Hi guys, Is there a way to have dynamic compact metrics? For instance, 1,500 becomes 1.50K, and 2,510,000 becomes 2.51M Right now, it seems we can only statically set it as one or the other (K or M). Cheers!

Is there a way to select/click to filter charts ? I have a use case where a table lists all the categorical values along with other metrics in a table. Is there a way to make the other trend lines be filtered to One specific value that is selected.

Hello Lightdash team, We are using metric labels extensively to provide translated and user-friendly names for our business users, and this has been working very well across the platform. However, in Canvas, metric names do not appear to follow the configured labels and instead display the original metric names. This can be confusing for users, especially in multilingual environments where labels are used for localization. It would be great if Canvas could either: • Display metric labels instead of the underlying metric names, or • Provide an option to toggle between metric names and labels. This enhancement would improve consistency across Lightdash and make Canvas easier for non-technical users to understand. Lightdash Version: v0.3187.0 Environment: On-premise Thank you for your great work on Lightdash!

📣 Heads-up: Lightdash is moving to Node.js 24 — Wednesday 29 July 2026 TL;DR — most customers don't need to do anything. Lightdash Cloud and standard Docker self-hosting upgrade automatically. Anyone using the Lightdash CLI — on their local machine or in CI/CD — will need to take action. We're giving about a month's notice so you can schedule it. What's changing On Wednesday 29 July 2026, Lightdash moves to Node.js 24 (the current Active LTS). We're moving off Node.js 20, which reached end-of-life in April 2026, to stay current on security patches, performance, and long-term support. On the day, we'll also publish the exact Lightdash version that requires Node.js 24 and update our GitHub Action templates. What you need to do ☁️ Lightdash Cloud — Nothing. We manage the runtime and roll it out for you, with no downtime or action needed. 🐳 Self-hosted — standard Docker / Helm — Nothing beyond your usual upgrade. From 29 July, the Lightdash image ships with Node.js 24 built in; pull the version we announce that day and deploy as normal. 🛠️ Self-hosted — custom images, modified deployments, or building from source — Please verify your environment on Node.js 24 before 29 July. A base-image bump can pull in other toolchain updates, so it's worth checking ahead of time rather than on the day. ⌨️ Lightdash CLI users — From 29 July, the CLI will require Node.js 24, whether you run it on your local machine or in CI/CD. There's nothing you need to change before then — this notice is just so you can plan it in. When the change lands (we'll confirm the exact version that day), you'll need to be on Node.js 24: • Locally: upgrade your machine to Node.js 24 (e.g. via nvm, fnm, or nodejs.org). • In CI/CD: run your CLI steps on Node 24 — e.g. actions/setup-node with node-version: 24 (also becoming the GitHub-hosted default). If you use our GitHub Action templates, pull the latest — we'll update them for Node 24 on 29 July. • Either way, you can skip managing Node entirely with our standalone CLI binaries, which bundle their own runtime. A note on currency formatting (COP & HUF only) Node.js 24 ships updated international formatting data (ICU/CLDR). As a result, Colombian peso (COP) and Hungarian forint (HUF) amounts in server-generated exports and scheduled deliveries (CSV/Excel, emailed/Slack reports) will display without decimal places — e.g. "COP 1,235" instead of "COP 1,234.50". This matches the current international standard for these currencies and how modern browsers already display them. No other currencies or formats change. If your COP/HUF reports feed downstream automation that expects two decimals, please adjust before the upgrade. Questions? Reply to this thread — we're happy to help with CLI or self-hosting upgrades.

github.com/lightdash/lightdash/pull/25022 I implemented the label support for the filter autocomplete which solves the issue raised by @Jose Rego. Would someone review the PR?

Hello team! With our users, a feedback which we continuously get is “I always struggle finding my dashboard when they are in different projects”. Would it be possible to allow search across different projects? Like a toggle at the organisation level, to allow our people to find dashboards with more ease. Scenarios are like: Antoine works in our marketing department, and has access to the marketing project. He also has access to some data from our sales team (i.e. in the “Sales” project)… but to access it he always needs to switch projects at the top. Which he doesn’t always remember to do. And he may also have access to some data in the “Product” project, in the “Finance” project… It’s a pain to remember where exactly is each chart, each dashboard.

Multiple dbt projects in one Lightdash project (beta) You can now connect more than one dbt project to a single Lightdash project. If your models are spread across several dbt repos, you no longer have to choose one or maintain a workaround that stitches them together. Add each repo as a dbt source, and Lightdash merges them into a single set of explores on every deploy and preview. Your agents see the whole picture, and AI writeback opens its pull request against the repo that actually owns the model. To try it, go to Project Settings → Connection settings and find Additional dbt sources. Two things to know while this is in beta. Additional sources need to be GitHub repositories for now. And model and metric names have to be unique across your sources, so if the same name is defined in two projects the deploy will stop and tell you which ones clash. Self-hosted: upgrade to

0.3275.0

or later, then add

multi-dbt-sources

to your

LIGHTDASH_ENABLE_FEATURE_FLAGS

environment variable. This is a beta and we are actively looking for feedback. If you run dbt across multiple repos, we would love to hear how this holds up against your setup and what is missing. Reply here and we will take a look.

