We’re kicking off Big Data Summer with highly-requested updates that will speed up your analytical workflows and allow you to customize Hex in small-but-mighty ways. Let’s get into it. 😎
Want to define a ratio of sums? A weighted average? Or some other custom measure to use in charts and pivots?
Editors and Explorers can now use Hex's Excel-inspired calculation language to define dynamic aggregations within explorations and Explore cells.
These aggregations automatically update based on how you group your data — without using any SQL or Python. And since explorations show the row-level data behind the calcs, your teammates can easily jump in and build on your work.
You can now format numeric input parameters as a percentage or currency, limit the number of decimal places, and add commas using the thousands separators.
The input parameter’s output is still a plain numeric value under the hood — this feature just makes numbers in your app easier for stakeholders to read and interact with.
Pro tip: Hold the shift key to increase the step size when adjusting a number, and hold the alt key for smaller steps.
When you’re writing a complex query, it’s a hassle to break out a whole new cell just to test or debug part of the code. Now, you can iterate in place with run selected SQL.
Just highlight the snippet you want to run, and hit Command+Enter (or the run button) to execute only the selected code.
With scrollable charts, you no longer have to choose between showing all your data or keeping your viz tidy. Now, you can set a smaller viewport that shows the most relevant data, and let users scroll to explore the rest.
No more stuffing extra data into an “Other” bucket that feels like a black box to viewers. Scrollable charts allow you to keep the long tail within reach without taking up too much real estate in your published apps.
Chart tooltips just got a major glow-up: format and reorder values, add aggregated columns, rename labels, and view color swatches that match legend colors — all in a few clicks.
Still seeing the old UI? Just switch from Manual to Auto mode to reset and unlock all the new goodies.
Faster package installs: Use !uv pip install instead of !pip install for package installs that are 10-100x faster. ⚡
Hide chart titles: You can now toggle off x- and y-axis titles, facet titles, and legend titles directly in the UI. Say goodbye to hacks like adding a space to hide titles.
Drag columns from tables in explorations: Just like from the sidebar, you can now drag and drop column headers from tables into drop zones to configure an exploration.
Link from published app cell to notebook cell: From the three-dot menu of a cell in a published app, you can click “View in notebook” to jump straight to the corresponding cell in the notebook — no scrolling needed!
Allow empty dropdowns: You can now make dropdown inputs optional. Instead of forcing a default selection or creating a “None” option, just toggle on “Optional” to let users clear the selection and reset the value to null. All existing dropdowns will stay required and behave as they have before.
File upload improvements: Got more CSVs to add? Drop them into the Files sidebar, even if there are already files there. We also changed the default sort so your most recent uploads show up right at the top.
Export Snapshots as a PDF: You can now export full Snapshots directly as PDFs to share with your stakeholders.
Enhanced run stats: View run stats in debug mode for published apps; see SQL cell query execution times, connection times, and cache retrieval times; identify execution type of non-code cells in the run stats panel; and hover over a SQL query to see the result set size in a tooltip.
Bulk column formatting, file upload inputs, and more
Format columns in bulk, let app viewers upload Excel, PDFs, & images, and format number input cells.
This week’s updates are all about making the everyday workflows in Hex faster and easier. We went back to the basics with much-loved, highly-requested improvements that cut down on clicks, scrolling, and workarounds — so building in the notebook and app builder feels smoother than ever.
Your database returns date columns in “2025-05-07” format, but your stakeholders prefer “7 May 2025.” You’ve got better things to do than reformat those columns one by one.
With new bulk column formatting in tables and explorations, you can filter by type, hit “Select all,” and apply formats across multiple columns at once. We did the math, and that’s way fewer clicks than before. 🤓
You can now hide and show columns in bulk, wrap and unwrap text in bulk, see the number of columns selected, and clear all selections with a single click.
We also added a new tab for quickly pinning and reordering columns in a list view instead of dragging columns across a mile-long table. As always, we’d love to hear what you think!
We’ve seen some truly inventive workarounds for uploading Excel files with non-tabular data. But those days are behind us.
File upload inputs now support more than just CSVs, opening the door to richer interactivity (and some pretty cool AI workflows, too).
Choosing Excel as the upload type returns a dictionary with a dataframe for each tab — great for multi-sheet .xlsx files.
Or the file upload type lets you upload PDFs, images, Parquet files, whatever you feel like — and returns a binary you can use in downstream Python code.
When you’re troubleshooting or trying to understand the data fueling a chart or pivot cell, sometimes you just need to get back to the source: the dataframe.
Now, instead of command-F’ing your way through the notebook to find where it’s defined, you can simply click “Go to defining cell” (or use the alt-click shortcut) to jump straight there.
It’s like the reverse of hovering over a cell output variable to see where it’s used. Now you can traverse the DAG in both directions, making it a whole lot easier to navigate large projects.
