Eric Wilson's Blog

Social TV – The Second Screen

Eric Wilson — Thu, 02 Feb 2012 03:04:14 +0000

The chance that you or someone you know has a tablet or smartphone is growing. The chance that those same people are using one or both devices while they watch traditional programmed television is growing even faster.

The Second Screen, as it is being called, is a new and compelling avenue for content providers, community builders and advertisers to come together.

Why is the second screen driving this change? Three reasons:

Flexibility.

Connectivity.

Data.

That’s So 1980

The old guard Set Top Box (STB), which is what the majority of households have installed, is difficult if not impossible to upgrade once it has left the factory. This is chiefly because the ability to perform meaningful over-the-air (OTA) upgrades or changes (i.e. how applications found on your smartphone are updated with the latest release) just doesn’t exist. Without OTA, the ability to change and adapt to new customer feedback is anemic at best.

Connectivity for the STB device - downstream to the home – is good. Bandwidth is constrained, however, on the upstream channel. This constraint limits the size and frequency that user data can be fed back to the operator. If the operator doesn’t know what is happening in the ecosystem, moment to moment, how can they create a compelling experience to the user? Social networks might as well not exist either, as there is no way to connect into them.

Lastly, user data is the super fuel that is required to transform the viewing model from ‘push’ to ‘interactive’. STBs dearth of data necessarily creates a disconnect between service provider and service consumer. They don’t know you and therefore can’t tailor an experience for you.

New And Improved Formula – Meet Social TV

There are now more than a few social TV companies vying for space on your mobile device or tablet. Because these fledgling enterprises are creating dynamic applications on powerful, connected devices they have the grand opportunity to deliver a personal media experience never before imaginable.

Given this new freedom of interaction and deployment, a few strategies are emerging that aim to create that winning user experience.

Companion Services

Got cable? These companies provide cool new ways to interact with the programming, share the experience with friends, buy related items and even earn points for being a loyal watcher. Nice.

Zeebox

This is a clever app for your iPad that seems to be focused heavily on social interaction. Find out what your friends are watching, invite a friend to watch a show with you or buy the games, music or other merchandise related to the show. Check out this video for a comprehensive overview.

Viggle

This iPhone app is all about checking in to a show – and they have some powerful tech to help you do this automagically. After checking in, users earn reward points for the length of time they spend watching (points are redeemable for physical goods – cool), can find out about which shows are being talked about, post updates to Facebook and more. While not connected to the social graph yet, it is surely in the works.

Cable Cutting

Frequency

If you want to go completely outside the bounds of connected television, then Frequency offers up a compelling product. The iPad app can connect into your social graph, pull in and organize video and social conversation while providing an elegant navigation metaphor. More can be seen in this review below.

What’s In It For Me?

As I’ve mentioned before on my blog, doing something smart with the user generated data will go a long way to create differentiated experiences. I’m looking forward to the possibilities.

Machine Learning And The Startup

Eric Wilson — Fri, 27 Jan 2012 03:13:50 +0000

Orthogonal Contexts

I’ve been involved with two things these past few years that, in my mind, have very similar execution success characteristics: creating a startup company (SC) and a recommender system based on Supervised Learning (ML).

Looking at how to implement each from scratch reveals interesting parallels.

The Cold Start

The challenge that is faced when first implementing a recommender system (recsys) in a new environment is that the ML engine is not very smart about what items to recommend. This makes some sense in that without rich historical data about how an item has performed (e.g. purchases, shares, ratings) there can be no concrete statistic on an items success probability. Historical data can be analyzed and used to help train the engine but that data’s usefulness is quickly obviated as new information is gathered from the live system.

A specific lack of information means that out of the gate there are some postulates guesstimated and models built which are based on the best information available. It is well understood that these models will change, features will be dropped or added and a next generation model should be in the works as soon as possible.

The same issue typically faces a new startup. Not much is really known about the best plan for success. Which products will win users over or land clients with eager, open checkbooks? Forecasting revenues for a brand new venture seems nearly an exercise in mental masturbation. New products envisioned by the ambitious team have no proof points.

Much like the ML implementation, understanding that the initial products and business models are literally a jumping off point is important. Not betting it all on an unproven theory will save time, energy and money.

