The company, which describes itself as “a group of humanists and computer scientists,” is basically trying to fuse machine learning with human understanding. It’s backed by Skype co-founder and über-angel Janus Friis, who lent a glowing quote to the launch release:

“Futureful makes it fun to surf the web again. Everyone should try it. And this is only the beginning of what Futureful can become.”

So, how does that square up with the reality of using the app?

I’ve been playing around with it a bit pre-launch, and I have to say the simplicity of the interface appeals, particularly when you compare it with the iPad version of StumbleUpon. That said, it can be a bit confusing at first, and that’s partially a data issue.

What you’ll see is an in-app browser, much as you’d get with Twitter or StumbleUpon, but with two rows of tags at the top. The first row is what Futureful’s artificial intelligence engine is proposing, based on the personalized information it gleans when you sign in using Twitter or Facebook. The second row takes in the tags you’ve picked.

That’s where it gets confusing. If you have multiple tags, sometimes they will all light up — if two are activated, for example, that means the piece of content you’re looking at covers both subjects. Sometimes, though, only one will be activated at a time. The example where I saw this happen involved the tags “swarm” and “robots”. This should really produce something relating to swarm robotics, perhaps, but it doesn’t.

“If you do some queries, a lot of them will combine,” co-founder Marko Anderson explained to me. “That might lead to content that’s older or less relevant. The system makes a decision that an article is closely related to ‘swarming robots’, but we’re not including that in the query.”

Similarly, sometimes if you swipe down to check out a second page related to your chosen tag, it will default to a Google search results page. As Anderson noted, this is about populating Futureful, and that’s hard to do before it has a significant number of users.

“Certain topics are very well covered and certain aren’t,” he said. “The technology is in very good shape, but it’s always a data problem with search. Now that we’re getting it out there, we’re going to continually ramp up the index. Users can also add content. It’s not alive yet until many people are dynamically using it.”

But is it fun? Yes, in a curious way. I like that it doesn’t regurgitate the same content over and over, and Friis isn’t blowing hot air when he talks about getting a sense of things to come. It’s a very different kind of search from Google, to be sure – think serendipity rather than looking for something specific – but the mechanism is intriguing, and it may well evolve into something useful down the line.

Speaking of Futureful’s future, Anderson told me the immediate next steps will involve more countries and more platforms – iPhone first, as that version is pretty much developed, then Android. But the company is also playing around with various data visualization ideas, perhaps around trending topics and so forth.

Which sounds good, as this team is clearly pretty smart and imaginative, but I just hope the Futureful app retains its elegant simplicity, rather than going too far down the ‘what kind of search would you like?’ route. There’s a nugget of something powerful here, and it needs to develop in its own way.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• NewNet Winners and Losers of 2009
• Carrier IQ and the continued erosion of operator trust
• The future of mobile: a segment analysis by GigaOM Pro

Futureful has been under rather stealthy development for two years, and the team is backed and mentored by Skype co-founder Janus Friis. It’s a bit like StumbleUpon, in that it’s an app that contains a browser (as opposed to being a browser – you can’t enter a URL) and is designed to help the user find new content.

However, Futureful is all about semantic tagging and artificial intelligence. As you browse, the app presents subject tags in a row at the top – click on a tag, and you get taken to another related page with its own set of tags. So, clicking on a ‘Silicon Valley’ tag may take you to a tech story, with the fresh tags above it including something like ‘Moore’s Law’. It basically provides an intelligent chain of content discovery.

What’s more – and here it really differs from StumbleUpon, in my experience – you never see the same content twice.

“It’s a new way to consume content,” co-founder Marko Anderson said. “We want the interface to be as simple and fluid as possible. Based on my usage of the service, I get very different things coming to me. If I choose ‘Silicon Valley’ and somebody else does too, the content might be different.”

I’ve not yet had a chance to play with the beta extensively, and of course it would take significant usage before it’s possible to evaluate how successful the algorithms are, but – if it works – Futureful could be a great way to kill time and learn new things. If you’re an iPad user, of course.