Now when you share a link to a cell, you can choose to preserve the current inputs. No need to explain which filters to set or how to recreate a specific state.
Your stakeholders land on exactly what you meant to show, so you’ll hear fewer “wait, I’m not seeing that” and more “whoa, very insightful.”
For teams on the Enterprise plan using OAuth data connections, it’s now much easier to see which projects need authentication, thanks to the new OAuth summary in the notebook view.
At a glance, you’ll see whether credential sharing is enabled, who owns the token, and which OAuth connections are in play. No more digging through a long list of data connections to figure out what’s what.
When you’re viewing a published app, just click the three-dot menu to see a simplified OAuth summary there, too.
Hex now supports a native GCS integration, so you can access files and pull in assets straight from GCS without exposing keys in your Python scripts.
Just like with the Amazon S3 integration, Admins can set up GCS as an external file integration, and Editors can import files from the data browser. Depending on your configuration, you can query, write back, or download files right from the project.
Navigate back to the source cell: When viewing an exploration, you can now use the dropdown next to its name to easily trace it back to the original cell that the “Explore-from-here” came from.
New default navigation for Favorites: Links in the Favorites section now default to published apps, and there’s a new nav button that’ll whisk you straight to the notebook view. ⭐
Preserve scroll position in multi-select filtering: Before, selecting multiple items would bounce you back to the top of the list each time. Now, you can filter away without the constant re-scrolling!
Shortcut for changing text case: When editing SQL and Python cells, you can now use a keyboard shortcut (ctrl+shift+U) to toggle selected text between ALL CAPS and all lowercase for a quick case change.
Smarter placeholders in single value cells: Instead of seeing “NaN” when values are still loading or undefined, you’ll now see more helpful placeholders in the notebook view. For null values in published apps, viewers will see “No value.”
Copy code to clipboard from markdown cells: Hover over a code block in a markdown cell, and you’ll now see a handy copy-to-clipboard button for grabbing the content you need.
Borders for selected text cells: We added subtle borders around text cells when selected to make the focus state more distinct.
Drill-downs, custom app themes, and more
Drill into charts, customize Hex apps to match your brand, and get early access to AI-powered self-serve analytics.
“Why did this spike happen?” “Why does my data look like this?” We’ve all had those moments — staring at a published app, wishing we could dig into the underlying data to answer our follow-up questions or sanity-check a surprising trend.
With drilling, now you can. Just click on the bar (or point, or slice) you’re curious about, choose how you want to break it down, and Hex will take you straight into a drilled-down exploration.
From there, you can keep drilling to get more granular — then, drill up to return to a higher-level grouping. No need to add filters, replace axes, or wrestle with your data. Hex takes care of the busywork, so you can zoom in and out easily.
Editor and Explorer seats on Team and Enterprise plans can drill from charts in published apps, explorations, and pivots in Explore cells. Give it a spin, and let us know what you think!
When you’re presenting a production dashboard to execs, the last thing you want is to get derailed by a styling nit-pick because the font is off. 🫠
Now on Team and Enterprise plans, Admins can create custom app themes with your team’s font, background and accent colors, and a chart color palette for a branded look and feel.
Editors can apply custom app themes in the app builder to unify branding, create a consistent experience for stakeholders, and yes — finally get your charts to match those PowerPoint slides!
ICYMI, we shared a private preview of Ask Magic in Explore last month — a conversational AI interface deeply integrated with Hex, your data, and your workflows.
Magic helps you get to insights faster by drafting analyses from scratch, surfacing relevant published apps, editing existing explorations, and even answering questions right in Slack.
We’re looking for more beta customers to kick around the tires, so drop us a note at [email protected] if you’re interested in early access.
We’ve built a lot of features into Hex to help close the Sharing Gap and champion a noscreenshots.club culture. But sometimes, a static, point-in-time image is called for.
Until now, grabbing an image from a published app meant right-click gymnastics that left out the legend and title, or keyboard shortcuts that flood your desktop with a bajillion screenshots.
Now, there’s a built-in button that lets anyone copy an image of the entire chart straight to the clipboard. Perfect for sharing in Slack, slides, docs, or anywhere static visuals go to live their best life.
All we ask is that you consider embedding first — for fresh data, better reproducibility, and those nifty on-hover tooltips. 😉
Since launching Semantic Model Sync, we’ve seen teams connect their dbt MetricFlow and Cube models to power governed, self-serve metrics in Explore — allowing stakeholders to answer their own questions with confidence.
Now, we’re making it easier to navigate those imported semantic models. You can search for dimensions and measures inside datasets from the data browser, and jump straight into an exploration from the search results.
We’ve expanded sorting capabilities to make chart behavior more consistent. Now, when you switch from a bar chart to a line chart (or any other flavor), your x-axis sort stays put.