Learning And Adapting

The ML engine learns how to provide more accurate recommendations over time by getting feedback from the ecosystem. The engine will suggest a variety of items, based on statistical analysis, that it believes will match a users interest at a particular time and for a specific context. The results of this endeavor are collected as log data and are returned to the ML system. This data set represents the success and failure of the previous set of predictions. As real time passes, the ML engine consumes the latest available data sets and (if properly designed) the accuracy of the engine’s future predictions becomes more acute.

Likewise the model through which these data sets pass may need to be updated to reflect new learning. Testing a new model against the previous model to reveal its efficacy (A/B testing) is an absolute prerequisite. Only after a rigorous testing will you know if the new direction you have chosen will provide better quality or if you should stick with the status quo.

Building a company on a hunch or in a complete vacuum is a recipe for failure. As one of my favorite startup bloggers Steve Blank said in no uncertain terms, “Get out of the building”. It is imperative to canvas the market you are trying to address to see if the widget you want to produce has value.

Getting feedback about the direction you are taking the venture toward or the features that you want to add to the product will be worth more than any blue-sky session at a whiteboard. Being agile, testing and iterating quickly over product ideas and features as data comes in from the field will produce dividends and should help steer you toward a successful path.

Measuring

It is important to the success of the ML project to establish KPIs before coding begins. These metrics should be meaningful and not academic or esoteric. Typically measuring the users reaction to the recommendations is the best metric; e.g. is your revenue per user or page view going up?

Based on clear, measureable success criteria the model can be refined and tuned to best perform.

Peter Drucker said it best, “What gets measured, gets managed”. Without measurable success metrics for your fledgling business how will you know if you are succeeding? You should create KPIs that have actionable meaning for your business. These KPIs should necessarily cover more than just your product or service. Internal KPIs for marketing, IT, business development etc. can be captured and presented in a company dashboard. Clear goals and metrics align the people getting the work done and provide the impetus for acceleration or course correction.

Rinse And Repeat

For both ML and SC, relevance is determined by maintaining a contemporary view of the ecosystem that it serves. Hence:

Learn->Adapt->Measure->Learn->Adapt->Measure…

iPhone 4S Battery Life

Eric Wilson — Tue, 15 Nov 2011 01:55:36 +0000

I had assumed that with the advent of owning such a powerful computing device that there would be drawbacks. Battery life seems to be the clear winner. It takes some juice to run this little gem.

Although Apple has released an update to iOS that may address some of the issues, I wondered if I couldn’t also help out my good friend the iPhone by doing some basic housekeeping.

I’ve been in and around mobile development going on eleven years now (yipe!) and there are some common reoccurring memes with respect to getting the most out of mobile device battery life. Mostly they center around making your device do fewer things. Simple enough.

Turn Off WiFi

Unless you are cruising at some well-known available network it’s best to turn WiFi off. Your device will be looking for networks to join, generally sniffing around, all with convenience in mind for its owner. Turning WiFi off when you’re not in range of a well-known network (e.g. home, work, Boingo) should conserve energy.

This behavior is similar to the no_wireless_signal battery drain issue of years past. Essentially, when consumers entered a building where the wireless carrier signal was lost, the mobile device would continuously scan for a connection. The result either being a carrier signal was found or the battery was drained to empty.

More often than not it was the latter.

Shut Down Background Apps

The iPhone is good at multitasking quite a few apps by allowing them to run in the background. The question I had for myself though was; how many of these apps are actually still running and do they really need to be?

Fortunately there is a facility that allows you to check to see what’s running and more importantly stop an app when you aren’t using it.

Double click on the home button. Doing this will present a taskbar which contains all running apps. This is a cool facility for switching between running apps (just tap the app you want to switch to). Swipe left or right on the taskbar to see more apps – trust me there will be more.

Now what? Look for apps that you aren’t using regularly and shut them down.

To shut down running apps, touch and hold one of the app icons and you’ll see a red circle with a minus sign appear. If you tap an icon the app will be stopped. Neat.

Manage Push Notifications

This is a useful feature when managed conservatively.

However, it seems that every app has something to say. All this chatter makes your mobile device perform network related activity and will be, of course, yet another source of battery drain.

Turn off unnecessary notifications.

