Google BigQuery gets smarter with large results, productivity improvements, and updated pricing

Posted by Ju-kay Kwek, BigQuery Product Manager

We know that today, more than ever, businesses need ways to store and rapidly analyze vast amounts of data and are looking for ways to accomplish this without huge infrastructure investments. To help make this possible, recent BigQuery features include the ability to join across multi-terabyte tables, and the ability to connect popular analysis tools such as Tableau®, BIME® and Excel®. In the past few months we’ve seen several interesting use cases enabled -- Shutterfly improving their customers’ experience, Gamesys understanding complex user behaviors, tracking and mapping the world’s ships, and monitoring Google I/O 2013 using a real-time sensor network.

Today we’re announcing another update to BigQuery packed with new capabilities.

• Large results: run queries that return arbitrarily large numbers of rows and save them as a new table for follow-up analysis. 
• Window functions: take advantage of built-in functions like Rank and Partition to create sophisticated statistical analyses with far simpler SQL than before. 
• Query caching: now recent queries that are re-run return a cached result when the underlying table is unchanged, providing more cost-effective analysis.

Gamesys, who previewed these features, was able to efficiently identify different cohorts of gaming customers and understand how to create a better in-game experience for distinct groups of users. "Rank and Partition are our 'go to' functions for examining player behaviour over time”, said Tom Newton, Director of Social Gaming at Gamesys. “These new functions combined with large results sets and query caching, help us efficiently and cost effectively improve and scale our analysis to create actionable intelligence that drives product enhancement.”

We’ve also rolled out a host of UI improvements, including the ability to validate a query and estimate its cost prior to running it, and to save frequently used queries. And thanks to recent operational improvements, we’ve been able to double existing query quotas.

Finally, BigQuery customers will have new pricing options starting in July. Data storage costs in BigQuery are becoming even more affordable for everyone, going from $0.12/GB/month to $0.08/GB/month effective July 1st. Furthermore, in addition to the existing on-demand rate for interactive queries, customers with higher-volume usage will soon be able to opt in for tiered query pricing. This provides more economical and predictable cost for interactive queries. Customers who are interested are encouraged to contact a sales representative.

You can get details about these new capabilities and more in our Developer Blog and in our updated product documentation. Got an inspiring use case? Share it in the blog comments or with our community using the #BigQuery tag on Google+.

READ MORE - Google BigQuery gets smarter with large results, productivity improvements, and updated pricing

News Ltd’s Traderoo Powered by Google App Engine

Posted by Joshua Lowcock, Head of Commercial Platforms and Products for News Limited

(Cross-posted on the Official Google Australia Blog)

Editor's note: Today’s guest blogger is Joshua Lowcock, Head of Commercial Platforms and Products for News Limited, an Australian media company. See what other organizations that have gone Google have to say.

News Limited is one of Australia’s largest media companies, spanning newspapers, magazines, online, and subscription TV. We publish over 140 online and printed newspapers in major Australian cities including Sydney, Melbourne, Brisbane, Adelaide, and Perth, as well as in suburban areas.

Classified advertising is a key revenue stream across all our markets, but traditionally booking and billing classifieds had been a manual and time-consuming process. We wanted to implement a solution that would allow customers to serve themselves by placing ads online.

Google App Engine has enabled customers to do just that. We chose Google App Engine as the application because it is easy to build, easy to maintain and simple to scale as the user base and data storage grows. Functionalities within the Google App Engine environment, such as Google BigQuery, have also been useful. We can do an in-depth analysis of our ads and item pricing, as well as provide an internal reporting tool, all using BigQuery.

The end result is a self-service, production booking and billing system - www.traderoo.com.au - which we have developed on Google App Engine. It’s proving to be a real winner for both our business and our customers. It’s fundamentally changed the way customers engage with our company, creating a more usable experience and superb responsiveness. It’s easy to use, and gives more control over ad content, as well as the ability to publish ads online immediately. Online ads are free, while print ads are optional and require a small fee, but complement online ads by extending the advertiser’s reach.

When customers book ads using the Traderoo website, they get automatic email notification from the platform that tells them how their advertisement is performing. Traderoo is optimised for PC, laptop, smartphone and tablet, so the browser and ad placement remain consistent, no matter what device our customers are using.

The real advantage for us is that our classified business has achieved faster time to market, lower costs and less overheads in the form of call centre time and manual data entry. The site has been a huge success, and we look forward to continuing to use Google App Engine as we develop Traderoo further.

READ MORE - News Ltd’s Traderoo Powered by Google App Engine