Share on LinkedInTweet about this on TwitterShare on Google+Share on FacebookShare on Reddit

Mode Team and logo

Mode will be part of every data scientist’s toolkit

Since launching the beta of Mode in June 2014, it has seen great traction among fast-growing companies like Twitch.tv, TuneIn, and Munchery who are looking to do custom analytics and better distribute data throughout their companies. They’re selecting Mode instead of building tough-to-maintain internal tools.

The public side of Mode, which has been compared to a “GitHub for Data Science” is being most leveraged for education and community building.  The Mode Playbook, for example, is the collection of SQL queries that the Mode team has open sourced to help other analysts tackle common behavior analysis like churn, retention, and feature usage.  

WHAT JUST LAUNCHED ON MODE? 

“The new Mode workflow is a huge step forward in terms of enabling Analysts to create and deliver their best work. But it also lays the foundation for big new features to come,” said Mode Analytics CEO Derek Steer. “The fact that we were able to to deliver such an improvement as a direct result of analysis done in Mode only reaffirms our beliefs in what we are building.” 

“Querying data directly using Mode has enabled us to develop and share deeper customer insights than we could with our existing analytics tools.” –  Conrad Chu, CEO Munchery (Corporate Customer)

WHY MODE  MADE THESE ENHANCEMENTS

Analyzing Analysts
The beta, launched in June 2014, gave them something they couldn’t get otherwise: usage data. By analyzing this data they were able to use the following findings to build a better experience for folks working with data:

  • Writing queries is an iterative process. Queries start with a median length of about 200 characters and grow to 800 characters over their lives. At the same time, early queries return, on average, 4 more columns than finalized queries. This insight inspired the design of a flexible query editor that could be changed at any point during a project, helping analysts optimize screen real estate.
  • Analysts and business users use Mode in very different ways. Analysts start on the Mode home page with much greater frequency than business users. This inspired the Mode team to tailor the home page to help keep analysts on the same page with an activity feed and query diffs.
  • 90% of “pins” (a feature we built to help analysts refer to previous work) were being used to pin database tables rather than reports. This analysis validated the common customer request for a schema browser.

HERE IS A COMPLETE WALK-THROUGH OF NEW FEATURES, from Mode’s Blog:

Introducing A More Powerful SQL Workflow

Mode Pic 1We’re on a mission to build analysis software that works the way today’s analysts work. That means building tools an analyst can share with every other person in her business.

After months of customer feedback, design prototyping, engineering, and, of course, data analysis, we are releasing big updates to make analyzing data with Mode even better.

The new Mode workflow gives you more powerful tools to explore data and simpler ways to share your findings with the right people. Today’s product updates also lay the foundation for new features to come over the next few months.

You can sign in to Mode for a quick walk-through of the updates, or read more below.

Death To One-Size-Fits-All SQL Editors

Many customers told us they like how lightweight and simple Mode is to layer on top of their SQL database. It was important to us to find ways to add functionality (keep the requests coming!) while maintaining these themes. To do so, we’ve split the workflow into two distinct steps: “Query” and “Report.”

In each phase, you can orient the panels to your preferred layout—and move things around as you go. Some folks like queries on top with data on bottom, and others like them side-by-side. Some queries start as just a few lines with a wide data sets and lead to long queries with just a few resulting columns. Collapse a panel, re-size a panel, change the background color, swap the layout. It’s your call.

MODE pic 2

Schema Browser? Check!

Now you can access robust information about your data source as you develop queries. Simply click “Browser” at the top left to open the Schema Browser. We weren’t content to implement any old schema browser—such a highly requested feature deserved some extra-special attention. Here we adapted pinning functionality to help you group the tables you need at the top of the Browser panel.

MODE pic 3

Preview Reports As You Build Them

Once you’ve gathered your data, click “Present” to switch over to the report builder. It’s WYSIWYG so there’s no need to jump between browser tabs to see how your report looks. Use the Chart and Table buttons to format report components, or open the HTML panel to add your own custom code. There’s a lot to come on this front!

Mode pic 4

Share When You’re Ready

After watching customers navigate the report publishing process, we’ve made it easier to share information with the right people at the right time.

Now when you name a report it is saved privately. If you want to collaborate with other people, just share the report link. Using the new Star feature, your collaborators can bookmark the report for easy access.

When you’re ready for anyone in your organization to discover the project, select “Share with your organization” from the Share menu. To keep your analytics team in the know, new and updated reports now appear in the new Activity Feed on the home page. You can also click on the time stamp of any update to see the how the report has changed since its last edit:

Mode pic 5

Thanks To You! 

The new, more powerful workflow is in huge part thanks to the incredible feedback and insights customers have shared with us since day one. Let us know what you think of the updates via email, Intercom (the “?” when you’re working in Mode), or give us a shout at @modeanalytics on Twitter. –  Please stop by www.modeanalytics.com

Share on LinkedInTweet about this on TwitterShare on Google+Share on FacebookShare on Reddit

Blog Publisher / Head of Data Science Search

Founder & Head of Data Science Search at Starbridge Partners, LLC.