Saturday, June 16, 2012

[G] New In-Ads Notice Label and Icon for the Google Display Network in Europe

| More

Inside AdWords: New In-Ads Notice Label and Icon for the Google Display Network in Europe

Last year in the U.S., we launched the AdChoices Icon on the Google Display Network. AdChoices is an industry standard "i" icon that expands to “AdChoices” when users move their cursor over the icon. Giving users clear notice about the ads they see is a high priority for Google and for many in the online advertising industry at large. Users who click this label are taken to a page where they can learn more about online advertising and the ads they see.

We’re now rolling out the AdChoices feature across Europe, supporting industry-wide efforts to provide more information to consumers. We'll soon start to change our “Ads By Google” in-ads notice icon to a new icon that expands to an "AdChoices" label. These notifications were developed by the Interactive Advertising Bureau (IAB Europe)-led Self-Regulatory Program for Online Behavioral Advertising to proactively give users notice and choice about the ads they see. With the adoption of a common icon users will see on ads across the web, we hope to show our support for the industry-wide initiative and to increase users' understanding about their choices.

Over the next few weeks, we'll start showing the new icon and label on most ad formats across the majority of European-language sites. Over time, we’ll expand the notice to ensure that all European-language publisher sites in the Google Display Network come within the Self-Regulatory Framework. This will be the single largest rollout of the "AdChoices" label to date, and we're encouraged that others in the industry are also adopting the label.

Users who click on the "AdChoices" label will be taken to a page where they can learn more about online advertising and the ads they are shown. This page will also link to the Ads Preferences Manager, where users can control the types of interest-based ads they see on the Google Display Network. We think this rollout will help users better understand the ads they're seeing, and we look forward to seeing widespread adoption of this label throughout the industry.

Posted by Jason Bigler, Product Management Director

[G] In Support of Legislative Transparency

| More

Google Public Policy Blog: In Support of Legislative Transparency

Posted by Seth Webb, Senior Policy Manager

Google believes that policy-making should be grounded in sound analysis of data. We take this to heart -- it’s why we launch tools like the Transparency Report, which shows when and what information is accessible on Google services around the world. Similarly, when governments are transparent with their legislative data, their citizens can be more active participants in the political process.

Last year, for example, the U.S. House of Representatives identified transparency as one of its top priorities, and since then it has taken several steps towards becoming more open. The House now streams and archives video of committee hearings, and it shares draft legislation for public consultation online.

As part of its ongoing effort to promote openness and transparency, the House of Representatives voted for an appropriations bill that directs a task force to examine and expedite the process of disclosing large amounts of legislative data to the public. Even before the bill was passed, Congressional leadership issued a statement on the importance of transparency and requested for the task force to begin its work immediately.

We believe the ability to download bulk legislative data in formats like XML on a regular basis provides tremendous benefits. Website and app developers can use such data to provide up-to-date information on bills. Researchers can use it to perform studies. And politically-curious citizens can use it to follow legislation moving its way through Congress.

We've seen positive transparency efforts throughout the U.S. government. The White House, for one, recently issued a Digital Government Strategy that called for data from offices in the executive branch to be made more easily accessible by application developers.

New information platforms make it easier for the American public to watch and participate in their government, which strengthens the political process as a whole. We applaud Congress for the work that it's done to promote openness and look forward to a future of increased legislative transparency.

Friday, June 15, 2012

[G] Third Market Algorithms and Optimization Workshop at Google NYC

| More

Research Blog: Third Market Algorithms and Optimization Workshop at Google NYC

Posted by Nitish Korula and Vahab Mirrokni, Google Research, New York

There are fascinating algorithmic and game theoretic challenges in designing both Google’s internal systems and our core products facing hundreds of millions of users. For example, both Google AdWords and the Ad Exchange run billions of auctions a day; showing the perfect ad to every user requires simple mechanisms to align incentives while simultaneously optimizing efficiency and revenue.

We think that research in these areas benefits from close cooperation between academia and industry. To this end, last week we held the Third Market Algorithms and Optimization Workshop at Google, immediately after STOC 2012. We invited several leading academics in these fields to meet with researchers and engineers at Google for a day of talks and discussions.

As a recent winner of the Godel prize, Éva Tardos from Cornell led off with a discussion of how to achieve efficiency in sequential auctions where bidders arrive and depart one at a time instead of all bidding simultaneously.

Eyal Manor, Google engineering director for the Ad Exchange, gave an overview of the design and functioning of the exchange. This was an opportunity to have questions answered by the absolute expert, and the participants took full advantage of it!

Costis Daskalakis and Pablo Azar from MIT and Tim Roughgarden from Stanford talked about different aspects of Optimal Auctions in Bayesian Settings. Costis talked about efficient implementation of optimal auctions in a class of combinatorial auctions. Both Tim and Pablo discussed optimal auctions in Bayesian settings with limited information. Tim, our other Godel prize winner, promoted the idea of designing simple auction rules that are independent of the distributions of buyers’ valuations, and Pablo presented optimal auction rules using only the mean and standard deviation of buyers’ valuations.