Location, Location, Location

Location based activity makes your device perform a lot of work - your iPhone has to carry on an intense data conversation with cell towers or GPS satellites (in the case of turn by turn directions the location requests are constantly happening). Limit the frequency with which you perform location based activity. Be especially mindful of those apps that perform regular location updates to / from the network.

Feed The Need

If you just can’t stop doing what you are doing then maybe you need to get some extra juice to help out until you can get next to a power outlet.

I’ve been using a combination case protector / external rechargeable battery from mophie. So far it’s been working quite well. It’s smart enough to either use it’s own power first before using the iPhone’s or can be used to charge the iPhone’s battery.

KloutVille

Eric Wilson — Sat, 29 Oct 2011 00:33:52 +0000

Klout is really a MMORPG

It struck me today, as I witnessed colleagues scrambling around to find out how they could repair their damaged Klout scores, that Klout has created the largest MMORPG on the planet.

What do I mean by that?

Let’s take a look at some excerpts about MMORPG’s from Wikipedia to see the parallels:

“Although modern MMORPGs sometimes differ dramatically from their antecedents, many of them share some basic characteristics. These include several common themes: some form of progression, social interaction within the game, in-game culture…”

Now let’s apply these some of these feature definitions to KloutVille:

THEMES

“Often these elements are developed using similar tasks and scenarios involving quests… and loot.”

There are daily quests that all players in KloutVille can partake in: CombatRecommend, CombatShare and CombatCreate (RSC). RSCing is how a player progresses to ever increasingly higher levels of Wealth in KloutVille.

Loot is a bounty that can be granted if the Game Master deems worthy the level of RSC activity and the total number of connections a player maintains.

PROGRESSION

“In nearly all MMORPGs, the development of the player’s character is a primary goal. Nearly all MMORPGs feature a character progression system in which players earn experience points for their actions and use those points to reach character “levels”, which makes them better at whatever they do.”

The goal in KloutVille is fundamentally to develop your online character. As such, you’ll want to develop a plan around RSCing and you’re going to need a kingdom to elevate the impact of your efforts. You’ll want to connect your character in KloutVille with as many other ‘worlds’ that you dwell in (Planet Facebook, Twittersphere, GooglePlusLand) thus enlarging your kingdom and your Wealth. You may also be rewarded with combat-useful items such as the Influencer Potion.

“The accumulation of wealth (including combat-useful items) is also a way to progress in many MMORPGs, and again, this is traditionally best accomplished via combat. The cycle produced by these conditions, combat leading to new items allowing for more combat with no change in gameplay, is sometimes pejoratively referred to as the level treadmill, or ‘grinding’.”

Productive, daily RSCing can prove itself to be an extraordinarily difficult task. A player needs to maintain a pace and velocity such that a one’s Wealth is not diminished due to atrophy. This cycle has revealingly lead many players down the path of grinding. Some of the more popular grinding RSCs:

Weather
Airline Updates
Song Lyrics
Coffee quality
#FF
I can’t believe they arrested me for this again – check out this pic (Okay, used more infrequently but still grinding – no?)

“This sometimes forces players to change their real-world schedules in order to “keep up” within the game-world.”

I’ve seen players enter into a full-blown panic when they realize that they haven’t RSC’d for a day or two. All real-world work efforts comes to a halt whilst they logon and care for their KloutVille character. Take care to manage your time here wisely.

SOCIAL INTERACTION

“MMORPGs almost always have tools to facilitate communication between players. Many MMORPGs offer support for in-game guilds or clans (though these will usually form whether the game supports them or not).”

A foundational element of the KloutVille experience is the harnessing of social interaction from within the player worlds: replying or retweeting in Twittersphere, sharing in the magic Circles of GooglePlusLand and ascending the Walls on Planet Facebook. Bring your clans together and gain strength.

ROLEPLAYING

“Most MMORPGs provide different types of classes that players can choose. Among those classes, players are encouraged to roleplay their characters, providing rules, functionality and content to this end.”

If you want to be able to roleplay as a particular Meme Magician, you are going to want to invoke the magic hashtag. Given a particular success or frequency rate of your hashtag casting, the Game Master may decide that you have earned the right and will grant you the Influencer Potion for the Meme you seek.

