Looker: An Introduction
The use of Business intelligence technologies and strategies effectively, is what determines a company’s position in this data-driven world. A number of BI tools are available today, but what makes Looker different?
But first, a brief history lesson
As a data enthusiast, it is remarkable to see how far we have come with respect to data technology. It has seen tremendous changes over the past 30 years, and it’s still changing at an exponential rate. Databases used to be expensive and very slow, but now they are fast and extremely cheap. And the icing on the cake? Lots of them reside on the cloud. That means a lot of the old constraints don’t exist anymore.
Here’s the not-so-great news. Even though databases have evolved rapidly the data analytic tools haven’t kept pace. Most of them are still designed to work with the slow and expensive databases. This can hinder companies from getting the value they want from their data, in a timely manner.
The landscape of business intelligence/analytic tools has seen 2 prominent waves. The first wave saw the likes of Business Objects, Cognos and Micro Strategy dominating the landscape. These tools guaranteed standards, scalability and governance. By data governance we mean who can take what action, on what data, in what situations, using what methods.
With the emergence of the second wave, we saw Tableau, Power BI and Qlik dominating the landscape. These tools promised self-service, flexibility and agility. Though these are great features to have, companies are having a hard time finding a middle ground. A tool that has the best of both worlds; self-service + governance.
Enter Looker ✨
Looker is Google Cloud’s cloud-native Enterprise BI platform. It provides secure access to near real-time data whenever you need it. It is a multi cloud data platform so it provides the flexibility to deliver impact through data regardless of the cloud platform currently in use. So rest assured, you can host Looker in the cloud of your choice with either Google Cloud, AWS or Azure.
Why Looker?
One thing that makes Looker future-proof is its database agnostic nature. It is optimised to run with whichever database you have in place. Looker supports over 50+ SQL dialects. One of the reasons that makes this possible is the use of the JDBC driver pattern. This helps Looker cater to a wide set of enterprises and businesses, regardless of the databases in place.
Looker also allows for the re-use of SQL queries. You can define your data model once and reference any piece of it anywhere. This way, you can avoid writing the same SQL queries repeatedly.
Looker provides pre-built analytical templates called Looker Blocks to help you move from data to dashboards as quickly as possible. They are pre-built pieces of code that you can leverage to accelerate analytics. All the way from optimized SQL patterns to fully built-out models to custom visualizations, there are a bunch of Looker blocks to choose from that can serve as a starting point for quick analytics.
This ease of use facilitates nearly everyone to be able to make use of data effectively without consulting the analytics team. This essentially gets rid of the “bottleneck” in case the analytics team is just a one-man army. When people need access or need to derive insights, they no longer have to wait in line.
Looker supports two-factor authentication and integrates with LDAP and SSO(supporting SAML, OneLogin and Google Apps). This is of great use to companies that have invested in user authentication tools and need access controls in place.
Built into the core of Looker’s platform are fine-grained access controls which provide three levels of data governance:
- Model level- limits which models users have access to.
- Group level- limits what content users have access to in Looker.
- Role level- sets specific feature functionality and data an individual has access to in Looker.
This layered approach to data governance can be of particular value to industries with specialized security requirements and privacy considerations.
Aggregate awareness is one of the reasons that makes Looker such a cost-efficient tool. Many organisations store their data at a finer granularity level. When queries are run to derive insights on say a year’s worth of data, this can significantly increase the cost of running the query. For very large tables, Looker developers can create smaller aggregates tables of data, grouped by various attributes. These tables essentially act as summary tables that Looker can use for queries, instead of going through the original large table. When implemented strategically, aggregate awareness can significantly speed up the average query by orders of magnitude.
Looker also provides support for embedded analytics. This allows users and customers to explore data embedded in an iframe in any HTML-formatted webpage, portal or application. The iframe then executes the entire Looker application, requesting the data which is necessary to display the query.
At first glance, you may feel that embedding presents privacy and security concerns. To mitigate these concerns, Looker offers a variety of embedding methods depending on the level of authentication you feel is required by the users accessing your data. For more information head on over to Security best practices for embedded analytics
Next Steps
The possibilities are endless with Looker, and to fully appreciate its functionality we must see it in action. Big query and Looker is a great place to start if you are curious about the combined possibilities of Big Query in conjunction with Looker.
Love to hear your feedback!