Bobby Kleinberg from Cornell and Gagan Goel from Google NYC presented recent work on pricing with budget constraints. Bobby’s talk was about procurement auctions where the auctioneer acts as a buyer with a budget constraining her procurements. Gagan, on the other hand, discussed Pareto-optimal ascending auctions where the auctioneer is selling to budget-constrained buyers. This has direct applications in Google AdWords auctions as advertisers aim to increase performance while staying within budget constraints.

With our mission of organizing all the world’s information, Google needs superior algorithmic techniques to analyze extremely large data sets. We had two talks on new algorithmic ideas for Big Data. From academia, Andrew McGregor gave an introduction to the new field of graph sketching. Though a graph on n nodes is O(n^2)-dimensional, Andy described how to find interesting properties of the graph (such as connectivity, approximate Minimum Spanning Trees, etc.) using only O(n polylog(n)) bits of information. These algorithms were based on clever use of the homomorphic properties of random projections of the graph’s adjacency matrix. In the next talk, Mohammad Mahdian from Google MTV explained a new model for evolving data; even a ‘simple’ problem like sorting becomes interesting when the order of elements changes over time. Mohammad showed that even if element swaps occur at the same rate as comparisons, one can compute an ordering with Kendall-Tau distance O(n ln ln n) from the true ordering at any time, very close to the optimal Ω(n).

Later, Mukund Sundararajan from Google MTV discussed algorithmic problems in interpreting and presenting sales data to advertisers. He challenged us to design flexible human-friendly optimization algorithms that can be adopted and tuned by humans. Toward the end of the workshop, Varun Gupta, Google NYC postdoctoral researcher, gave a short presentation about the use of primal-dual techniques for online stochastic bin packing with application in assigning jobs to data centers.

We also discussed some of the main activities in the algorithms research group in New York, like the use of primal-dual techniques in online stochastic display ad allocation at Google and large-scale graph mining techniques based on MapReduce and Pregel. Corinna Cortes, Director of Research in New York, and Alfred Spector, VP of Research and Special Projects, gave short speeches. Corinna talked about our statistics, machine learning, and NLP research groups in New York, and Alfred challenged us to design mechanisms to take into account fairness in allocations and pricing. For more details, see the blog post by our colleague, ‘Muthu’ Muthukrishnan.

Part of what makes Google a fascinating place to work is the wealth of algorithmic and economic research challenges posed by Google advertising and large-scale data analysis systems. These challenges define research directions for the computer science and economics research communities. Workshops like this and our weekly research seminars help us continue collaborations between Google and academia. We hope to post videos of this workshop shortly, and look forward to organizing many more such events in the future.

[G] Calling all Beliebers! Ask Justin a question on YouTube Presents

| More

YouTube Blog: Calling all Beliebers! Ask Justin a question on YouTube Presents

You’ve seen him grow up on camera, transforming from a talented little kid posting videos on YouTube into the star of his own documentary and a global pop phenomenon. Now Justin Bieber is ready to come back to where it all began for an exclusive Q&A with YouTube Presents, and he wants to answer your questions.

What have you always wanted to know about the Biebs? Starting today, and running until midnight ET on June 20, you have the chance to ask Justin Bieber all those burning questions. Head over to and pour your heart out—Justin will hear you! And be ready to tune in to on June 21 at 7pm ET to watch the interview -- with none other than Jimmy Fallon -- live.

YouTube Presents gives music fans an opportunity to get more into the music and artists they love with live performances and interviews at venues around the world. Check out recent performances and concerts with artists including Taylor Swift, The Shins, Ziggy Marley, Lenny Kravitz and more on

And remember -- never say never!

Sarah Bardeen, music community manager, recently watched “Justin Bieber’s Backstage Interview at Capital FM’s Summertime Ball 2012.”


[G] New In-Ads Notice Label and Icon for the Google Display Network in Europe

| More

Inside AdSense: New In-Ads Notice Label and Icon for the Google Display Network in Europe

(Cross posted to the Inside AdWords Blog

Last year in the U.S., we launched the AdChoices Icon on the Google Display Network. AdChoices is an industry standard "i" icon that expands to “AdChoices” when users move their cursor over the icon. Giving users clear notice about the ads they see is a high priority for Google and for many in the online advertising industry at large. Users who click this label are taken to a page where they can learn more about online advertising and the ads they see.

We’re now rolling out the AdChoices feature across Europe, supporting industry-wide efforts to provide more information to consumers.  We'll soon start to change our “Ads By Google” in-ads notice icon to a new icon that expands to an "AdChoices" label. These notifications were developed by the Interactive Advertising Bureau (IAB Europe)-led Self-Regulatory Program for Online Behavioral Advertising to proactively give users notice and choice about the ads they see. With the adoption of a common icon users will see on ads across the web, we hope to show our support for the industry-wide initiative and to increase users' understanding about their choices.

Over the next few weeks, we'll start showing the new icon and label on most ad formats across the majority of European-language sites. Over time, we’ll expand the notice to ensure that all European-language publisher sites in the Google Display Network come within the Self-Regulatory Framework. This will be the single largest rollout of the "AdChoices" label to date, and we're encouraged that others in the industry are also adopting the label.

Users who click on the "AdChoices" label will be taken to a page where they can learn more about online advertising and the ads they are shown. This page will also link to the Ads Preferences Manager, where users can control the types of interest-based ads they see on the Google Display Network. Our tests of this new icon and label showed that they should not have any effect on ad performance. We think this rollout will help users better understand the ads they're seeing, and we look forward to seeing widespread adoption of this label throughout the industry.