Sort categorical x-axes any way you like for all chart types and series types (including line, area, multi-series, and dual y-axis). Also, pie charts now support custom and size-based sorting. 🥧
More improvements to sorting logic are on the way (including color and facet sorting), so stay tuned. In the meantime, let us know if you have any feedback!
On the Enterprise plan, Admins can now plug in their own OpenAI API Key for Magic’s model calls for all inference and generation requests. BYOK is a great option if you’re looking for more control over governance and billing. Just head over to your Magic settings page to try it out.
Python 3.11 with Pandas 2.0 is now available as a kernel in your project dropdown, and it’s the default for new projects.
EU data residency support for Magic: We now support full EU data residency, helping customers in Europe meet local data sovereignty requirements when using Magic.
Improved default labels for weekly data: Charts with weekly datetime buckets used to default to labels like ”2025 W14” (kinda cryptic). Now, they show a more human-readable format like “Mar 31, 2025,” marking the start of the week.
New relative date filters: You can now filter by “this year,” “this month,” and “this week” for those quick, intuitive time frames.
Enhanced pie chart tooltips: Hovering over a pie chart now reveals the slice’s percentage.
Introducing Embedded Analytics: powerful data experiences for your customers
Embed powerful data experiences and stunning visualizations for your customers.
Delivering data analytics and insights to your customers shouldn’t take weeks or months of in-house product development.
Introducing embedded analytics from Hex — making it faster than ever to ship first-class data experiences for your customers, and grow new revenue streams.
Build, customize, and embed stunning dashboards into your web app using:
🔒 Row-level security for showing customers only their own data
💅 Custom app themes for a white-labeled UI to match your branding
🧩 Embed API with quick and easy pass-through authentication
📈 Scalable pricing for use cases of all sizes
And more!
Need to see it to believe it? Join us for a live product demo and Q&A on April 17th, where we'll build and embed a Hex app in under 10 minutes.
Welcome to our first release notes of spring — and it's a big one, with new features and improvements across almost every part of the product that broaden (or should I say widen?) what you can do with charts, published apps, Explore cells, and more.
A single data app often serves a bunch of different audiences who care about different results. Introducing Saved Views: a way for users to save combinations of inputs and filters when viewing a published app, so they can get to insights that are relevant to them, faster.
On Team and Enterprise plans, you can create saved views, list them for others to use, and schedule them to re-run on a regular cadence for fresh data. Our team’s favorite new workflow? Get personalized conditional notifications based on a saved view.
Until today, you needed to write extra pre-processing in Python to make charts from wide-format data (i.e., spanning multiple columns of data). Now it only takes a click to:
Create stacked or grouped bar and column charts from multiple columns
Facet by multiple columns
Plot multiple columns on the x-axis
In this new era of wide-format-friendly charts, the need to “tidy up” your data using the Pandas .melt() operation has effectively melted away. 😀
🔔 Notifications with CSVs & configurable screenshots
On Team and Enterprise plans, you now have two new notification options:
Include a CSV attachment by selecting a table display in a published app.
Send screenshot attachment of specific tabs or individual cells that triggered a conditional notification.
Mix and match CSVs with PDF or PNG screenshot attachments to send notifications to yourself and that VP who wants daily updates on new customer signups delivered straight to their inbox.
Hex makes it easy for code, no-code, and queries to live together in a canvas made of atomic building blocks that stack and engage with one another.
That’s why it's a rite of passage for every cell type to return a dataframe. Now that Explore cells do exactly that, you can truly integrate them into your notebook to unlock tremendous new workflows.
Got an exploration you want to dig further into? Just throw it into a notebook and start writing SQL or Python against the output of the Explore cell, or chain downstream no-code cells to your heart’s desire.
Make one small tweak to a component, and suddenly every downstream reference is out of date! With great power comes great responsibility, and now you’ve got options depending on how authoritarian you’re feeling. 😈
Our new UI for managing a component’s project references provides useful metadata to help you understand how components are being used and which projects to prioritize updating.
If you're an Admin, this comes with the ability to force update all (or some) project drafts to use the latest version of a component.
Feeling democratic? You can notify component consumers that a new version is available and let each project’s Editor decide whether to update.
We’ve seen customers try to wrangle components with hundreds of downstream references. Now, it can be as easy as a few clicks to keep components consistent across projects.
Write code faster with AI-powered inline completions in Python cells. Magic Typeahead is now out of beta and readier than ever to give you powerful, tailored suggestions that take into account your project context as you’re typing.
View SQL generated by Explore: Hex generates SQL under the hood to power analysis performed with Explore. Instead of pulling up a separate tab with your data warehouse’s query history, you can now see the compiled queries directly in Hex by clicking the “View generated SQL” option in the Explore UI’s three-dot menu.
Dropdown display labels: This feature we announced last time to enhance dropdown input parameters needed a bit more love — but it’s ready for use now. Thanks for your patience!
