TLDR
Row Zero is a new spreadsheet that looks and feels like Excel but is 100x faster. You can easily import multi-GB CSV, JSONL, and Parquet files or connect directly to Snowflake, Redshift, Postgres, and S3. Users use us to explore big data and build live dashboards, forecasts, financial models.
Long version
As an engineer working on Amazon S3, I spent a lot of time in Excel because, even as a dev, it was the fastest way to explore data, build a model, or share a forecast with my business partners. My Excel usage was was plagued by three pain points:
- Poor cloud integration - The inputs I needed were in Snowflake, Redshift, S3, or behind APIs
- Poor performance - Excel starts to choke beyond 100k rows
- VBA sucks - When formulas got complicated, I wanted helper functions. But that meant VBA. Yuck.
I love the richness and responsiveness of Excel, but I had to give up too much power to get it. This felt like a false choice, so I started working on it. Queue Rocky training montage.
Today we're announcing Row Zero, a spreadsheet that looks and feels like Excel/Google Sheets, with some huge improvements:
Speed
We're 100x faster than desktop Excel and 1000x faster than Google Sheets. Users have imported data as large as 270M rows and we can go up to 1B.
Connectivity
Workbooks run in AWS. Because the workbook's sitting next to your data, you can load hundreds of millions of rows from Snowflake, Redshift, and S3 in seconds. We spin up the workbook process in the AWS region closest to you so it still feels as responsive as a desktop app.
Python native
Our formula language is Excel-compatible, but formulas can also call custom Python methods (defined in a separate Python code window), which may reference popular modules like pandas, numpy, scipy, and yfinance. Python types are seamlessly marshalled to the spreadsheet type system.
Other papercuts
Imported data is immutable by default. We have data types for datetimes and JSON. Pivot tables are dynamic. Formulas can reference visible-only data with the new tilde operator (e.g. ~A1:A10 will exclude filtered and hidden rows in that range).
If you're using Excel/Sheets to view the results of SQL queries, explore data, or create financial models and forecasts, you'll love us. If you use Jupyter for data exploration but miss the responsiveness of a spreadsheet, you'll love us.
Architecturally, we're different from BI tools. Rather than delegating compute to the data warehouse, we just load everything into RAM because it's faster. Behind the scenes our control plane spins compute resources up and down dynamically to keep costs under control.
Competition
Because the data space is more crowded than a Tokyo subway, here's a short list of competitors and how we're different:
Enterprise BI tools (Tableau, PowerBI, Looker, Sigma) - These apps are dashboard-first. They're powerful, expensive, and rigid. They delegate compute to your data warehouse and are slow to update. They require additional training, and doing anything non-standard is only possible if you're a power user.
"2nd generation notebooks" (Hex, Deepnote) - These apps start with the Jupyter paradigm and bolt on features in an attempt to make them accessible to business users. We're a spreadsheet-first UI that everyone knows how to use.
Other "Spreadsheets" (Airtable/Smartsheet) - These aren't spreadsheets in the style of Excel. They're more database/project management tools. They're not what you'd use to explore data or build a model/forecast.
The team
Before I wrap up, I'd like to recognize the world class team behind the product. Our engineering team is 5 ex-principal engineers from Amazon S3, Tableau, and Airtable. Row Zero's one of those software products that's only possible to build with superstars, and we've got an entire team of them. It's an immensely fun, talented, and thoughtful group to work with. I'm incredibly proud of what we've built and am giddy to see what we come up with next.
Give Row Zero a try and let us know what you think!