CULTURE

“Since MMORPGs have so many elements in common, and those elements are experienced by so many people, this has led players of many games to expect “buffing” or “nerfing”, which describe the strengthening or weakening, respectively, of particular game elements. (“Buffing” also refers to in-game effects that temporarily enhance performance; both usages come from a core meaning of increasing power levels.)”

This week saw both buffing and nerfing take place. It was kingdom wide.

Those that were buffed, bonded together in song and jubilance. They were the chosen ones.

Those that were nerfed experienced the full wrath and might of the Game Master. True it was that a portion of the nerfed were merely wounded, while others were cut cleanly in half. They were now the downtrodden, the forgotten.

Avoid the nerfing at all costs.

What’s In A Number

Ultimately I am looking for tools and information that can better instruct me on tangible efforts that I can undertake in order to improve my reach and better understand my audience. Tools like HootSuite and bitly track the reach of the links I provide – knowing reach helps gauge the overall popularity of a particular message or blog post. Understanding what’s popular can be prescriptive when approaching the next topic or post issued.

Crowdbooster provides some clever metrics including times of the day that you may want to post (as determined by the success of previous posts during that same time period). If I not only understand what makes for a successful post but also when I should release it, I can be assured that I’m making the best effort for my audience to actually see it.

Google+ has Ripples which visually represents how a post spread and to whom. This is really more for my entertainment purposes but seeing that something I posted made it to many groups of people that I don’t even know is, well, kinda cool.

The bottom line is that the measurements presented from the companies I mention above are very clear. How they arrived at such results is not up for debate. Why this data can be directive for future posts is self-evident.

I don’t get these kinds of practical questions answered by Klout, so until that day arrives Klout will be a fun game of strategy and intrigue where the goal is to outsmart the machine.

But perhaps the kind of insight I describe above is not the end goal for Klout. Perhaps what is being established is an enormous pool of opt-in users who can be micro-targeted by advertisers. Interestingly, this same group of users can concomitantly market and propagate brand messages by means of social trust to their own personal circles. Hmm.

It’s all speculation. But I do know one thing; my KloutVille character needs help getting a KloutBFG to…

No Virginia, There is No Privacy

Eric Wilson — Mon, 10 Oct 2011 01:45:17 +0000

Frictionless Exposure

The announcement of Facebook’s CEO Mark Zuckerberg that he characterized as “frictionless experiences” got me thinking: Does the notion of “privacy” exist in today’s digital world? A world where your unique self and its interesting voyage through life are now fully recorded, analyzed and monetized. In truth, we are all individually one giant step closer to starring in our own Truman Show.

Really?

Yes – and to support this notion let’s take a quick look at a few example services we use in our daily lives.

SOCIAL WEBSITES

As a logged in Facebook user, websites (who integrate with the Facebook platform) already have the ability to know who you are when you visit. Before the latest enhancement to the Facebook platform, you could ‘like’ a page and if your friends visited the same page they would come to know that you liked it.

All this explicit user action information was of course also sent back to Facebook.

The new changes in the Open Graph API may now allow websites to automatically post an update stating what article you’ve just read – even if you didn’t explicitly ‘like’ it.

This concept of automatically posting your personal event data will be extended as Facebook integrates video and music services, streaming all the things you watch, listen to and buy. And you may not be able to opt out even if you log out.

Now how are you going to approach your web browsing experience? I’m sure that you won’t mind if folks know about your Discovery Channel or Economist browsing habits. But say you’re planning a vacation trip as a surprise; you go to a vacation site and that information gets posted to your Wall. That automatic Wall post could potentially spoil your gift.

You get the idea.

FREE EMAIL

Google has been famous for its bold statement of never having to delete an email - yet another cool, free service from the data giant. Of course you have to be comfortable with Google data mining all of your email. Every single message is parsed and analyzed in the name of understanding you better.

Every. Single. Message.

The mining is done to provide you with better services of course [read better advertising].

SEARCH ENGINES

Since December of 2009 Google has been keeping track of all of your searches and has been building a profile on you. This is not data in the aggregate – it’s specifically about you. Google uses your profile to tailor your query results to what Google thinks is a best match for you as an individual. This means theoretically no two users search results are the same. It’s also means that Google can control the resulting data that you are presented with; both inclusive and, more importantly, exclusive.

Bing even goes so far as to fuse your Facebook friends with your Bing search results.

