks leads us to interesting insights, for example, did you know that the half-life of a link on Twitter is 2.8 hours?

Recently weʼve been exploring how content propagates (or “goes viral”) through social networks, particularly how the day and time something is posted affects the eventual amount of attention it will receive.

Note: All the plots are based on EST. You will see day of the week, starting with Monday, on the Y axis, and hour of the day, starting with midnight, along the X axis. For the first plot in each section, the darker the blue block, the more traffic on average links posted during that hour received in the following 24 hour period of time. White blocks, show when links got less traffic. For the second plot, the darker blue represents when the site is most active, which we calculate based on number of clicks on Bitly links coming from these social platforms.

Links posted from 1pm to 4pm result in the highest average click throughs. The peak time of the week was on Wednesday at 3pm. Links posted after 8pm and before 8am will have more difficulty achieving high amounts of attention. As with Twitter, avoid posting on the weekends.

Facebook traffic peeks mid-week, 1 to 3pm. While traffic starts to increase around 9am, one would be wise to wait to post until 11am. Traffic from Facebook fades after 4pm. Despite similar traffic counts at 8pm and 7pm, posting at 7pm will result in more clicks on average than posting at 8pm.

Tumblr

Tumblr likes to party! This network shows a drastically different pattern of usage from Facebook and Twitter. One should wait until at least 4pm to post. Also postings after 7pm on average receive more clicks over 24 hours than content posted mid-day during the week. Friday evening, a no-man’s land on other platforms, is an optimal time to post on Tumblr.

Bitly traffic from Tumblr peaks between 7pm and 10pm on Monday and Tuesday, with similar traffic on Sunday.

Conclusion

It’s easy to see that just like your neighborhood restaurants, each social network has its own culture and behavior patterns. By understanding the simple characteristics of each social network, you can publish your content at exactly the right time for it to reach the maximum number of people.

What’s a true click? Humans vs. bots.

A common question we get from bitly users is, “Why are my bitly click counts different from those reported by other platforms?”

Click counts can appear to be substantially higher when bots, crawlers and other automated systems are included in “click” counts. These systems are both ubiquitous and largely invisible on the web, and can account for the majority of click-like events on some links. That’s why our data science team spends lots of time working on classifiers that filter out clicks from non-human agents such as bots and crawlers. What we record on the info page of a link are what we call true clicks which are counted by bitly only when a human clicks on the bitly link.

This heavy filtering allows for a more accurate understanding of how many times your content was actually viewed. When you look at your click count on a link, you know that bitly clicks represent humans who actively engaged with your content. On the social web, identifying something as basic as a human click can require some heavy lifting. We make a big investment in data science and engineering, so that we create a system that counts clicks accurately and makes the analytics you see more valuable and insightful.

Tracking Links Across Social Networks

The social web can be a big, complicated place. Once your content is published, your audience will circulate it in ways you can neither control nor predict; a link shared to Twitter will be copied and pasted into Facebook, AIM, email, LinkedIn, and beyond. Single-platform analytics don’t see all this activity, but bitly allows you to track and optimize your efforts across every social network.

For example, we can use bitly data to visualize how traffic jumps from one large social network to another. The above visualization shows click traffic to a small web publisher’s bitly link from Twitter and Facebook over a period of 60 minutes. This link gets its initial traffic from Facebook, and then sees a sudden spike in click traffic from Twitter about 30 minutes after it first receives traffic from Facebook. Because bitly measures real human clicks consistently across every platform, we can provide a unique window onto how links are circulated from one social network to another.

Above is a traffic visualization for a link shared in a Promoted Tweet by a large daily deal site. We can visualize click traffic across social networks to show that, after the link was initially shared on Twitter, it spread organically to Facebook, where it received substantially more click traffic. Because bitly’s analytics are calculated in realtime, we can identify a viral lift like this as soon as it occurs, wherever it occurs.

Announcing Bitly Enterprise 2.0

Facebook has 800 million active users. Twitter has 300 million.  For many businesses, social media has evolved from the side project of a single marketing intern to a complex strategy executed by a large team.

As the social web has grown, bitly has grown right along with it. Half of the Fortune 500 companies active on social media use bitly, as do 75% of the world’s largest media companies. bitly now hosts 20,000 whitelabel domains, and our Enterprise product is used to track social media traffic to over 10,000 websites. bitly’s API is one of the most-used on the social web, powering social sharing for millions of users across every major social network, email, IM, and SMS.

With an ever-evolving social web comes the need for more robust and comprehensive social media tools. Too many social media teams are still using 10 different dashboards to measure different engagement metrics across different platforms. Today, a year after rolling out our initial Enterprise offering, we are launching our first point release of Bitly Enterprise. We’ve incorporated feedback from our clients to create a tool that social media teams of any size can use to share content, discover actionable insights, and optimize messaging to maximize social media ROI.

Access
Bitly Composer

When we asked our Enterprise clients the biggest challenge they face with using social media tools, the most common response we heard was “dashboard fatigue.” As social media teams grow in size, it is increasingly important to have a tool that can be used by many individuals within an organization both to share content and to measure the performance of that content across the social web.

To help streamline account management and social media sharing workflow, we’ve built these functions right into the Bitly Enterprise 2.0 dashboard. You can set up unique permissions for each member of your team, manage multiple Twitter and Facebook accounts (including Facebook pages), and schedule posts for future dates and times.

Insight
Reputation Monitoring

We recently announced a reputation monitoring service built on top of our new social search platform. Using bitly’s reputation monitoring service is now as easy as entering a term you’d like to monitor into your Bitly Enterprise dashboard. For each term added, we track the volume and sentiment of popular content that is travelling through the entire bitly ecosystem. We even backfill 7 days of data for each new term you add, so you can begin tracking the conversation around a particular phrase right away.