Hi there, do you have a plan to support dynamic pivot table like redash.io/help/user-guide/visualizations/pivot-table-visualizations ? It’s very useful for analysts and we don’t have to create hundreds of charts.

🗑️ Issue issue #10503 is no longer planned to be fixed.

No information provided

Check out the linked issue for more information.

github.com/lightdash/lightdash/issues/25933 github.com/lightdash/lightdash/pull/25935 I made a follow-up issue and PR for the labeled dimension and autocompletion. Would someone review it?

Hey there, i've made an PR to optimize Athena warehouse to use it's result-reuse capability. Is there anything i can do to help get this reviewed?

Hi team 👋 I'd like to share a request about virtual views. Date/timestamp columns in virtual views don't get time intervals (Day / Week / Month / Quarter / Year) the way date dimensions from dbt models do. Because of this, we can't switch date granularity on virtual view charts, and date filters always fall back to the day-level picker — for example, there's no way to pick "August 2026" with the month picker. This might be technically challenging, but I find date granularity really useful, so it would be great if time intervals worked on virtual view date columns too 🙏

Hi team lightdash party parrot I have another request, this time about date filters. cool panic There are two things I'd like to share: 1. The "is" filter on date fields only accepts a single value — for example, in the month picker I can only select one month. String fields allow selecting multiple values in the "is" filter, so it would be great if date fields worked the same way, e.g. selecting June and August to compare specific months. 2. On dimensions with a time granularity (week / month / quarter / year), the "is" filter shows a granularity-aware picker, but "is between" always falls back to the day-level calendar. Picking mid-period boundaries (e.g. mid-month on a month dimension) can silently exclude periods, so it would be nice if "is between" also respected the granularity. Thanks for considering 🙏

Hi everyone, I would like to ask if we can have two behavior for default filter: • The default when chart is opening and • The default when user actually click into the filter My use case is that I want some default value, e.g. the data for past 7 days so user won't need to select it. But if user do want to select, I normally found that they have to switch from "In the last" section to "Between" before they can filter the wanted range. Can we somehow have option to set those default? So I want default value is 7 days but if user click into the filter button it will shows between as default.

Hi! Question about aggregate awareness. Our marts are pretty big, so in dbt we build each one at several aggregation levels (

agg_level_0

to

agg_level_3

, same schema, coarser levels drop dimensions). Right now we expose them in Lightdash through a

UNION ALL

view with an

agg_level

column, and users have to pick the right level with a filter. It works, but it's easy to get wrong and it's not a great experience, specially for the end user. I saw somebody already asked for this in github.com/lightdash/lightdash/issues/16168. Is it something you're planning to work on anytime soon? We're considering building it ourselves otherwise, so if it's coming we'd rather wait or help out. Thanks!

🗑️ Issue issue #19994 is no longer planned to be fixed.

No information provided

Check out the linked issue for more information.

Hi team 👋 I have a request about dashboards with tabs. There are two things I'd like to share: 1. When a dashboard that has multiple tabs is delivered to Slack via Scheduled Deliveries, clicking the "Open in Lightdash" button always opens the first (leftmost) tab, not the tab that was actually delivered. It would be great if the link took us directly to the delivered tab. 2. When we edit a chart on a dashboard tab via "Edit chart" and then click "Return to dashboard" in the top right, we land back on the first tab instead of the tab the chart belongs to. It would be nice to return to the tab we were editing. Thanks for considering 🙏

Hi there, It would be great if we can have the static link to dashboard. Right now the only can get static link using slug for dashboard but not for project. eg.

<http://lightdash.com/projects/{project-uuid}/dashboards/{dashboard-slug|lightdash.com/projects/{project-uuid}/dashboards/{dashboard-slug> OR dashboard-uuid}/tabs/[NO TAB SLUG CAN BE USE HERE]

It would be great if we can replace prject uuid with project slug so we can have consistent link between env (e.g. we have link in dev env and prod env without have to change prj uuid in link for ref) It even more useful if we can use tab slug to go to the wanted tab

<http://lightdash.com/projects/project-slug-01/dashbards/db-slug-01/tabs/tab-slug/|lightdash.com/projects/project-slug-01/dashbards/db-slug-01/tabs/tab-slug/>

as above requests https://lightdash-community.slack.com/archives/C067S48BHRP/p1786534711058579

🗑️ Issue issue #19603 is no longer planned to be fixed.

This didn't pick up support in over six months, so we're tidying up the backlog. If you'd still like it, comment here and we'll reopen.

Check out the linked issue for more information.