Posted by Jason Bigler - Product Management Director


[G] YouTube automatic captions now available in Spanish

| More

YouTube Blog: YouTube automatic captions now available in Spanish

Cross posted from the Blog de YouTube en Español

Last year, YouTube had more than 1 trillion views, or about 140 views for every person on earth. As the world tunes in to YouTube, we want everyone, in every language, to have the same opportunity to enjoy YouTube. So today, we’re expanding our language accessibility to add automatic captions in Spanish.

When a video has recognizable speech, you’ll see a “CC” button appear in the bottom of the player, which will instantly add captions of the video in Spanish. Just look for this icon and click “Transcribe Audio.”

The hundreds of millions of Spanish speakers in the world are the latest to see the auto-caption feature, adding to other available languages of English, Japanese and Korean. You’ll find auto-captions available on more than 157 million videos, with videos being added every day. We’ll continue to refine our speech recognition technology, and you can learn more about how it works here. See it in action on this video, by clicking the CC:

If you want to see YouTube videos in even more languages, you can combine auto-captions with our auto-translate feature to generate subtitles in more than 50 languages. For creators, upload a Spanish transcript with your video and we’ll automatically create timecoded captions. You can even download the automatic captions, all from your Video Manager.

We’re launching new countries and languages all the time, as we work to make YouTube accessible and enjoyable to all.

¡Nos vemos en YouTube!

Hoang Nguyen, software engineer, recently watched "Casillas: 'Si la Eurocopa hubiese sido en 2011, habría habido más problemas.'"


[G] 10 top Dutch universities adopt Google Apps for Education

| More

Official Google Enterprise Blog: 10 top Dutch universities adopt Google Apps for Education

Posted by Grainne Phelan, Google Apps for Education Team

The sun was shining in Amsterdam and the tulips were in blossom today as 10 top universities in the Netherlands came together to announce a group migration to Google Apps for Education. This group movement, which was negotiated with the help of SURF, the Dutch higher education and research partnership for ICT-driven innovation, represents the second example of a national “wave” of universities making the transition to cloud computing together.

This group of Dutch universities join a community of schools and universities across the world who have adopted Google Apps for Education to help stay ahead of the curve in the education sector. 66 of the top 100 universities in the United States have already adopted Google Apps, and other top institutions across Europe, including the University of Warsaw and University of Sapienza in Rome, have recently transitioned to Google Apps for Education.

The universities that have gone Google today are: Open University; University of Groningen; University of Amsterdam; Utrecht University; Tilburg University; University of Twente; Delft University of Technology; Avans; HZ University of Applied Sciences; and NHL. The educational research organization TERENA (Trans-European Research and Education Networking Association) also announced their migration today.

The University of Tilburg previously used Microsoft’s cloud solution, and transitioned to Google Apps for Education for staff and students at the beginning of June. Their IT director, Corno Vromans, is feeling positive:
“With the transition to Google Apps, the whole academic community, students and staff, have a new, modern working environment. With Google Apps, regardless of time and place, users can access their files and they can share and collaborate simultaneously. We are very interested in innovation and collaboration, so with Google Apps we are ready for the future."
Congratulations to the all the Dutch universities as they embark on their new journey with Google, and we look forward to celebrating the next wave of universities to join the club.

[G] Have your own YouTube party in Google+ Hangouts

| More

YouTube Blog: Have your own YouTube party in Google+ Hangouts

Ever been to a party where you start showing friends your favorite YouTube videos? Compete for the screen to see who knows the latest and greatest hit? Now you can get the same experience even if you’re a world away, with the new YouTube app in Google+ Hangouts.

Create, control, save and even share playlists of videos with your friends—all inside a Hangout. Just start a Hangout, have everyone load the YouTube app at the top of the screen, and start adding videos. It’s like your own VIP table at the world’s coolest YouTube party.

Crowdsourcing awesomeness

Everyone can add videos in the Hangout through a search tool in the app, or remove the videos you don’t like. All your friends in the Hangout can drag and drop videos to sort the order in the playlist, or skip forward or backward to play the next one. Click the “Push to talk” button to chat with the group to give props to the best curator, or to hand out reprimands to the friend who keeps adding the 10-hour Nyan Cat video.

Save and share those precious memories

If you like a video that’s playing, you can share the video with your Google+ circles at any time. Since great playlists are works of art, you can also save the playlist that you and your friends created to your YouTube account as public or private to enjoy later.

So gather the team to watch the last night’s game highlights, hangout with your friends for a haul-a-thon, assemble the audiophiles to review the newest tunes, or come up with a whole new way to enjoy (and let us know!). The new app is available worldwide in 60 languages.

Ullas Gargi, software engineer, recently watched “Ken Block Gymkhana Practice.”


Thursday, June 14, 2012

[G] Customize account management with AdWords scripts

| More

Inside AdWords: Customize account management with AdWords scripts

We know that you're always looking for more efficient ways to manage your AdWords accounts. We also hear that many of you would love to use your own external data to make decisions in your accounts. Today, we're excited to announce the limited release of scripts in AdWords. With scripts, you can make changes to an AdWords account by writing simple JavaScript programs.

