Communicators, Take Note: The Deplorable State Of Social Search Is Changing
The rate at which people share information (profound and profane alike) on sites like Facebook, Twitter and YouTube is staggering. 137 million status updates per day appear on Facebook, 230 million tweets are shared per day on Twitter and 72 hours of video are uploaded to YouTube every minute — all numbers which increase daily. No […]
The rate at which people share information (profound and profane alike) on sites like Facebook, Twitter and YouTube is staggering. 137 million status updates per day appear on Facebook, 230 million tweets are shared per day on Twitter and 72 hours of video are uploaded to YouTube every minute — all numbers which increase daily. No doubt, social media is alive and well.
Social media properties have clearly made the processes of creating and sharing content a compelling activity for everyone. Yet the story is altogether different when it comes to retrieving content. Ever tried to search for an old post in Facebook? Pretty difficult if the update was more than a month old. Ever tried to search for tweets older than a week using Twitter search? Mission impossible.
Fortunately, this sorry state of affairs is changing, with significant implications for all those who want to communicate — whether it be about people (e.g. reputation management), places, products or services.
Search Wasn’t Part Of Social Media DNA
Historically, social media sites like Twitter and Facebook focused their resources on real-time information publishing and consumption. The expectation was that users would scan and potentially interact with updates while they were fresh, within minutes or hours of their publication. Enabling the discovery of what was posted in the past just wasn’t part of these companies’ DNA — perhaps not too surprising, as search is complicated and expensive.
The one search area social media companies have generally excelled in has been people search (and email address book matching). Again, not a surprise — the success of these companies depends on the number of active users and the size of an individual’s network within the social media property.
There have been moments of hope, such as when Twitter acquired Summize in 2008. Yet, a full blown Twitter search didn’t emerge; searches were limited to a selected subset of all tweets from the last few days.
In its early years, Facebook allowed users to search on most, if not all, people attributes, including potentially sensitive ones such as religion, political affiliation and sexual orientation. Over time, data privacy concerns kicked in, as did search complexity thanks to the stellar growth of Facebook’s user base. Facebook scaled back people search options. Since Facebook introduced Newsfeed search, it has been limited to updates from the last 30 days.
Even the major search engines have not been very helpful. Microsoft’s Bing began to experiment with social media search in 2009, eventually launching Bing Social in a small number of countries (including the U.S., U.K. and Germany), which limited searches to subset of recent Twitter and Facebook data. In 2012, Bing replaced Bing Social with a social side bar, but again limited availability to just a very few countries.
Also in 2009, Google took the unprecedented step of paying Twitter for the privilege of indexing twitter tweets, only to change its mind once the agreement expired, killing their real-time search in the process.
For those who wanted to search comprehensive archives of social media sites, third party solutions like Topsy have been the best bet. Topsy and its competitors buy social media data directly from social media sites or through distributors such as datasift and gnip.
Social Media Companies Are Reconsidering Search
As online social networks mature, they’re beginning to reevaluate the value of the immense quantities of historical data and other content they’ve accumulated. In January, Facebook announced a completely new search experience, called Facebook Graph Search. Graph search allows users to search for information within Facebook using natural language search queries, e.g. restaurants in Berlin liked by my friends from Berlin. Results are ranked primarily using the relationships between people (friend, follow) and item popularity (likes). (Naturally, reality is slightly more nuanced, and it is a safe bet that not all likes are created equal.)
In comparison, traditional Web search engines like Google and Bing use links to a content item as a primary signal to determine its popularity. The power of Graph Search lies in the ability to perform complex searches which emulate the old-fashioned human behavior of asking a trusted friend for a recommendation.
Facebook’s Graph Search does have a few limitations. Graph Search only knows about data within Facebook and only about people, places, photos and interests. News feed updates are planned, but have not yet been included — a regression from the older Facebook search. Results from Bing’s Web search appear to supplement information not on Facebook (currently broken if accessing Facebook from Italy; I’ve reported this as a bug). Facebook notes that Graph Search is a very limited beta program for English (US) audiences.
Twitter Finally Revamping Its Own Search
Twitter is also giving greater attention to making historical data discoverable. Over the past few years, they have rewritten their search engine and improved the search interface, yet users were still limited to searching on tweets from the last week or so. In February, Twitter announced they would begin to show some older tweets in Twitter search.
They also noted that ultimately, their goal is to surface the best content for your query, which isn’t necessarily the same thing as the “completeness” some users may be expecting. Currently, Twitter is offering “Top” and “All” options to filter tweets, and I am seeing tweets from as far back as 2008. Will they be filtering out more than just spam and protected tweets? That remains to be seen.
LinkedIn Discovers Value Of Comprehensive Search, Too!
Still in doubt that social media properties have awakened to the importance of a robust search feature as a key to unlocking the value of their data assets? In March, LinkedIn joined the fray as well, announcing a significantly improved search system. The race for better social search is on!
Google Social Media Properties Come With Search Built In
It shouldn’t be too surprising that Google’s social media properties have included rich search options since long before their rivals, from limiting a search to Events in Google+ to 3D movies over at YouTube.
What Communicators Need to Know: SEO Isn’t Limited to Google & Bing
Now that social media properties are revitalizing their search features, communicators need to start seriously considering how to ensure their social media collateral is easily found by searchers within a social media site. WPP head, Martin Sorrell, noted that marketers have been slow to move spend from print to digital, even as the people they presumably are trying to reach have migrated to digital.
Ironically, when marketers do spend on digital communication, there is often a serious disconnect between investments and user behavior. In Search Engine Marketing (SEM), most spend is on paid ads (PPC/SEA) while most user clicks are on organic SEO, i.e. earned media, results.
You know better. SEO isn’t just for search engines any more — it’s also for social media as part of a broad Social Media Optimization strategy. In the conclusion of this two-part series, I’ll look at social media SEO techniques applicable to multiple business cases and social media properties.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.