As Anderson pointed out, “mobile devices aren’t great for typing”. Add to that the need for enough space to show the tag row, and it makes sense that Futureful is launching as a tablet app.

“We started with web development, but the tablet is the ultimate consumption device, where the information finds you,” Anderson told me. “We haven’t ruled out a web version, but the smoothness in terms of the cleanness of the UI has just been better on the tablet.”

And Android? “We have no plans yet, but we have had some of the biggest players in the mobile industry ask us to build for them.” Same goes for the iPhone – maybe in the future, but first things first.

Also, there’s no business model yet. That said, if Futureful takes off I’d imagine there would be many options for making money off it.

Here’s a video:

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The 2013 task management tools market
• The state of cross-platform media measurement
• Analyzing the wearable computing market

I spent two days last week watching talks at the IE Group’s Big Data Innovation event, and here are five messages that really resonated with me. Hopefully, they’ll help your business, too.

1. Hadoop isn’t for everything. This should be common knowledge by now, but it bears repeating. Usama Fayyad, CTO at ChoozOn, pounded this point home when discussing how even Yahoo — Hadoop’s biggest champion and Fayyad’s former employer (he was chief data officer) — learned this lesson the hard way. Yahoo was trying to do some advanced customer segmentation with Hadoop, he said, but found out it would be 50 times less expensive to do that particular workload with a more-traditional database architecture. The realization ultimately killed that project, which was resurrected as analytics startup nPario. Yahoo is now a paying nPario customer. (At Structure:Europe in October, we’ll debate the merits of Hadoop versus traditional relational databases onstage.)

nPario’s Hadoop-free architecture.

2. Big data makes data science easier. I found this one of the more enlightening realizations, thanks in large part to how its messenger — Daniel Wiesenthal, chief data scientist at Sparked.com — was able to so clearly delineate between the sometimes-overlapping concepts of big data and data science. Essentially, he explained, techniques such as support vector machines and neural networks are time-tested and proven methods for “sucking every last ounce of information from your data set,” even when those data sets are small, but the techniques are very complicated, they’re difficult to interpret and they tend to break at scale.

A sample decision tree.

However, big data lets data scientists use simpler modeling techniques such as decision trees and regression, while letting the volume of data account for accuracy (and statistical significance) rather than a super-complex algorithm. And, Wiesenthal noted, using general-purpose big data technologies such as Hadoop means data scientists can develop and test models faster because their infrastructure isn’t tuned to a specific algorithm or problem type, and it’s designed to perform well against large data sets.

3. “Sometimes it’s more important to know what to kill.” Software-as-a-service pioneer Salesforce.com uses its big data platform to monitor the uptake and usage of various product features, said director of product management Narayan Bharadwaj, but the goal isn’t only to predict what new features to add next. Rather, he explained, using data to determine what features aren’t doing help a company like Salesforce.com decide to put those resources into more-valuable features. “Sometimes it’s more important to know what to kill,” he said.

Bharadwaj didn’t address this point, but it seems a logical next step would be to analyze the characteristics of features that perform well/poorly to get a sense of what works and what doesn’t from a design perspective.

4. Context adds value. To put it another way, if users know why they’re being shown a particular piece of content or offer or recommendation, they’re more likely to check it out. As a senior data scientist at StumbleUpon, explained, his company invests heavily in big data technologies and data science techniques in order to put the most-relevant web content in front of each user, but knows it’s not enough to expect those users to just trust the service’s judgment. Sparked.com’s Wiesenthal made a similar point in his talk, noting that services such as Pandora and Netflix are popular in part because they actually tell users something about themselves when recommending similar content.