SMARTPHONE DEVICE MAKERS

Earlier this year Apple and Google got into privacy trouble when it became known that both companies were collecting and storing location data on their customers; sometimes even when the location feature had been ‘turned off’ by the phone owner.

WIRELESS CARRIERS

Here is an except from the latest Verizon Wireless contract agreement:

What kind of data is used to prepare business and marketing reports?

Verizon Wireless will use the following categories of information:

Mobile Usage Information:

Addresses of websites you visit when using our wireless service. These data strings (or URLs) may include search terms you have used.

Location of your device

App and device feature usage

Consumer Information:

Information about your use of Verizon products and services (such as data and calling features, device type, and amount of use)

Demographic and interest categories provided to us by other companies (such as gender, age range, sports fan, or pet owner)

We use this Information in a manner that does not identify you personally to prepare these reports for our own and third party purposes. Verizon Wireless may also share location information with other companies in a way that does not personally identify you so that they can produce limited business and marketing reports.

All of your activites are tracked and ‘aggregated’. There is some variability with respect to opt-out but pretty much your information is collected and that’s that.

ONSTAR

Concierge services and navigation company, OnStar, made an announcement that it will continue to keep track of an OnStar equipped vehicle’s speed and location information – even after the consumer discontinues the service.

That’s what I call frictionless.

CREDIT CARDS

Law enforcement agencies can and do seek real-time surveillance of credit card transactions. Fourth Amendment be damned, as it takes merely a subpoena to gain access to your account. Although cash may be ‘inconvenient’ it’s one of the last ways to try to avoid prying eyes. Something to think about next time you and a friend use Square to even up the bar tab.

Hiding in Plain Sight

It seems that Millennials are far less concerned than previous generations when it comes to privacy. The amount of personal information freely shared is unprecedented and to date it seems that Zuckerberg’s Law is proving to be true.

Users actively participate with physical check-in services like Foursquare and Yelp that publish the time, date and location of the establishment they are sitting at.

Then there are virtual check-in services with the likes of Miso and IntoNow where you broadcast to the world what you are watching on TV.

The ACLU will do its best to try and protect your rights but as personal information becomes more available and readily accessible most likely theirs will become a Sisyphean task.

The digital age isn’t something you can hide from, but perhaps it is a place that you can negotiate with more prudence; unless of course you subscribe to former Sun CEO Scott McNealy’s proposition.

His solution? “Get over it.”

#ThankYouSteve

Eric Wilson — Thu, 06 Oct 2011 04:58:24 +0000

Wonderful tribute by Boing Boing with the link to the statement from Apple’s Board of Directors.

iGrumble

Eric Wilson — Wed, 05 Oct 2011 06:52:32 +0000

The Reveal

The long awaited announcement from Apple about its next generation iPhone happened today. Prior to this morning’s conference at the Cupertino offices, pundits and fanboys alike had been whipping the mobile population into a froth with the feature speculations, design assumptions and capabilities wish lists. It was almost impossible not to hear or read about ‘the latest iPhone rumor’.

In short, the run up to the day of reveal was exhausting.

Unfortunately it is rare when the dream matches up to reality. So it seems there is a lot of grumbling and unhappy mobile customers riffing about the shortfalls of the launch.

I don’t understand why – there are some fantastic features that I have to say I’m pretty satisfied with.

What am I excited about?

WORLD PHONE!

I have been holding on to my dear old BlackBerry Tour 9630 because it is a world phone. How great it is to have both a GSM and a CDMA radio inside. Since I’m a Verizon customer I can roam seamlessly on Vodafone when I’m outside the U.S. The new iPhone is also a world phone and this multi network roaming capability removes the last of my ties to BlackBerry.

Sweet.

SIRI

I have been working with ML technologies similar to those that power SIRI and am really happy to see AI making it into mainstream products. I think SIRI has a lot of promise and I’m looking forward to seeing it mature.

New Camera

8-megapixel camera that supports HD (1080p) video recording, up to 30 frames per second with audio.

I find that there are so many spontaneous situations in which I want to use my phone’s camera; having 8-megapixels will provide me just that much more richness of detail. Who needs a digtial camera anymore?

