Twitter: From ‘chirps from birds’ to shouts from the White House.
On December 6, 2016 then President-Elect Trump sent a tweet about the costs of the future Air Force One, abruptly ending in “Cancel Order!” These 137 characters sent Boeing stocks tumbling in pre-market trades. Although the stock recovered later that day, it is remarkable that any singular person can impact world markets and dominate global headlines through one social media website — twitter — a network that got its name, in part, from “short bursts of inconsequential information” like “chirps from a bird”
But how did this page rise from, essentially, a semi-public group text message to a global network of over 319 million monthly users? What is it about the site that has allowed it to transcend the digital space to impact culture and politics? And what does the future hold for Twitter, or, more importantly, what does twitter hold that might shape the future?
Texts between friends: Content
Twitter began in 2006 during the pre-smart-phone-days of SMS text messaging. Founder Jack Dorsey had the idea of sharing the short messages with a larger group of friends and, for the first few years, it was exactly that — a service to share short messages between circles of friends and acquaintances.
At the time, SMS text messages were limited to 140 characters and Twitter maintains that as their bread-and-butter. But saying that a 140 character short text field is the core data of Twitter would not capture what allowed it to explosively grow from its early days.
The turning point came in 2007 when Twitter had a substantial presence at SXSW. During the conference, Twitter staff set up two large screens where tweets would be displayed of those attending. Attendees could see what everyone else was up to, maintain an ongoing dialogue, and interact with presenters. But how could that be grown-to-scale beyond manually inputting someones username to an event or group? Even if you were to search for key words, how could you sort through all the relevant tweets that would contain those key words (e.g. south, southwest)?
Anatomy of a Tweet: Pattern
From an organization and taxonomy perspective, the hashtag changed everything. The simple HTML-activated hashtag allows users to do the work of sorting content into self-generated headers. Similar to how Google broke the search engine mold by prioritizing how many times a page is referenced rather than volume of relevant keyword text, hashtags allows a user’s Tweet, for the first time, to have the capability to stretch beyond their circle of manually added friends and acquaintances. The tool is also powerful since users can add more than one hashtag allowing their tweet to “be filed” under more than one header. The trend of adding more than one hashtag rose in 2009. By January 2011, there were an average of 1.3 hashtags in every tweet.
A second tweet entity, popular in the early days of the website, is embedding a URL as a link. While this is a driver for content generation, it doesn’t do much in the area of organization and taxonomy. The total amount of tweets that contained a URL peaked in 2009 at around 30% and then steadily declined. While the volume of embedded links is relatively low, the use of URL’s do enhance knowledge and perceived trust on the meta-data level by being able to link content that goes into much greater breadth and depth.
Another key component within those 140 characters is the “mention”. The use of a “@” symbol followed by someones twitter name creates a hyperlink to their account and, more importantly, shifts a tweet from a one-way announcement to a dynamic conversation. By 2011, over 50% of tweets contained a mention which was another important tipping-point of the site becoming more “conversational” in feel. Compare that to the initial home page which prompted users by asking “what are you doing?”
Twitter encourages the use of mentions by having a “reply” button at the bottom of every tweet. This is next to two other buttons — the “favorite” star (comparable to a “like” on facebook) and the re-tweet, added in 2014. These features serve three important purposes. First, it helps Twitter identify patterns within its content and differentiate that from “pointless babble”. Second, it provides feedback for users to engage them with the site and, thus, generate further content. Finally, the “favorite” and “retweet” feature establishes trust as users discern what is important. While the psychology research is mixed, it boils down to a simple perception — more retweets is more credibility.
The “favorite” and “retweet” tools positioned twitter to generate a high volume of content at an incredible speed while hashtags and mentions allowed this content to be organized, searchable, and quickly form knowledge that blur the boundaries between the digital and physical worlds.
Twitter breaks the news: Pattern to Information
As perceived trust in institutions (including the media) declined in the 2000’s, Twitter provided users the ability to share news in a peer-to-peer format. This, combined with the rise of smart phones with cameras, led Twitter to become a galvanizing force in the way news stories break worldwide. Twitter itself acknowledges the Miracle on the Hudson and The Arab Spring as key events in its history due to the sheer speed and volume of content generated around breaking stories— all of which is made searchable, shareable, and discoverable through the HTML-enabled hashtags and mentions.
Engaging users: Navigation
From the foundation organizational core data, Twitter continually tweaks the way users see content.
Twitter’s navigation has always lent itself to feeling “conversational” with the continuous scrolling of tweets in reverse chronological order. Between 2007 and 2009 there were relatively few changes. The horizontal menu bar at the top is for profile management and the navigation as a whole lends itself to promoting volume of tweets and followers rather than organization with the exception of the under-utilized ability to save searches and create lists. User feedback in the form of alerting users as to how others engaged with their tweet indirectly guides them to generate content that cuts across one-way announcements to their friends circle (e.g. more hashtags, more people will see your tweet, more retweets). This feedback takes on different forms as the site has developed.
Coinciding with dramatic growth, Twitter released a new homepage in 2010 . There is a clearer emphasis on trends with a continually updated side panel. The tabs at the top of the feed allow users to sort between passive browsing (Timeline) and active conversations (mentions and retweets). There is better use of the screen space and user thumbnails under the followers/following counters mark the beginning of a slow transition towards more visual content. In 2011, Twitter allowed users to share images and throughout 2016 they invested in other forms of visual content (Vine, Periscope) however the core data for content organization remains the hashtag and the interaction elements (retweet and mentions).
In early 2016, Twitter leveraged its massive amount of content and user information to target specific content to users by moving away from a reverse chronological feed. In its place are highlighted posts from across twitter that are predicted by an algorithm to be of interest based on categories that a user has interacted with. This, combined with linking users to others in their extended network, provides a more “directed” experience — all of which still driven by the core data pillars of hashtags, mentions/retweets, and keywords.
Today, the majority of twitter users come from mobile devices. Similar to the desktop navigation, the default landing page is the algorithm produced feed of likely interests with a horizontal navigation bar at the bottom of the page. There is an emphasis on “discovering” new content. Users now need to make additional taps/clicks to see their chronological feed that used to be the sole navigation of the original page. While new users will see content from broader categories, returning users will see very targeted content based on their past activity. There is a structural emphasis on engaging returning users with notifications regarding their own tweets. This can be extended to other alerts beyond the app itself.
The Future of Twitter
A key turning point in Twitter’s history was their IPO in 2011. As with other publicly traded social media platforms, that raises the question “how can this make money?” Well, the business model plays a clear role what tweets are “suggested” and what a user sees by default verses has to hunt around for. For the future, it is critical that Twitter not abrasively post sponsored content that deviates from the “conversational” and “scroll & discover” feel of the platform. Also, the predictive analytics shouldn’t make users feel as if they’re being “watched.”
Another future consideration of Twitter is galvanizing what responsibility the site has to censor “fake-news” and, more importantly, defining what “fake news” really is. Beyond the clearly fabricated news stories, what responsibility, if any, does Twitter have in reducing the echo-chamber of content that validates our existing our world-view? No matter what this future might look like, it is clear that Donald Trump will want to have his voice included — 140 characters at a time.