SocialThe World Welcomes 2014 on Twitter, Reveals Usage Patterns

The World Welcomes 2014 on Twitter, Reveals Usage Patterns

This analysis of how New Year's transpired on Twitter around the globe as each region welcomed 2014, examining popular keywords, as well as elements of tweets (retweets, links, hashtags, pictures). Here are six key takeaways about Twitter usage.

New Year’s Eve is possibly the only event that is celebrated all around the world regardless of religion, culture, and race, but with one important characteristic: it’s based on time zone.

While religious events, like Christmas, Rosh Hashanah, or Ramadan have a time-zone importance (they are celebrated at a specific time every year), they lack the global all-encompassing nature since they are specific to a religion.

Global events like the World Cup, the Olympic Games, or even the Oscars, might have the global importance but they aren’t time-zone specific; they happen at a given point in time regardless of the rest of the world’s time zone.

New Year’s Eve is the one event that has both characteristics – it’s global (or as global as it gets) and it’s time zone specific. And why is that important? Well, because it creates an almost 20-hour celebration craze on social media that allows us to see how one, scheduled event impacts the world in different ways on social media.

To look into those usage patterns, I analyzed several aspects of the celebration as it transpired on Twitter around the globe as each region welcomed 2014. I analyzed volume of keyword mentions as well as some elements of tweets (like the existence of an image, a link or a hashtag) and charted them across the time zone of the regions that celebrated the New Year’s Eve.

In all the charts below, the X-Axis represents midnight in each time zone in GMT (Greenwich Mean Time). For example, the data that corresponds with GMT +0 is the activity on Twitter when midnight “hit” London.

“Happy New Year”

Happy New Year by Time Zones

The term “Happy New Year” trended heavily on Twitter throughout the New Year’s Eve celebrations. As midnight hit every time zone, the term spiked, indicating the level of social media usage in those regions.

As the graph shows, when midnight struck in China and Japan, Twitter exploded with more than 1.6 million tweets that included the term “Happy New Year”. What’s even more impressive is that the vast majority of these tweets weren’t in English but yet the English term “Happy New Year,” trended.

I next analyzed how the term “Happy New Year” trended in other languages: Russian, Chinese, Spanish, and Arabic.

Happy New Year in Russian Chinese and Spanish

These trends align perfectly with the regions that celebrated New Year’s Eve and their most common spoken language. The Spanish term spiked when Spain celebrated New Year’s Eve and then again as midnight arrived in South America. The Russian term spiked, well, in Russia, and the Chinese in China.

Since the Arabic term showed low volume, we graphed it by itself to look for any interesting trends.

Happy New Year in Arabic

While the volume is low, the term trended in the regions you would expect it to trend: the Middle East and North Africa (Egypt, Lebanon, Jordan, Iraq). The big spike in GMT +9 can only be explained by usage in Indonesia.

Other Popular Keywords

Next, I looked at some of the most popular terms and how they trended on Twitter over the 20-hour of global New Year’s celebration.

Midnight and Kiss Trend on Twitter by Hour

Unlike other terms that trended in spikes over the 20 hours, the term “kiss” spiked mostly on the East Coast. My only explanation to that is the celebration in NYC and the kiss-on-midnight tradition as the ball drops in Times Square.

Countdown Trend on Twitter by Hour

I expected the term “countdown” to spike on the East Coast – just like the term “kiss” – but interestingly enough it peaked in Asia and Europe an hour (or less) before midnight reached China, Japan, and London.

NYE Trend on Twitter by Hour

The hashtag #NYE didn’t trend on Twitter. While it showed high volumes by itself, it didn’t take off as a trended term and showed minimal peaks throughout the night, eventually dying off when midnight reached the West Coast.

Twitter Keyword Volume by Hour

Somewhat expected, the term “2014” had the highest volume on Twitter, more than any other term. Fittingly, the term “2013” slowly disappeared as the actual year ended.

The term “2014” followed the similar trend the term “Happy New Year” followed, revealing similar usage patterns on Twitter: Asia, led by China and Japan, show strong usage followed by the East Coast and the UK region.

Tweet Elements

Lastly, I looked at the most popular tweet elements (retweets, links, hashtags, and pictures) and how they trended by time zone.

Tweet Elements Trend on Twitter by Hour

On average, retweets (RTs) were included in 43 percent of tweets showing the conversational nature of Twitter, especially in social events like New Year’s Eve. Links, included in almost 25 percent of tweets, followed by hashtags (19 percent), and embedded media (15 percent).

Overall, it seems like the usage patterns of these tweet elements didn’t vary much by region, although the peaks in RTs in Asia and in links on the West Coast are interesting.

The peak in RTs in Asia can be explained by the sheer volume of tweets that drove more RTs, However, you could argue that the RTs were what contributed to the overall volume.

The peak in percentage of links that happened on the West Coast can be explained by the fact that at that point, the celebrations all over the world were over and more content was produced that could be shared on Twitter, therefore the increase in shared links as percentage of total tweets.

Embedded Media Pictures vs Video

Embedded media are pictures and videos that show up right in a tweet. Those would include pictures from TwitPic, images uploaded directly to a tweet, Vine videos, and YouTube videos.

Looking at the distribution of those you can see that pictures constitute a whopping 92 percent of all embedded media vs. only 8 percent of videos. In addition, images uploaded directly to a tweet (as represented by a URL that starts with “pic.twitter”) are the preferred method of embedding media into tweets.

Distribution of Links to Other Social Networks

Lastly, we looked at the distribution of links from tweets to other social networks during the 20-hour New Year’s Eve event.

Expectedly, we found that Instagram, which has a simple integration with Twitter, ruled the outbound links. Second was Tumblr and third was Path, both social networks enable sharing on Twitter and make it easy enough to post and share directly on Twitter. Pinterest, Facebook, LinkedIn, and Google+ had some percentage of links point to them, but compared to the overall content shared on Twitter, those were negligible.

Key Takeaways

What can we learn from this data, other than the fact that people love to party and share it on Twitter? Here are a few takeaways:

  • Twitter usage in Asia is astronomical.
  • Even in non-English speaking countries, English terms can still trend (see “countdown” and “Happy New Year”).
  • Twitter is becoming the de facto universal online chat room where people actually have short conversations on (see percentage of RTs in tweets).
  • Make it easy to share stuff on Twitter, and people will (see Instagram, Tumblr, Path).
  • Twitter is becoming more and more about sharing media on Twitter (as opposed to using Twitter to link to media off Twitter) and having real interactions on Twitter. If you’re marketing on Twitter, develop content that can stay on Twitter.
  • Not on Twitter? What are you waiting for?

Any other observations? Share them here.

Resources

The 2023 B2B Superpowers Index

whitepaper | Analytics The 2023 B2B Superpowers Index

8m
Data Analytics in Marketing

whitepaper | Analytics Data Analytics in Marketing

10m
The Third-Party Data Deprecation Playbook

whitepaper | Digital Marketing The Third-Party Data Deprecation Playbook

1y
Utilizing Email To Stop Fraud-eCommerce Client Fraud Case Study

whitepaper | Digital Marketing Utilizing Email To Stop Fraud-eCommerce Client Fraud Case Study

1y