1080p HD video. What else is there to say?

iMessage

I use BlackBerry Messenger everyday. It’s a fantastic tool for communicating and sharing. Well, now I have the equivalent functionality and more with iMessage. I can see who’s reading the messages I have sent, do group chat, send pictures and videos. Everything I loved about BBM is now present with iMessage.

The comments about ‘what happened to iPhone 5’ will surely continue to bounce and tweet over the coming days. Guess that means there will be less of a preorder queue for me. :)

The End of the BlackBerry Prayer

Eric Wilson — Wed, 28 Sep 2011 22:07:33 +0000

7 of 9

It’s amazing to me still how much my BlackBerry device has become my indirect taskmaster. How you ask? I have ‘cleverly’ connected it to both my work and personal email accounts. I’ve installed applications like Foursquare, Twitter, Skype and LinkedIn. I have a few folks on the BlackBerry IM service as well. I regularly use the mobile web. My mobile device is my portable office – even when I’m in my office.

I am completely connected to everything and everyone.

More often than not I have the device set to vibrate. With the variety of chirps, beeps and pleasing chimes going off I certainly wouldn’t want it to disturb anyone.

It’s a real time saver since everything I need is right there in the palm of my hands. When it summons me I respond almost immediately. I find that I’m often staring down at it, holding it with both hands, practicing the BlackBerry Prayer.

This past weekend I was happily tweeting, checking in, responding to emails when, in an instant, my device was given a lobotomy: The track ball would no longer scroll down.

I Am But A Pinball

I like to think I’m a planner. I have good intentions each morning for the new day ahead. I subscribe to the GTD philosophy and employ many of the techniques for productivity. Indeed I get a lot done but my good friend the mobile phone always seems to be interrupting my rhythm.

The phone’s notifications can really get me bouncing around as I try to keep up with all of the real time events streaming in that cry out for attention. As such, I’ve become expert at managing the random flow of notifications and data. Things do get done.

But by days end my circuits are fried.

At the end of the day, I need to unplug from the grid and reconnect with the world around me. Family and my four-legged friend take precedence now.

This was my daily routine up until the track ball stopped scrolling. No longer could I update statuses; I couldn’t select Skype conversations. I couldn’t navigate to my apps and fire them up. I couldn’t surf the web. Through some clever keyboarding I was able to get to priority email but my phone basically became just that – a phone.

The Horror.

print “Hello World!”

It seems I have gotten something back since this device malfunction. At least for a while I can’t be interrupted or distracted because, well, I can’t do anything about it. I have to wait to respond. I am ‘disconnected’ in a sense – free.

This situation could be a boon for what becomes a priority and what is not. It may well be a cure for my self inflicted A.D.D.

The iPhone Event Horizon

I have read blog posts where folks write about ‘managing your inbox’ and how email can keep you from getting any work done. I’d argue that it’s not just your inbox, but the sum of all event notifications that you sensibly have to negotiate on a daily basis – SMS included.

Now, my BlackBerry replacement plan has been to purchase the iPhone 5 iPhone 4S. Of course that plan started early in the year, so I have been patiently waiting. It seems I won’t have to wait too much longer.

I do envisage myself exercising this new skill of moderation and prioritization when the iPhone finally arrives – Angry Birds be damned.

Recommender Systems are the New Oil Refineries

Eric Wilson — Wed, 21 Sep 2011 02:24:32 +0000

From Crude to Shrewd

I read a post by a colleague of mine, Richard Kastelein, who covers the Social TV space in which he proclaims “Scene-Level Television Metadata – Tagging TV – is The New Oil in the Industry”. There is a great deal of merit to this train of thought, and indeed having access to this kind of metadata about television programming has been a long time in coming.

I like his New Oil analogy and, given my interest in recommender systems, thought it appropriate to continue the analogy meme with the statement that Recommender Systems will be the New Oil Refineries.

What does this mean?

Crude oil has limited applications until it as been processed and refined into myriad useful products. Likewise, metadata itself provides a modicum of utility, but its full potential becomes realized when combined and processed with usage behavior.

Until recently, comprehensive television usage data was not captured, as there was no pipeline to transport the information. A burgeoning well of data was left untapped because of the limitations of the Set Top Box (STB) platform itself. The STB had limited data bandwidth, processing power and the majority of its logic was burned into its firmware – making it inadaptable and expensive to upgrade in the field.