With scripts, you can:
  • Use external inventory data to either change bids or pause/unpause keywords. 
  • Output account statistics to a spreadsheet from which you can create reports and visualizations. 
  • Use stats trends over several weeks to change keyword or ad group bids.
Scripts run in the Google Apps Script infrastructure, and we're enabling Apps Script integration with three services: you can use Google Spreadsheets and Url Fetcher to integrate with external data, as well as email results of a script execution from the script itself.

To get started with scripts, visit the Campaigns tab, and select “Automation” then “Scripts” from the left toolbar:

To make the most of scripts, we recommend that whoever is writing them has a basic understanding of JavaScript.

In the coming weeks we plan to enable this feature for everyone. In the interim, we are accepting applications for access. Applicants will be whitelisted in batches as we ramp up usage in the system. In order to apply, you will need an AdWords customer ID (not a My Client Center ID). Please refer to our developers site for more instructions and best practices for using scripts.

Posted By: Prashant Baheti, AdWords Product Manager

[G] Take advantage of mobile education opportunities

| More

Inside AdWords: Take advantage of mobile education opportunities

Last week, in partnership with the Learn With Google program, we held our first “Mobile Education Week” webinar series for your agency or business to learn more about the tools and resources to help you succeed on mobile.

Below are video recordings of these webinars - check them out, and share with your colleagues.

Our Mobile Planet: Understanding U.S. Smartphone Consumers
Our first webinar provided an overview of the 2012 Our Mobile Planet smartphone research. Watch this video if you’d like to learn:
  • How smartphones are transforming core consumer behavior around the world, and the way we connect with others, stay informed, keep ourselves entertained, shop and navigate the world around us.
  • Implications for advertisers and strategies businesses can use to win the moments that matter with mobile.
Ideal for: global agencies and marketers

GoMo: Mobilize your Site with Quickly and Easily New Tools from Google & DudaMobile
The second webinar was a joint effort between Google’s GoMo initiative and DudaMobile on how it is now even easier for small businesses to create a free mobile-friendly website.

View this video if you’d like to learn:
  • Why mobile sites matter
  • Review Google's resource
  • New: How to create free mobile sites for your small business directly from, using a tool powered by DudaMobile
  • Examples and best practices for mobilizing your site using the free DudaMobile tool
Ideal for: small businesses and agencies interested in free tools to create a mobile site

Introducing Mobile Apps Inventory in AdWords
The final webinar shared the new opportunity available with the availability of AdMob inventory to AdWords users.

This video is great if you’d like to learn:
  • How AdWords now makes it easy for you to extend your campaigns to reach users inside mobile apps.
  • Review mobile-specific targeting, ad units, and reporting than enable you to optimize your ROI.
Ideal for: marketers and agencies looking to promote and target mobile apps

Want to learn more about how to help your business or agency succeed on mobile? Stay tuned to all new Mobile Ads news at the Google Mobile Ads blog and YouTube channel.

Posted by Sonja Lee, Product Marketing, Mobile Ads

[G] Chrome and the New Shiny

| More

Google Chrome Blog: Chrome and the New Shiny

You always want Chrome to look great, no matter what device you’re using. Apple recently announced a new laptop with a Retina high-resolution screen, and we’re committed to polishing Chrome until it shines on that machine.

The Chrome Canary channel already shows the early results of this work, bringing basic high-resolution support to Chrome. We have further to go over the next few weeks, but we’re off to the races to make Chrome as beautiful as it can be.

Posted by Nico Weber, Software Engineer and Chief Apple Polisher

[G] Ads Integrity Alliance: Working together to fight bad ads

| More

Google Public Policy Blog: Ads Integrity Alliance: Working together to fight bad ads

Posted by Eric Davis, Global Public Policy Manager

Today StopBadware is announcing the formation of an industry partnership to combat bad ads. We’re pleased to be a founding member of the Ads Integrity Alliance, along with AOL, Facebook, Twitter and the IAB.

Since its beginnings in 2006, StopBadware has enabled many websites, service providers and software providers to share real-time information in order to warn users and significantly eliminate malware (such as viruses, phishing sites and malicious downloads) on the web. We believe that the Ads Integrity Alliance can make a similarly important contribution to the goal of identifying and removing bad ads from all corners of the web.

In 2011, Google alone disabled more than 130 million ads and 800,000 advertisers that violated our policies on our own and partners’ sites, such as ads that promote counterfeit goods and malware. You can read more about our efforts to review ads and also see the numbers over time. Other players in the industry also have significant initiatives in this area. But when Google or another website shuts down a bad actor, that scammer often simply tries to advertise elsewhere.

No individual business or law enforcement agency can single-handedly eliminate these bad actors from the entire web. As StopBadware has shown, the best way to tackle common problems across a highly interconnected web, and to move the whole web forward, is for the industry to work together, build best practices and systems, and make information sharing simple.

The alliance led by StopBadware will help the industry fight back together against scammers and bad actors. In particular, it will:
  • Develop and share definitions, industry policy recommendations and best practices
  • Serve as a platform for sharing information about bad actors
  • Share relevant trends with policymakers and law enforcement agencies
Bad ads reduce trust in the web and in online advertising. The web puts the world’s information at your fingertips and has given everyone a platform to speak, listen, engage and unite. The growth that businesses generate from online advertising has enabled an enormous part of this platform. We think the web is worth fighting for, which is why we strongly support the Ads Integrity Alliance’s efforts to tackle bad actors who seek to damage it.