If I like Metallica, I like …

5. Transaction data trumps search data. Mok Oh, chief scientist at PayPal, discussed the chain of events that begins with product searches and ends with purchases, and how it becomes increasingly difficult to determine signals when you start at one end of the chain and work your way toward the other. PayPal is trying to traverse this gap, however, beginning with the transactions it processes and using the other data at its disposal (both internally and from external sources such as Facebook and Gnip) to try and figure out who its customers really are and what they really want. He argues this is easier than, say, Google trying track users from search through purchase — unless, of course, they actually purchase something using a tool like Google Wallet.

Mok Oh discussing PayPal’s Customer Genome analysis.

I think the greater lesson, though, is to make lemonade from the lemons that are your data. Assuming a company’s greatest data resource is the data it has gathered specific to its own business, a path toward big data success is to use that data as a starting point and then get creative figuring out ways to glean more insights from it.

Feature image courtesy of Shutterstock user Bruce Rolff.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• A near-term outlook for big data
• Dissecting the data: 5 issues for our digital future
• Big Data Marketplaces Put a Price on Finding Patterns

Shareaholic's survey (click to enlarge)

Pinterest is now driving more referral traffic on the web than Google+, YouTube, Reddit, and LinkedIn — combined. That’s according to Shareaholic’s January 2012 referral traffic report, which is based on aggregated data from more than 200,000 publishers that reach more than 260 million unique monthly visitors each month.

In January Pinterest was responsible for 3.6 percent of referrals tracked by Shareaholic, up from 2.5 percent during the previous month. That means the site is quickly gaining ground on Twitter, which drove 3.61 percent of referral traffic in January, down from 3.62 percent in December. Pinterest’s ascent has been especially rapid when viewed through a longer lens: The site owned just .17 percent of the traffic in Shareaholic’s July survey.

Not surprisingly, Facebook is holding steady at the top of Shareaholic’s survey, as it was responsible for more than a quarter of all referral traffic in January. Next in line was StumbleUpon, with 5.07 percent. It bears mention that while the Shareaholic survey is global, in the United States market alone StumbleUpon has in the past unseated Facebook as a top driver of referral traffic.

It’s exciting to see a relative newcomer growing so quickly in the web space. While the web’s more established companies are quite powerful these days, the fact that a startup like Pinterest has successfully established its own foothold shows that the competitive landscape is still alive and mainstream users are open to trying things from new players.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• NewNet 2012: companies and technologies set to disrupt
• 12 tech leaders’ resolutions for 2012
• How publishers must adapt to multiple content discovery options

It’s the largest and most comprehensive branding and redesign initiative StumbleUpon has made in the company’s history, CEO Garrett Camp told me in a recent interview:

“We realized we weren’t conveying the experience well enough through our visual appearance. The power of our technology is remarkable, but now we’re finally doing justice to that with a modern, slick front-end. Finally, our front-end is as good as our back-end.”

The new StumbleUpon sign-in page (click to enlarge)

For StumbleUpon users, the changes will be very noticeable from the moment they visit the site (for now only the web version of StumbleUpon has the new look, but Camp says relaunches of its mobile apps are on the way.) The redesign is aimed at bringing StumbleUpon’s more granular features — such as the newly-implemented ability to Stumble according to specific interests — to the surface, Camp says. It’s also meant to show users content that they may find interesting, stuff they may not have previously noticed with the site’s previous look. Profile pages have more photos and social data, the site surfaces more personalized recommendations based on your current preferences and activity, and the new search bar (called a StumbleBar) makes it easier to search according to keywords and interests.

Essentially, it’s designed to make StumbleUpon more “sticky” than ever. And having played with the new version of the site for the past couple of days pre-launch, I have to say it succeeds at reaching that goal. The new interface strikes that elusive note in design: It makes the app simpler and more intuitive to use while at the same time giving the user more options to choose from.

Along with the redesign, StumbleUpon has launched a new feature called “channels” that lets users subscribe to StumbleUpon streams made by brands, websites, and noteworthy people. The feature is launching with more than 250 channels including ones from Details Magazine, The History Channel, and photo website 500px. For now it’s completely free for brands and celebrities to host channels, but StumbleUpon says it will be selective in giving the green light to prospective channel owners.