Over The Top (OTT) solutions have arrived and are aimed squarely at solving the STB challenge. These second screen companion devices (e.g. iPads, tablets) provide the facilities to capture quality data as well as the feedback mechanisms for retrieving that data in a timely fashion.

OTT solutions will deliver the missing data pipeline and concomitant data elements that are the fuel needed to power sophisticated, data hungry recommendation platforms.

What are the applications?

Recommender systems have the capacity to analyze and decision against incredibly large datasets. As noted above television content can be one of the recommended products delivered – but so can advertising; be it imbedded, pre or post roll.

Ready, Fire, Aim

If it is indeed true that 82% of TV Ads generate negative ROI then surely it makes plain sense to have a powerful machine learning system in place that can accurately target content to users in real time.

Television advertisers just don’t understand me

The combination of new user interaction models, the availability of standardized TV metadata and powerful machine-learning infrastructures will deliver a more personalized experience to each user. Indeed as users move en masse to tablets for companion viewing and in some instances primary viewing, or as systems allow for user logins, there will be an even greater opportunity to understand the individual. This personalized experience will reach across all content – including advertising.

Not Just “New” Search

One of my favorite VC bloggers, Mark Suster, posted an article on his site not quite a year ago; The Future of Television & The Digital Living Room. He touches on the subject of better search in in his 7^th point about Content Discovery and the improvement of the user experience. What’s missing from his discussion is the introduction of recommendation engine services. Better Search will not be the difference in next generation TV. The power of understanding the consumer at a heretofore-unavailable scope via this new OTT data rich model of usage interaction combined with a wealth of television metadata is how we’ll be equipped to offer meaningful, relevant content discovery experiences.

Consumers need relevant content served up to them - let’s fire up the refinery and make it happen.

Just Say NOcial

Eric Wilson — Thu, 01 Sep 2011 00:38:31 +0000

Social Is The New Black

One of the most abused phrases I hear these days is Social + [insert any noun here]. Conventional wisdom appears to regard communication between people as a bold new concept. This exciting new medium adds value to any business (check out the astronomical valuations of the privately held companies on SharesPost).

Sarcasm aside -

so·cial (adjective) : offering opportunity for interaction; allowing people to meet and interact with others in a friendly way

The truth of the matter is that social is asked about whenever I discuss a recsys implementation with clients – no matter the vertical. Some opportunities lend themselves well to an articulate social component plus a recsys while still others struggle to find social relativity. Thus, I have proactively endeavored to find concrete strategies in which to use ‘social’ data in productive and meaningful ways, especially with respect to an implementation of a recsys.

Do I Really Want Recommendations From My Friends

We may have 500 friends on Facebook or Google Plus but do we really want each and every one of them driving recommendation decisions for us? Imagine the experience you have every time you visit your wall, how it’s littered mostly with things that aren’t directly or immediately interesting to you. Now imagine that same experience applied to music, books or TV. Simply adding friends to the mix will not add value; more likely it will invoke churn from your service.

Hey friend, what do you like?

There are more informed approaches for incorporating the data from the social graph that can deliver a positive uptick in consumption. One such approach is the representation of a user’s friend set in order to reinforce a recommendation. This means we still rely on a user’s behavior within an ecosystem (supervised learning) to arrive at a recommendation decision but choose to complement the result with a list of friends that have also engaged with the recommended item. In this way we benefit from both machine learning and the social graph – we just don’t use the social graph to drive the decision. Cool.

Filter My Friends

Another interesting approach examines the relative usefulness (conversion percentage) of your friends’ recommendations to you. Observing the success and failure of your friends’ recommendations to you effectively turns them into a channel from which statistical probabilities of success can be rendered.

I like what you picked!

This success metric can be used to rank order future recommendations from ‘high quality’ friends and can cleanly winnow away the noise.

A beneficial side effect of using a qualified friend recommender is the possibility for more prolific experiences of serendipity. The facility to surprise and delight a consumer cannot be underestimated. While we are refining mathematical pathways to successfully discover interesting elements that fall outside the known model of the user, the incorporation of high quality external influencers is a welcome commodity.

Social DOES Have Value

When more than a hand wave toward social is given we can discover many influential uses for social data. The overuse of the term will surely relax over time – I will just have to hold my tongue until the next big buzzword takes its place.