(Cross-posted from the Official Google Blog)

[G] AdSense now available for websites in Slovenian and Estonian

| More

Inside AdSense: AdSense now available for websites in Slovenian and Estonian

If you have a website in Slovenian or Estonian, you'll now be able to earn money through the AdSense program. To get started, review the basics of AdSense and our policies, then sign up for a free account. Already an active AdSense publisher? Simply place the AdSense ad code on your sites in Slovenian or Estonian to start showing targeted ads.

If your application for AdSense with a Slovenian or Estonian website was previously disapproved for being in an unsupported language, you’re welcome to resubmit it by logging in with the original login details you provided.

For now, you’ll be able to access the interface and help resources in English, as we heard from your feedback that you’d prefer to begin using the program in the meantime. Rest assured that we’re working to provide help content in these languages in the near future.

We look forward to welcoming our Slovenian and Estonian publishers to the AdSense program!

Posted by Ulrike Jung -- Product Sales Lead


[G] Recap of NAACL-12 including two Best Paper awards for Googlers

| More

Research Blog: Recap of NAACL-12 including two Best Paper awards for Googlers

Posted by Ryan McDonald, Research Scientist, Google Research

This past week, researchers from across the world descended on Montreal for the Conference of the North American Chapter of the Association for Computational Linguistics (NAACL). NAACL, as with other Association for Computational Linguistics meetings (ACL), is a premier meeting for researchers who study natural language processing (NLP). This includes applications such as machine translation and sentiment analysis, but also low-level language technologies such as the automatic analysis of morphology, syntax, semantics and discourse.

Like many applied fields in computer science, NLP underwent a transformation in the mid ‘90s from a primarily rule- and knowledge-based discipline to one whose methods are predominantly statistical and leverage advances in large data and machine learning. This trend continues at NAACL. Two common themes dealt with a historical deficiency of machine-learned NLP systems -- that they require expensive and difficult-to-obtain annotated data in order to achieve high accuracies. To this end, there were a number of studies on unsupervised and weakly-supervised learning for NLP systems, which aim to learn from large corpora containing little to no linguistic annotations, instead relying only on observed regularities in the data or easily obtainable annotations. This typically led to much talk during the question periods about how reliable it might be to use services such as Mechanical Turk to get the detailed annotations needed for difficult language prediction tasks. Multilinguality in statistical systems also appeared to be a common theme as researchers have continued to move their focus from building systems for resource-rich languages (e.g., English) to building systems for the rest of the world’s languages, many of which do not have any annotated resources. Work here included focused studies on single languages to studies aiming to develop techniques for a wide variety of languages leveraging morphology, parallel data and regularities across closely-related languages.

There was also an abundance of papers on text analysis for non-traditional domains. This includes the now standard tracks on sentiment analysis, but combined with this, a new focus on social-media, and in particular NLP for microblogs. There was even a paper on predicting whether a given bill will pass committee in the U.S. Congress based on the text of the bill. The presentation of this paper included the entire video on how a bill becomes a law.

There were two keynote talks. The first talk by Ed Hovy of the Information Sciences Institute of the University of Southern California was on “A New Semantics: Merging Propositional and Distributional Information.” Prof. Hovy gave his insights into the challenge of bringing together distributional (statistical) lexical-semantics and compositional semantics, which has been a need espoused recently by many leaders in the field. The second, by James W. Pennebaker, was called “A, is, I, and, the: How our smallest words reveal the most about who we are.” As a psychologist, Prof. Pennebaker represented the “outsider” keynote that typically draws a lot of interest from the audience, and he did not disappoint. Prof. Pennebaker spoke about how the use of function words can provide interesting social observations. One example was personal pronouns like “we,” whose increased usage now causes people to feel the speaker is colder and more distant as opposed to engaging the audience and making them appear accessible. This is partly due to a second and increasingly more common meaning of “we” that is much more like “you,” e.g., when a boss says: “We must increase sales”.

Finally, this year the organizers of NAACL decided to do something new called “NLP Idol.” The idea was to have four senior researchers in the community select a paper from the past that they think will have (or should have) more impact on future directions of NLP research. The idea is to pluck a paper from obscurity and bring it to the limelight. Each researcher presented their case and three judges gave feedback American Idol-style, with Brian Roark hosting a la Ryan Seacrest. The winner was "PAM - A Program That Infers Intentions," published in Inside Computer Understanding in 1981 by Robert Wilensky, which was selected and presented by Ray Mooney. PAM (“Plan Applier Mechanism”) was a system for understanding agents and their plans, and more generally, what is happening in a discourse and why. Some of the questions that PAM could answer were astonishing, which reminded the audience (or me at least) that while statistical methods have brought NLP broader coverage, this is often at the loss of specificity and deep knowledge representation that previous closed-world language understanding systems could achieve. This echoed sentiments in Prof. Hovy’s invited talk.