In all it’s a good move for StumbleUpon, and it’s one that seems long overdue. Once you compare the new look of the site to the old version, you realize how much was hidden under the surface. StumbleUpon has been growing at a rapid clip lately — the site now gets 33 million monthly unique visitors and has more than 20 million regular users, about half of whom joined in the past 10 months — so it is right to make sure it is putting its best face forward to make sure the success continues.

Here is a before and after comparison of StumbleUpon’s homepage design for a signed-in user:

My homepage on the old StumbleUpon (click to enlarge)

My homepage on the new StumbleUpon

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Post-IPO strategies for LinkedIn
• NewNet Q1: Content Farms and Niche Networks on the Rise
• NewNet Winners and Losers of 2009

StumbleUpon is sure to have some pretty disappointed users come the end of next month thanks to its decision to shut down some features of the service. The company says it’s shutting down groups, photo and HTML blogging, and the ability to select themes, all of which are going away as of Oct. 24. StumbleUpon says the decision is about being honest about what it’s good at (social discovery and recommendations) and away from what it’s not (trying to be Tumblr, Twitter, Facebook or MySpace).

So what’s going to change exactly? The ability of StumbleUpon users to customize reviews of the stuff they find on the site. That means no adding HTML or photos to them anymore. StumbleUpon will still let users write reviews and comment, but they’ll be plain text only from now on. Previous reviews that have been customized in any way will live on, but they’ll be converted to plain text reviews, according to the site’s FAQ. Users also will no longer have the ability to add themes or a background image to their profiles on the site. Soon they’ll all look the same: white background with black lettering only.

Some of these are features that are widely used (photo blogging) while others are less popular (groups, themes) among the site’s 15 million users. Over the years, the service added those features before there was Tumblr or Twitter, or a large chunk of the world was on Facebook. Now StumbleUpon says it realizes its team doesn’t have the time or resources to devote to features that they willingly admit other services do a lot better.

“Other platforms frankly do a really good job of [customization and self-expression] and that’s what they’re all about. While certain [StumbleUpon users] may not want to go, other services do offer a better experience of that,” said Marc Leibowitz, Stumble Upon’s vice president of marketing and business development, in an interview Thursday. Essentially, they want their users to come for the social discovery, and go elsewhere for other stuff. Leibowitz said that they know based on how people share StumbleUpon URLs that many of their users are already using Twitter, Facebook, Tumblr, and others for blogging anyway.

Attempting to put the changes in perspective, he noted how different the social web is today versus when his service got its start:

It’s amazing how StumbleUpon started in a day and age when the searching we take advantage of now didn’t exist. Google was just getting started. There are things and microblogging services that hadn’t even been conceptualized when StumbleUpon was already a few years old. The things we built out of necessity because they didn’t exist elsewhere are no longer warranted.

As reasonable as it sounds, it can’t come as very welcome news for users who’ve spent tons of time customizing their StumbleUpon review sites with HTML, or those that are just used to using StumbleUpon a certain way. But the company says it’s “exploring possible export options” for those who want to save their PhotoBlog before it disappears from StumbleUpon’s servers next month.

“We’re talking about a small, albeit passionate, group of users that were using these bells and whistles…it was costing us time to maintain these things that would be better applied to improving the discovery experience,” Leibowitz said. “We regret doing anything that disappoints our users, but this is being done to simplify the service and make it easier to use for a broader audience.”

Thumbnail image courtesy of Flickr user Frank Gruber

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• NewNet Q4: Platform mania and social commerce shakeout
• NewNet Q4: Platform mania and social commerce shakeout
• How social discovery is transforming entertainment

The new feature, called the Explore Box, allows StumbleUpon users to surf through streams of content related to specific keywords or phrases they type in. By typing in keywords such as “New York City” or “Katy Perry,” you can now use StumbleUpon to peruse the web through a more narrowly directed content stream. Previously, you could only explore the web using StumbleUpon within certain sets of around 500 topics. With the explore box, you now have the option of viewing StumbleUpon content within hundreds of thousands of much more granular interests.

