Welcome Data Eng Listeners!

Lyft's story
Lyft understands that every investment in leveraging their data more effectively has a huge payoff—it’s how they reached IPO status, and it’s why they invested in building tools like Amundsen.

It’s also why they chose Mode. Lyft uses Mode to find the nuggets of gold in all that data.

Your story
As your data grows in volume and complexity, you can make the correct foundational choices to keep data workflows streamlined and your team’s time to value low. Below are 3 reasons to rethink data discovery:

3 Reasons to Rethink Data Discovery

Don’t waste your time on a data dictionary.

Data dictionaries are a noble concept, but rarely work in practice. Because data grows and changes quickly, they’re notoriously hard to maintain. And an out-of-date data dictionary, in which people aren’t sure what they can trust and what they can’t - or worse, assume they can trust it when they shouldn’t - can be worse than no dictionary at all. But there’s a second problem with data dictionaries, which is they’re very difficult to get right for everyone. What’s useful a sales rep who’s six weeks into the job likely isn’t the same as what’s useful to a data scientist who’s steeped in your data. Some people need the quick-start guide; some people want detailed technical documentation. And both are expensive to build and maintain.

Catalog questions, not data.

Instead of cataloging your data, catalog what’s universally useful - questions. Focus on tracking the question you answer, rather than the data that you use to answer them. Questions are more directly valuable to your business - they’re the ends to your data’s means. Questions also spark inspiration better than data discovery tools. We’re much more likely to have an aha moment reading about things we’ve learned than we are spelunking through (even very well documented) data. And finally, answers to questions can often serve as a living documentation of your data. By seeing how related questions are answered, you can figure out data structures from context.

Define concepts, not schemas.

The place where documentation can be helpful is around core concepts and metrics. A few well-defined metrics keep on people focused on what truly matter to the business. If you have a few metrics that people know well, when they get new questions, they’ll relate their work to those metrics. If you instead focus on data discovery and documentation over these core concepts, everyone will find different numbers to inform their decisions. While your organization may feel more data driven, everyone will be driving it in a different direction.

How to Mode:
Connect your data warehouse.
Analyze with SQL, Python, or R.
Share across your organization.

Finally, a data platform you’ll want to live in

See how our Notebook and SQL Editor improve the speed and quality of iterative analysis

Mode adds value to any data stack

See how customers are using it

Your business isn’t drag-and-drop simple

Self-service dashboards are limited. Take reporting to the next level with D3, matplotlib, ggplot, and more.

Explore Mode’s completely customizable reports

"It used to take days for our analysts to go from a request for data to a report. With Mode it often takes minutes."

Denis Zgonjanin, Engineering Lead — Data Presentation, Discovery, and Governance

We don't want to eat your data stack

Mode plays nicely with others so you can deploy the best possible solution for each aspect of your infrastructure

Automate everything.
Equip everyone.

On demand

Refresh data anytime

Interactive

Help everyone explore data

Slack

Deliver data right where you need it

Permissions

Get the right data to the right people

Support

Live in-product chat

Learn from the Mode community with tutorials, example reports and visualizations.

Learn SQL

Pull, aggregate, filter, and join from databases.

Try it

Learn Python

Explore data visualization, modeling, and analysis.

Want to kick the tires for free?