Ever since the early days of Google, Googlers have had a presence at NAACL and other ACL-affiliated events. NAACL this year was no different. Googlers authored three papers at the conference, one of which merited the conference’s Best Full Paper Award, and the other the Best Student Paper:

Cross-lingual Word Clusters for Direct Transfer of Linguistic Structure - IBM Best Student Paper
Award Oscar Täckström (Google intern), Ryan McDonald (Googler), Jakob Uszkoreit (Googler)

Vine Pruning for Efficient Multi-Pass Dependency Parsing - Best Full Paper Award
Alexander Rush (Google intern) and Slav Petrov (Googler)

Unsupervised Translation Sense Clustering
Mohit Bansal (Google intern), John DeNero (Googler), Dekang Lin (Googler)

Many Googlers were also active participants in the NAACL workshops, June 7 - 8:

Computational Linguistics for Literature 
David Elson (Googler), Anna Kazantseva, Rada Mihalcea, Stan Szpakowicz

Automatic Knowledge Base Construction/Workshop on Web-scale Knowledge Extraction
Invited Speaker - Fernando Pereira, Research Director (Googler)

Workshop on Inducing Linguistic Structure
Accepted Paper - Capitalization Cues Improve Dependency Grammar Induction
Valentin I. Spitkovsky (Googler), Hiyan Alshawi (Googler) and Daniel Jurafsky

Workshop on Statistical Machine TranslationProgram
Committee members - Keith Hall, Shankar Kumar, Zhifei Li, Klaus Macherey, Wolfgang Macherey, Bob Moore, Roy Tromble, Jakob Uszkoreit, Peng Xu, Richard Zens, Hao Zhang (Googlers) 

Workshop on the Future of Language Modeling for HLT 
Invited Speaker - Language Modeling at Google, Shankar Kumar (Googler)
Accepted Paper - Large-scale discriminative language model reranking for voice-search
Preethi Jyothi, Leif Johnson (Googler), Ciprian Chelba (Googler) and Brian Strope (Googler)

First Workshop on Syntactic Analysis of Non-Canonical Language
Invited Speaker - Keith Hall (Googler)
Shared Task Organizers - Slav Petrov, Ryan McDonald (Googlers)

Evaluation Metrics and System Comparison for Automatic Summarization
Program Committee member - Katja Filippova (Googler)

[G] GA Advocate Justin Cutroni Answers Your Analytics Questions

| More

Google Analytics Blog: GA Advocate Justin Cutroni Answers Your Analytics Questions

The following post originally appeared on Justin Cutroni’s Analytics Talk blog.

There are a lot of GA users. As a matter of fact, in Google’s Q1 2012 earnings call it was revealed that GA is being used on 10MM sites. That’s a lot of data and and a lot users!

Those users generate a lot of questions. Recently, I solicited questions on Google+. I hope you find the answer useful in your daily use of Google Analytics.

Please note, if I did not get to your question it was for one of two reasons:

1. It was posted after I started writing the post

2. It may have been a bit too specific. I contacted a few people directly about those questions.

On to the answers!

I'd love to use the multi-channel report in GA, but the 30-day cookie doesn't work for websites who offer a 30-day trial and want to track all of a customer's touch points through free trial and subscription. What suggestion do you have for recording campaign touch points outside of the 30-day window?

Justin’s Answer

This is a great question, and we hear that request a lot. The Multi-Channel funnel reports, and the attribution modeling tool, both use a 30-day lookback window. In reality this has nothing to do with a cookie, it’s how Google Analytics processes the data on the back end.

If you’re looking to identify activity that happened 30-plus days prior to conversion you need to work outside the bounds of Multi-Channel funnels and create something that stores activity and date. You have a couple of options: custom variables or events. 

The easiest way is to use a visitor scoped Custom Variable. Store some type of marketing-touchpoint list in the CV. Then use the Custom Variable reports to look at which paths generated conversions. The hard part is you’ll need to update the custom var with referral info  on every visit. This means custom JavaScript to update the cookie.

Another option is to use an event, If you already have some way to identify your visitors across sessions consider storing the referral information in your system. Then push out some events that list all of the touch points when the final conversion happens. This technique requires a lot more server side code.

I have data coming from a number of different sources so i'm trying to tape together the best possible picture I can of a multi-touch universe. I have a number for the total interactions via a campaign (including some content campaigns which an interaction is an impression) and I want to create an influencer metric using a combination of last touch point / total interactions and first touch point. I am working on a number of full attribution models, but in the meantime. Using the 3 numbers I have (first touch point conversion / Last touch point conversions / total interaction conversion) how would you come to a 'nice' metric that gave an indication of the 'importance' of a given campaign.

I have something in mind - but wanted to pick the brain of some bright things :)

Justin’s Answer

What a fantastic question. 

I think I would do it the same way you are: using ratios. If you’re looking for one number to represent the importance of a campaign, based on the number of first/last/total interactions, I would use a ratio. 

First Touch Influence = First Touch Conversions / Total Conversions

Last Touch Influence = Last Touch Conversions / Total Conversions

You've probably noticed that this is almost the same way that Google Analytics calculates it's assisted/last ratio. But it's simple and easy to understand. Plus, depending on the data you have available, you could also segment these metrics.

You can also create a benchmarks internally using un-segmented data or historical campaign performance. I usually don’t use a lot of compound, custom metrics, but this is fairly easy for anyone to understand.