The Explore Box feature is launching on StumbleUpon’s web app Wednesday and should be deployed on its mobile apps in the near future. Here’s what it looks like to use (click image to expand):

In a recent interview, StumbleUpon CEO Garrett Camp told me that the reasons behind launching the Explore Box now are two-fold. First, customers have long been asking for a way to use StumbleUpon to find content on more specific things, such as their favorite celebrities or sports teams. And secondly, the service had to grow large enough for the deployment to make sense. Camp said:

“We’ve had a prototype of this for a while, but earlier on we felt like we weren’t getting enough results. We couldn’t really do it at the beginning because we needed a bigger index. In the earlier days, if you had typed in a specific term, you may have gotten just five results. Now, we have millions of pages of content, and it will just get better over time.”

Camp stressed that the company is certainly not abandoning its core goal of bringing serendipity to web surfing. “With the explore box, it still feels like stumbling upon things. But it feels like more focused stumbling,” he said. He was quick to emphasize that the explore box is not meant to be a search feature. “I think the term explore is better; it connotes that feeling of active seeking, but you still don’t know exactly what you’re going to get.”

Additionally, Camp told me StumbleUpon has started to bring in certain RSS feeds that will help bolster the amount of long-tail content it offers through the explore box as well as its flagship product. Previously, StumbleUpon only included pages that had been explicitly added by users.

In all, the explore box is just the latest example of how well StumbleUpon has been able to iterate very useful new features since it spun out from eBay  two years ago. As Camp told me in an interview last month, “Now that we’re independent again, you can have an idea here and make it happen very quickly.”

Here is a video of the Explore Box at work:

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Google doesn’t like walled gardens — except its own
• Pinterest: signs of staying power
• Facebook’s IPO filing: ideas and implications

StumbleUpon CEO Garrett Camp

At nearly ten years old, StumbleUpon is certainly well past the point of attracting users by being a new and hot trend. But the San Francisco-based web discovery engine has proven that slow and steady growth can still lead to success: Earlier this month, StumbleUpon surpassed Facebook to become the biggest traffic driver among social media startups in the US.

Given Facebook’s status as a media darling, its being unseated by the decidedly lower profile StumbleUpon has taken some people by surprise. I sat down last week with StumbleUpon co-founder and CEO Garrett Camp to find out exactly how the company has ascended to this point — and where it plans to go in the weeks and months ahead.

Here are a few main factors that have led to StumbleUpon’s recent strength:

• It’s a labor of love – Camp co-founded StumbleUpon in 2001 while doing post-graduate work in his native Canada. Since then, the company has been through its fair share of ups and downs — and Camp has been there through it all. StumbleUpon was essentially run out of its co-founders’ bedrooms up until late 2005, when it took on $1.5 million in seed funding and moved to San Francisco. Shortly thereafter in 2007, StumbleUpon was sold to eBay for $75 million. But like so many big M&A deals, the integration did not ultimately work out, and in 2009 Camp and one of his original co-founders Geoff Smith bought StumbleUpon back along with a group of investors.

   
  Camp’s dedication to StumbleUpon through thick and thin seems unique in an industry where the label “serial entrepreneur” is often worn as a badge of honor. The CEO told me he has stuck with StumbleUpon for one basic reason: It’s still the best place for him to turn his dreams into realities. “At the end of the day, I just really like the service,” he said. “To this day, when I have an idea, it always seems like StumbleUpon is still the best place to implement it. Now that we’re independent again, you can have an idea here and make it happen very quickly.”

• Technology comes first – More than two-thirds of StumbleUpon’s 80 person staff is technical, said Camp, who himself is trained as a software engineer. “There’s so much technology under the hood. We’ve spent so much time making the system, and by now it’s pretty complex and advanced.”

