Видео

Glad it was helpful!

Thanks!

Thanks for watching! Metabase keeps getting better and better!

Are you building your questions in the query builder or with SQL? These features are not available for SQL-based questions. If you are using the query builder and still don't have these features available, please reach out to our support team!

@@AlexYaroshMetabase I just had to log out and back in and they were there :)

Hah, noted :) I'll go slower in the future. Thanks for the feedback!

Thank you! Glad you found it useful!

Thank you!

This should be fixed in 51.3! If you're still encountering this issue on 51.3+, please report it on our github. Thank you!

Passing in code to render charts with Python and R is not supported since they don’t run natively in the browser. As for Google Colab, that could definitely work in the future, but it’s still an open feature request for them: github.com/googlecolab/colabtools/issues/1225

Passing in code to render charts with Python and R is not supported since they don’t run natively in the browser. As for Google Colab, that could definitely work in the future, but it’s still an open feature request for them.

JSON unfolding is actually enabled by default in MongoDB. You can use any nested JSON field in filters, summaries, custom columns etc. Metabase will generally display JSON fields unfolded into columns as well, but there's currently a bug with display settings where if you select to display *all* the fields in the document, the nested fields will not unfold. The workaround is to uncheck any one field from the field selector in the query builder editor (the downward error next to the collection name in the "data" block) - as long it's not "select all", then the nested JSON fields will unfold into columns as well. This issue is on our radar and we're working on it!

True! Offset is actually implemented exactly using LAG/LEAD

You bet! Head over to chart settings (gear icon on the bottom left of chart view) and switch to the "Display" tab. There should be a setting called "Stack values to show" and you can pick "Segments" or "Both" options to display either just the values for individual stacks, or individual stacks AND total.

@@metabasedata thanks for the reply but does this require a particular version? because I can only see "Show values on data points- Toggle for yes or no" & "Values to show: All or Some"

Great callout, thank you! In the underlying query the ranges were ensured to be non-overlapping, but it was not communicated with labels. I updated the dashboard, thank you!

Yes you can install Metabase on windows.

@@matthewhefferon can you please share instructions,it would greatly help

@ can u please share steps for installation

@@kunalkrishneel9478 Go to metabase.com and under docs you can find the installation instructions.

Good question! I think actually the most straightforward way to do this is using a new feature we added in the latest release of Metabase, v51: multiple breakouts by the same field. In v51, you'll be able to add just one aggregation "Cumulative sum", and then group it by both "Year" and by "Month of the Year". This way accumulation will happen only within a single year along the months

Hey @AM-oj8yp, to get the best help, could you please post the details on our Discourse forum: discourse.metabase.com/

if i do left join answer is not coming kindly help

Hey there! To help you out better, could you post your question on Metabase’s Discourse with a few extra details? 1- When you say “blank,” do you mean empty, null, or something else? 2- What happens when you run the syntax-do you see an error or something unexpected? 3- Could you give a bit more info on how your left join is set up? These details will help us (and the community) give you the best possible answer:)

Hi @rafaelg8238 👋. RLS is just one part of our data sandboxing feature, which also includes CLS and extras like row modification and user attributes. We invest heavily in making permissions both secure and user-friendly, that is why it’s a paid feature. Here is the latest on permissions: www.metabase.com/releases/metabase-50#improved-permissions-with-more-control-over-how-groups-view-and-query-data

@rafaelg8238 RLS is just one part of our data sandboxing feature, which also includes CLS and extras like row modification and user attributes. We invest heavily in making permissions both secure and user-friendly, that is why it’s a paid feature. Here is the latest on permissions: www.metabase.com/releases/metabase-50#improved-permissions-with-more-control-over-how-groups-view-and-query-data

We don't have any video about RLS yet, but we are planning more video content and we'll take this into account 👍

Indeed, viewers without permissions to create questions can't change filter on a question, because changing a filter on a question creates a new question. But if you add a question to a dashboard, create a dashboard filter, and connect it to the question, dashboard viewers will be able to use the filter to change the data they see without changing the underlying question. Check out our video on how to use dashboard filters: ruclips.net/video/PO1WLfW9y_U/видео.html

Drill-through doesn't work on sql questions. If you build your questions with the query builder you should see the break out option.