With this release, we’ve made reputation monitoring more flexible, so you can further refine your search by location, domain, or the social platform (such as Facebook or Twitter) that is driving click traffic. For example, you can set up separate monitors for the word “pizza” in New York and California, to discover that New Yorkers are reading about Brooklyn’s vaunted DiFara being closed by the health department, while Californians are reacting in outrage outrage to the US Congress declaring pizza sauce a “vegetable.” With our improved reputation monitoring, you can discover conversations your audience is having right now, and laser-target your social media strategy.

Advice
Message Optimization

From day one, the primary goal of Bitly Enterprise has been to provide our users with realtime, actionable intelligence to optimize social media. While other social media tools are built around vague, platform-specific metrics, the intelligence provided by Bitly Enterprise is based on real clicks by real people, measured across multiple social networks and platforms. We have improved the core functionality of the original Bitly Enterprise dashboard to better show you the top-performing content that your users are sharing across the social web, and how it may be different from the content being shared by your social media team. By identifying potentially underutilized content, you can capitalize on the momentum being generated by your audience, independent of your own promotional efforts.

We have also incorporated built-in timing optimization, which actually listens back to your audience to see what they are talking about right now, and how many of them are actively engaged with social media. When the moment is right, Bitly Enterprise can use this information to automatically send out the content that seems most relevant to your audience at the very second it is most likely to drive traffic, whether or not you happen to be in front of your computer.

We’ll be rolling out Bitly Enterprise 2.0 to our existing customers in the coming months. If you have any questions, or want to try out any of these new features, please drop us a line at enterprise@bitly.com.

bitly and Verisign

One of our investors brought his teenage daughter by the office the other day. She was puzzled.

“Bitly has an office?” she asked her father. “I thought it was, you know, just part of the Internet.”

We took that as a great compliment — over the last three years, bitly links really have become a ubiquitous part of the web.  A recent Microsoft Research report even claims that short links can account for as much as 1% of the new URLs created on any given day. 

With that kind of scale comes a great responsibility in terms of technical reliability and redundancy. Which is why we’re pleased to announce a new agreement with Verisign, which operates two of the Internet’s root nameservers and much of the web’s DNS infrastructure. If there’s a single company that qualifies as the steward of the Internet, it’s Verisign.

Verisign and bitly already work in tandem nearly every time a short URL is clicked. On any given day, bitly translates hundreds of millions of short URLs into standard web addresses; last month alone, we handled 8 billion such redirects. Verisign takes those long URLs and translates them into IP addresses, resolving over a trillion monthly DNS queries, including every URL hosted on a .com, .net or .gov domain. 

These two steps constitute the core infrastructure of the social web, and bitly’s relationship with Verisign aims to make them completely reliable and blazingly fast. 

Beginning this fall, bitly’s primary data center will be hosted on Verisign’s global infrastructure. That’s saying a lot, since Verisign’s infrastructure has maintained the .com namespace for more than a decade with zero percent downtime.  

Verisign’s architecture has been integral to the growth and stability of the Internet at large, and we could not be more excited to work with them. Scientists at both companies are already poring over volumes of DNS resolution data — data that will help us answer fundamental (and fundamentally awesome) questions like: “what actually are the most popular websites on the Internet?” and “just how big is the Internet, anyhow?” 

Stay tuned, as we’ll be sharing our findings here.

Social Search

Today we’re pleased to announce the beta launch of bitly’s search platform and reputation monitoring service.

bitly shortens 80 million URLs every day, give or take a few. Now, with our new search technology, we’re crawling and classifying every URL we shorten to create an index of the most “viral” content on the web — content that’s broadly distributed, frequently-clicked, and trending at a high velocity.

Here’s an example of bitly search results for the query “onstar”:

You can see that there’s been some controversy about OnStar tracking its customers, and its former customers. The top results include news stories about the company, and a blog post by a single individual, a forensics scientist in Ohio named Jonathan Zdiarski, at zdiarksi.com. 

A Google search for “onstar,” by contrast, returns the company’s official website and its wikipedia page —  results based on Google’s pagerank algorithm, which prioritizes the pages which are linked to by the most authoritative sites on the web. Not zdiarksi.com.

This is the challenge of the realtime search space — many of the pages are so new, so fresh, that they don’t have any pagerank. A personal blog post isn’t authoritative in the way that the New York Times or Wired magazine is authoritative, but stories don’t find their way into Wired unless people like Jonathan Zdiarski speak out. Increasingly, they do, and often they reach a broad audience on social media before more conventionally-authoritative newsgatherers amplify their messages. 

So instead of pagerank we’re using a different filter — for any given search query, we display the stories that we predict will get the most attention over the next 24 hours. Then we use bitly’s analytics to refine our predictions in realtime. Our search technology is based on the the most valuable measure of engagement: the click.  

The first product we’ve built on top of this technology is a reputation monitoring service. We added a layer of sentiment analysis to our search results  and built an alert system that lets our customers know what people are saying about products, brands, or about any topic on social media. Unlike a typical clipping service, which lets you know the things that people have already said about you on Facebook, this is an early-warning system, designed to alert you in realtime to swings in volume and sentiment related to specific keywords. 

In addition to an email alert, there is a dashboard for all the keywords you’re tracking:

Reputation monitoring will be rolling out to our beta testers and current bitly Enterprise users over the next two weeks.  It’s just one part of our next bitly Enterprise release — Enterprise 2.0. More on that soon. Meanwhile, if you’d like to participate in the beta test for reputation monitoring, please contact beta@bitly.com. For more information about Enterprise, or to schedule a product demo, please go to http://bit.ly/EnterpriseSignup.

We’re looking forward to your feedback!