   
  Today, Camp said, StumbleUpon is just starting to reach the stage where enough people are online to really take advantage of the system. ”For the first several years we never made a proactive effort to talk to people. I was in Canada in grad school, and then in eBay there wasn’t a lot of press either,” he said. “In the last couple years people have really paid more attention to us; but it’s taken us years to get to this point.”

• It’s no one-trick pony – StumbleUpon’s roots are firmly in the web — but Camp says that’s really only the beginning. According to him, the company’s core technology can be fairly easily translated to many other mediums such as television, gaming, social networking, and beyond. It branched out in November with an app discovery service for Android. “The idea of providing recommendations without knowing exactly what you want can be useful in so many places,” he said. “We’ve built this backend, and now we can focus on one or two platforms at a time. But the only limitation [for growth] is the number of engineers we have.”

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Google doesn’t like walled gardens — except its own
• How social discovery is transforming entertainment
• Social media in Q1: commerce and discovery dominated

Source: StatCounter Global Stats – Social Media Market Share

For months, Facebook has been the biggest social media site that drives web traffic in the United States. Those days are over– for the moment, at least.

Web discovery engine StumbleUpon is now the biggest traffic driver among social media websites in the US, according to global web analytics service StatCounter. The company unseated Facebook at the top during June 2011, according to the latest StatCounter social media data. StatCounter tracks hits to over 3 million websites, and its social media data is gathered by analyzing every hit referred by a social media site.

This is certainly not StumbleUpon’s first brush with Internet fame. The company, which finds and recommends relevant web content, was originally founded in 2001 and sold to eBay for $75 million in 2007. But like so many big M&A deals, the acquisition did not go as smoothly as hoped. In 2009, StumbleUpon co-founders Garrett Camp and Geoff Smith bought the company back along with a group of investors.

In the two years since then, the company has been working quietly and diligently at its comeback. StumbleUpon closed on a $17 million funding round in March to expand its product offerings to the mobile and television spaces. The company has also been working on its revenue strategy, with the development of a “paid discovery” offering for advertisers. StumbleUpon briefly surpassed Facebook in StatCounter hits for two weeks back in February, but Facebook quickly regained the top spot.

This time, StumbleUpon’s turnaround is gaining traction at exactly the same time that Facebook’s new user numbers are reportedly slowing down, and Twitter seems to be dealing with stickiness problems. Clearly, something about StumbleUpon’s strategy is working very well. It remains to be seen how long StumbleUpon can retain the top spot — its competitors in the social media space aren’t known for going down without a fight.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Connected world: the consumer technology revolution
• Social 2013: The enterprise strikes back
• Examining the rise of crowd labor platforms in 2012

The BBC iPlayer first added social sharing last September, initially hooking into Twitter and Facebook at the same time it expanded the number of connected devices it could be found on. But by adding sites like StumbleUpon and Reddit, the BBC shows proves that even in video, Twitter and Facebook aren’t everything.

Online video providers are increasingly seeking new ways to allow viewers to promote the videos they’re watching with friends, in an effort to keep driving up the number of video views. While it’s been incredibly popular for viewers tuning into short-form and user-generated clips, social discovery is also being used by professional content providers as a way to boost traffic.

The addition also comes as sites like Reddit and StumbleUpon have focused on providing video recommendations in addition to text links. The former launched Reddit TV a few years ago to improve viewing videos on the site. And StumbleUpon created a Pandora-like service last year for discovering videos through its service.

The ease with which users can share and see what their friends are watching could be one reason why the iPlayer continues to post record numbers for video requests from its viewers month-after-month. Adding more social distribution outlets — and therefore increasing the number of discovery outlets — could grow its viewership even further.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Connected world: the consumer technology revolution
• Welcome to the New Paradigm: TV Makers Rule
• Report: NoSQL Databases – Providing Extreme Scale and Flexibility