Is it possible to get back city-state-country e-commerce data without the usage of API? Because we want to use these to track different type of buyers.

Justin’s Answer 

Unfortunately no. These dimensions do not exist in the UI. Also, a quick note, that Google will be deprecating the Data Export API on July 10, 2012. The Core Export does NOT contain the ecommerce geo-dimensions. Perhaps you use custom variables to collect the information that you need.

Setting up many goals is supported, even encouraged. What would you say is a good practice to divide the less important goals (clicking on something, a certain time on site) from the core business ones? (sales, lead generation), so the data doesn't get polluted. Thanks in advance!!

Justin’s Answer

I’m a neat-freak! I like things organized. So I would say yes. If you can group your macro conversions into one goal set, and your micro conversions into another goal set, it would make using GA easier.

"You can organize macro and micro conversions in Google Analytics using Goal Sets."

HOWEVER you do have the option to create custom reports. And when you make a custom report the goal sets don’t matter! So if your micro and macro conversions are a complete mess try using a custom report to organize things. 

I am trying to see how many hits I'm getting against my pages. The catch is that many of my pages are passed a query part in the URL, and I am completely uninterested in this query value. The way things appear to be working is that for each different query parameter, the page is counted as a different page. So the following are all currently reported as different, but I want them reported as the same page:




Even more than the answer, however, I want to know where in the documentation I should have been able to figure this out.


Justin’s Answer

You’re looking for pageviews, which is a very different thing.

Query string parameters are such a pain! I hate it when they magically start showing up in a report. Use the Exclude Query Parameter setting in Google Analytics. Simply enter a comma-separated list of query parameters and GA will strip them out of your data. You only need to enter the name of the parameter, not the value.

Use the Exclude Query String Parameters to remove unwanted query parameters from your content reports.

If you don’t know the names of the parameters, or if they are constantly changing, you might consider an advanced  filter. This is the nuclear option :) An advanced filter will strip off all the parameters, all the time, no matter what they’re named.

You can also use an advanced filter to remove all query string parameters from your content reports.

Hi. I use Google Analytics, and for some reason I get different results when I access in my office and in my home. What explains this discrepancy?

Justin’s Answer

That’ a tough one. I really can’t explain why you would see different data. Once the data has been collected and processed it does not change. My only suggestion is to make sure you are looking at the exact same profile. You might be looking at different profiles, thus seeing different data.

Do you have a post with a list of the different dashboards that you can "plug and play" ?

Justin’s Answer

You’re in luck! Here’s a list of a few dashboards you can add to Google Analytics

How does cross-domain tracking work in Google Analytics? Specifically, after putting the correct additions (trackDomain) to the Google Analytics tracking code, what does cross-domain tracking look like in the GA reports? We have clients that want this working for their sub-domain and their top-level domain ( &

Justin’s Answer

Sub-domain and cross domain tracking are two very different things! Check out this article to read about the finer points of cross domain tracking and sub-domain tracking.

As for how the data looks in Google Analytics, there’s really no difference. You’ll notice the sub domain or the secondary domain in the Audience > Technology > Network > Hostname report. And you should see all of the pages from both domains in the Content reports. 

I usually add an advanced filter to add the domain name to the content reports. This makes it easier for me to identify pages on different sites. If you need to separate the data you can create different profiles based on the hostname or used Advanced Segments.

Jon Darch asks:

This might be a stupid question, but when setting up a custom dashboard, how do I create widgets which show a metric (i.e visitors for the last 30 days) with the previous month's figure as a % up or down? I'm sure I've seen others doing this, but can't seem to figure it out! 

If you are able to offer any advice, that would be much appreciated :)



Justin’s Answer

Unfortunately you cannot add a “sticky” date to the dashboard. But I wish you could! You can manually do a date comparison, then you’ll see a % change in some of the widgets, like the tabular widget.

But stay tuned, we might have a better solution for that.

Hey Justin,

When comparing 2 date ranges in the adwords reports, the calculation for 'change in ROI' is misleading/incorrect if the ROI value for either the first or second date range is Negative. 

(for eg, week 1 has -10% ROI, week 2 has +30% ROI; in this case, the calculated '% difference' is -400%, however, I just turned a loss into a profit)

In this scenario, what alternative solution/calculation do you think is more apt? Also, does the GA team have any plans to use a more accurate calculation or even put a warning note against a scenario like this?


Justin’s Answer

Thanks for the heads up! I completely agree that this is not correct. We’re working to fix it. Stay tuned for an update.

Nice one. I created a goal with funnel visualization and few days later. I realized was a wrong goal. I created a new goal so how do I delete the old goal and it's visualization?

Justin’s Answer

Unfortunately you can not delete data from Google Analytics. Once the data has been collected and processed by our system it’s static forever. I would suggest de-activating the goal for a few weeks to ‘clear’ the data. Then add an annotation to remind everyone that there is some bad goal data in the reports.

Hello Justin,

When creating goals (taget URL), how to account for different routes through to the same target?



Justin’s Answer

Different paths to the goal are handled using a Funnel. When you create your goal you can also create a funnel to see how many people follow the defined path and how many people take other paths. The Goal Flow report will help you see people moving in and out of each step.

