Facebook launched Graph Search as a beta product for many reasons: its gradual rollout to only US English users; its limitation to people, photos, places, and interests (posts and Open Graph actions still aren't included); an attempt to manage what could be outsized expectations from both users and Wall Street. But it's also because Graph Search isn't available on most of the devices people use to access Facebook, and it won't be anytime soon.
Graph Search isn't mobile, whether on the web or through any of the company's smartphone or tablet apps. This is a problem. It sharply limits the usefulness of Facebook's new search in almost all of the use cases presented on Tuesday. It cuts off access for Facebook's fastest growing and most devoted pool of users. And it confirms the unfortunate impression that Facebook either doesn't understand how and why to develop for mobile, or simply isn't concerned with making it a priority.
Facebook's users may be mobile, but development still isn't a priority
Let's stipulate, based on Facebook's enormous purchase of Instagram and its ongoing introduction of all-new, mobile-specific features like free voice calling on Messenger for iPhone, that Facebook knows it has to deliver for its mobile users. Its leadership isn't willfully blind or worried that mobile devices will cannibalize Facebook use. The people who run Facebook aren't foolish or malicious. And CEO Mark Zuckerberg said in September, "we really are a mobile company. All the code we write is [for] mobile." So then why does the Graph Search beta have a hole in it the size of half a billion mobile phones?
Right now, mobile support is on the list to be added to a future version of Graph Search "tomorrow," along with support for all languages and indexing of posts and Open Graph updates. Facebook's mobile team, Mark Zuckerberg said Tuesday, has been busy for the past year updating its apps on all platforms from HTML5 webapps to proper native applications while the Graph Search team worked independently on its own product. Add that delay to the already extended timeline of "weeks and months" to roll out this version of Graph Search, and we can expect not to see mobile graph search for more than months — maybe even years.
"We are working on mobile updates for search. For local businesses around them, consumers can use 'Nearby.'"
Meanwhile, mobile Facebook still has search capabilities even more limited than what's already on the desktop. "People can still search on their phones, such as people’s names and pages, but without the search improvements announced today, including searching in simple phrases or combining multiple sets of search criteria," a Facebook spokesman told Talking Points Memo. "We are working on mobile updates for search. For local businesses around them, consumers can use Nearby." (When asked about mobile rollout for Graph Search, a Facebook spokesperson repeated a similar statement, adding that "We think it will take time for people to understand the full potential of Graph Search and we plan to have a very slow roll out to improve performance and collect feedback. This process will help us as we develop the mobile experience." Facebook was unable to comment on specific timetables, its mobile development process, UI and design issues, or other features of Graph Search.)
To be clear, it's absolutely common for mobile versions of software to have fewer capabilities than on the desktop, even on the web. Facebook certainly has taken its time in adding features, including seemingly elemental online staples like comments, to its mobile versions. But we know that Facebook has more than 600 million mobile users, including more than 100 million who access Facebook only on mobile devices. We know that users in the US (for now, Graph Search's only market) on average spend more time using Facebook on mobile devices than on the desktop. And we know that all of these numbers, plus mobile ad revenue, are growing — and that analysts and investors are watching this growth extremely closely.
From the use cases we've seen, Graph Search would arguably be more compelling on mobile than on PC. Where does it make more sense to search for "Bars in Dublin liked by people who live in Dublin," assuming you don't live in Dublin — on your laptop in Chicago, or on your smartphone in Dublin? Likewise, Instagram's success shows that people are interested in viewing their friend's photos on their phones as well as on their computers. Breaking news breaks whether you're on the road or at home. Finally, Facebook's competitors in specialized search, whether it's Yelp or Foursquare or the search side of Apple's Siri, have established presence on mobile. This is all before we mention Google, who's staked out mobile and local search as a primary focus for the better part of the last decade. Facebook is already playing catch-up, and yet it seems unconcerned about falling further behind.
Facebook's future: services that "layer blankets of value over its core topography"
Mobile social networking, multi-point telecommunications, ad delivery, and local search are Facebook's future. As John Battelle writes, these are the services that "layer a fresh blanket of value over [Facebook's] core topography." Without building up those layers, Facebook's growth is stunted.
Facebook will need to not just talk up mobile, but invest in it
Moving Graph Search into mobile won't be like flipping a switch. Facebook has to work out the user interface on each of its apps, which for tasks like photo search might not necessarily be intuitive. It has to determine whether it will support voice input, which would give Graph Search better parity but poses a range of technical and linguistic challenges that Facebook may not be able to tackle yet. It almost certainly can't do this with a mobile app team that doesn't have the resources to build native apps and work on mobile search at the same time.
Facebook's engineering culture is one where products are launched, tested, iterated, and re-launched, again and again. "We have to have a rapid development pace where we're always trying to deliver the best version of what exists today, but at the same time think of what doesn't exist — the things people don't know how to ask for, and that will be useful a year from now," Facebook's Director of Engineering Andrew "Boz" Bosworth told The Verge last August. Right now, robust mobile search on Facebook doesn't exist. But we've seen the beginnings of what search on Facebook looks like. If Facebook really wants to be a mobile software company, it's time to deliver.