🗑️ Issue issue #19706 is no longer planned to be fixed.

This didn't pick up support in over six months, so we're tidying up the backlog. If you'd still like it, comment here and we'll reopen.

Check out the linked issue for more information.

🗑️ Issue issue #19982 is no longer planned to be fixed.

There hasn't been any interest in this for over six months. If you'd still find it useful, let us know and we'll reopen.

Check out the linked issue for more information.

Hey gang! Something struck me yesterday when I found myself asking Claude to draw out a visualisation of the semantic layer and explores we've defined in code. Is there any plan for a visual representation of the Lightdash semantic layer inside the app? I miss the days of the old 'ERDs' you used to get with tools like PowerBI and SAP universe design tool: being able to see at a glance the tables, the joins between them, their cardinalities, and a sample of the data sitting in each one. Has this ever been suggested or discussed? Maybe I'm a particularly visual learner, but I think it would really help with onboarding new starters and with explaining to some power stakeholders how an explore actually works. I had a dig through GitHub first and found #17241 (semantic lineage graph) as the closest match, plus the older #53 and #121. Happy to add detail there instead if that's the better home for it.

🗑️ Issue issue #20066 is no longer planned to be fixed.

No information provided

Check out the linked issue for more information.

🗑️ Issue issue #10078 is no longer planned to be fixed.

No information provided

Check out the linked issue for more information.

Hi team 👋 I've collected a couple of feature requests from our internal Lightdash users and would love to share them with you🙏 There are two things: 1. Reorder fields in the "Selected" panel (top-left of the Explore view) Right now, users can reorder columns in the Results table, but not in the Selected panel. Since the Selected panel shows all the chosen fields at a glance, being able to reorder them there as well would be really nice. 2. Show the expected row count before running "All Results" export When exporting data with the "All Results" option, there's no way to know how many rows will be exported until the export finishes. Users would like to see the expected row count (or at least a rough estimate) before kicking off the export, so they can decide whether to narrow down the filters first. It would also help us avoid unexpectedly heavy queries on our warehouse 🥹 Thanks for considering 🙏

Hi Lightdash team, Is there a way to copy the contents of a published space into a user’s personal space? The use case is that a user has view access to a published space and wants to make changes to the dashboards in their own personal space. Currently, duplicating a dashboard doesn’t clone the underlying charts, so the copied dashboard isn’t independently editable. This is similar to Tableau/Power BI where user can download the workbook/report and make changes in the desktop version. Could we have a “Copy space” option that duplicates the space, dashboards, and associated charts?

Hi team 👋 We've had feedback from our internal users about how View underlying data behaves on our self-hosted instance (v2.114.1), and I'd like to raise a couple of requests. 1. We'd like to view tables with more than 50 fields The docs say the limit is 50 dimensions, and that anything beyond that needs to be set via

default_show_underlying_values

. That means writing the fields out in our YAML, but listing 50+ fields there duplicates what's already in our column definitions, so the maintenance cost feels a bit high. What we'd like to ask: would it be possible to make the limit adjustable via an environment variable? That way we could show all the fields without writing out the definitions. That would really help 🙏 2. Column order in View underlying data (both on screen and in downloads) Right now, unless we set

default_show_underlying_values

, the columns in the View underlying data modal come out in an order that doesn't follow order_fields_by, and the CSV export is the same. Would it be possible to have both the modal and the downloaded file follow the order_fields_by order by default? Also, something we noticed while testing: when default_show_underlying_values is set, the modal does follow the order we listed, but the CSV export still comes out in a different order, so the file doesn't match what we see on screen. It'd be great if you could look into this one too!

👋 Feature request: user attributes / RLS enforcement in SQL Runner We use user attributes (

sql_filter

,

required_attributes

) for role-based access to sensitive data. The issue: attribute filters don't apply in SQL Runner, so any Developer can query the underlying tables directly and see everything the project connection can see. In practice we can't give data builders Developer access without also giving them unrestricted access to sensitive data, which undermines the RBAC model for that data. What we'd love (any of these): 1. Enforce model-level

required_attributes

/

any_attributes

in SQL Runner, so tables a user can't access in Explore are also blocked there 2. Lightdash-side allowlist/denylist of schemas/tables for SQL Runner, attachable to roles, groups or user attributes 3. Separate permission to run SQL Runner vs. other Developer capabilities, available outside custom roles Looks related to #13908 (github.com/lightdash/lightdash/issues/13908). Is this on the roadmap, or has anyone solved it differently?

Hey all, is there any plan to support Starrocks? github.com/lightdash/lightdash/issues/7940 I saw this issue there seems to be a PROD ticket (PROD-4559) the question is by when? 🙏

Hello! In data apps, I would like to be able to load images from Shopify's CDN; the photographies can not be loaded unfortunately 😞 (I hold the link in a column) Are there plans to implement this? Thanks!