If you have multiple paths to conversion, and you want to get a sense of how people move through each pathing, you may consider creating a goal, with a different funnel, for each path. It’s easier to separate the data for analysis.

If your goal does not have a defined path you can use the Reverse Goal Path report to view the 3 steps prior to every conversion. Or try using the Flow Visualization report to explore other paths to conversion.

How would you go about investigating (or have any previous examples of) why a site appears to double count visits. Almost exactly 50% of visits have no landing page set and no pageview information and I am sure they are not real visits but something to do with how the site is set up.

Justin’s Answer

This is normally due to events or other hit types. The visit metric is incremented on the first hit of a visit. If the first hit is an event, and there are no other hits, then you would see lots of visits with no landing page or pageviews. So go check for some rogue events.

What do you think is the best practice for adding a mobile web site to the collection of sites/apps we track in GA? Should a site have its own UA-code, be a new property under UA-code, or just be rolled into Right now we track mobile apps separately from the website, but adding an m. site is not as straightforward.

Justin’s Answer

Before I get to the answer, a quick note on terminology. We use the term ‘web property’ to represent a unique Google Analytics tracking code. This is analogous to a UA number. So UA-1 and UA-2 are web property IDs. 

Each web property can have multiple profiles. A profile is a combination of data from a web property and settings applied to the profile. So UA-1-1 and UA-1-2 are both profiles for web property UA-1.

Now the answer!

I think yes, you should separate your mobile site into a new web property. The user experience for a mobile-optimized site is usually very different than a www site or a mobile app. As a result I would separate that data into a new web property so it’s easier to understand the behavior. If you need to combine the data from the mobile site with other data then you might consider using the API.

We have developed a mobile website & implement tracking code for mobile website

Now we are checking referral sources & found that our mobile website is showing as self referral.

We have verified the same using Firebug & value of UTMR is showing that referral is our own mobile website.

Any help from the community is very helpful for us.

Justin’s Answer

Referral information for the server side code (ie the WAP tracking code) uses server information to include the referral. For example, if you are using the PHP code then the GA uses $_SERVER["HTTP_REFERER"] to identify where the visitor came from. My guess is that there is some issue with that server variable and that’s why you’re not getting valid data send to Google Analytics.

I came across an issue about real time data of Google Analytics.

I am browsing some mobile websites using iPhone, iPad, Blackberry phones & at the same time when i am checking their real time data in Google Analytics Location is showing as United States, where as i am browsing from India.

Justin’s Answer

I've seen this too, and I've always assumed it's an artifact of where the mobile network connects to the public internet. I would say that, for some reason, the routing of your connection is changing the geo-location to the US, rather than India. 

I'd really like to know why it seems that I still can't create a profile that only includes traffic and transactions from a particular sub-domain (www vs www2).

Transactions include those from all domains, and the hostname is always still (not set).

Using a profile that filters according to transaction affiliation half works, since it shows $ numbers for that affiliate (each sub-domain has its own affiliate), but it also shows 0$ transactions for the other sub-domain - and it only shows traffic for visitors who convert. Annoying.

Other question - might as well abuse of your offer :)

Ever since I showed people how to use campaign tracking, they have been using it to track clicks from banners on the homepage to other pages on the same site. Now I've always been convinced that the best use for this is tracking outside referrers (emails, banners, etc). Would there be a better way for them to track the clicks on these internal pages (this works reasonnably well because they are able to modify the URLs themselves in the CMS)?

Justin’s Answer

You’re correct re: filtering transactions. The hostname dimension is NOT attached to transactional data. That’s why you can not filter transaction based on hostname. For a better solution, try adding an identifier to the transaction ID and filter based on the transaction ID, which is part of the transaction and product data. But we’re in the process of fixing this. I know it’s a huge hassle, sorry.

As to your other question, there are a couple of ways to track internal campaigns. You could use event tracking, but I like to re-purpose Site Search to track internal campaigns. Check out the article, it’s pretty easy to implement.

When are service accounts coming to the Google Analytics API? It's still way too hard to do something simple like have a web server pull a top 10 most viewed content list in JSON.

Justin’s Answer

While I can’t comment on what we’re working on, that sounds like a great idea. Let me see what we can do. And thanks for the awesome suggestion!

Hi Justin and thanks so much for the initiative. I just want to know when will GA share more tutorials on goal analysis.

Justin’s Answer

I love that you’re so focused on conversion analysis! Google’s definitely focused on launching more and more educational materials. You can start with our Introduction to Google Analytics webinar, the Goals configuration webinar, as well as the multi-channel analysis webinar (watch the official Google Analytics blog for the YouTube video). There's a lot of stuff on our YouTube channel. And we're working on new ways to create a better learning experience for users.

<iframe width="420" height="315" src="" frameborder="0" allowfullscreen></iframe>

How can I see a full report of the most popular time of day (hours with the most visits) on my websites?

Justin’s Answer

Use an Overview report. Then, look for the Hourly graphing option under the Date Selector. Here’s a screenshot.

How to graph traffic by Hour of the day. 

That's it.

Thanks everyone! Those were great. What a variety of questions.

That’s the first installment of Analytics Q & A. Stay tuned, we’ll do this again next month.

Posted by Google Analytics Advocate Justin Cutroni