Hello! We haven't recorded an Oracle database connection video but here is the link to the documentation if that's helpful www.metabase.com/docs/latest/databases/connections/oracle

@@metabasedata thanks!

It's currently not possible to join data from different databases in one Metabase query, but your database itself might already have the capability to access data from another database, enabling Metabase to only query just one database for joins! Here are some things to look into: - If one of your databases is Postgres, you can use Foreign Data Wrappers to query data from other databases (including other Postgres databses) from your Postgres - If one of your databses is Redshift, you can use federated queries for the same purpose - With Spark and Databricks (Databricks driver coming in the next release 🥳), you can set up JDBC connections to external databses - With BigQuery, you can use external data source - You can also use federated query engines like Presto/Trino (supported by Metabase) for dozens of other combinations

​@@metabasedata Can I create models from two databases and then create a new model by joining them? Isn't that right?

Thanks for the feedbackt! We don't have a specific video about that, but we are adding that idea to the backlog. In the meantime, there are some helpful tips here: ruclips.net/video/9ZQyS2ZEqa4/видео.html ruclips.net/video/TdgDcJzv8zg/видео.html

Great to hear you find this video useful 🙏 We are not on TikTok at the moment, but it's something we may consider if our community express interest. What data creators do you follow on TikTok?

After numerous attempts, I got this, a few points people need to keep in mind when replicating YOY like above, I just imported the .csv file into the database and it created lot of issues... 1. the date format after importing was not allowing me to get all the date filters like day, week, month, year, etc - FIX - do not directly import the file, create a proper table via SQL command wherever your database is hosted and then upload the data to the table. I tried typecasting the date field before that, but that didn't help 2. Now, the Revenue field that I used came in as DECIMAL value, which was not populating the graph, not sure for what reason, FIX - I recreated the table with new SQL command using FLOAT as the datatype, that worked perfectly 3. Finally when I wanted to filter the YOY using the Client, as you can see above, the Product or User fields are not used, I tried simply adding the Client to the filter along with the Date, this was pulling the data, but not visualizing because we need to select the data for X and Y axis, I did that, but on the dashboard it was not populating the data, FIX - first use the CLIENT filter and then the DATE, that gave the right comparison data, which can be used by Client filter on the dashboard. Hope this helps for someone recreating this scenario, which is very common. Thanks for this amazing tool, loved it!

Currently, Metabase will only show columns from the last stage of aggregation in dashboard filters. But good news is that we're extending these options soon! In the next release or two, you'll be able to add filters for any stage of aggregation. You can track progress on this feature here: github.com/metabase/metabase/issues/46519

Thanks for the suggestion! Feel free to upvote this feature request: github.com/metabase/metabase/issues/42594

Thanks for the feedback 🙏. We are actively working on this and goal line, here is our roadmap: www.metabase.com/roadmap

Great news - this feature is on our roadmap! Check our roadmap at www.metabase.com/roadmap. You can also track this feature request here: github.com/metabase/metabase/issues/6514

Currently, Metabase will run a separate query for every dashboard card

We're evaluating this idea now! Feel free to upvote the feature request: github.com/metabase/metabase/issues/21600

Update: this functionality should be included in the next major release!

Metabase currently does not support importing XLSX files. We recommend converting XLSX to CSV and importing a CSV, or working with Google Sheets and using BigQuery to query Google Sheets directly. Some docs that could be helpful: www.metabase.com/docs/latest/databases/uploads www.metabase.com/docs/latest/databases/connections/bigquery#connecting-metabase-to-google-drive-data-sources

Not at the moment, but feel free to submit a feature request! www.metabase.com/docs/latest/troubleshooting-guide/requesting-new-features

@@metabasedata Love Metabase, but one of the biggest shortcomings of Metabase is its lack of (actually 0) dynamic support of data/functions/variables etc. This is a must have if they want to compete with the bigger players.

We're working on adding more chart types to Metabase! You can see our roadmap here: www.metabase.com/roadmap. Stay tuned...

Superset is a developer oriented BI solution. Literally anyone can add a custom chart type without tweaking the code at all. On the other Metabase develops custom charts in great detail, so user can easily add it.