All Collections
Datatrics Analytical
Advanced Analytical options, tips & tricks
Advanced Analytical options, tips & tricks

This article is a must-read if you want to learn more about Analytical.

Xander Wijering avatar
Written by Xander Wijering
Updated over a week ago

Speeding up Questions and dashboard

Though Datatrics works hard to process all information when doing Questions, the performance when creating questions may be longer than expected, especially when a lot of data needs to be processed. You could do a few things to improve the speed of showing the data.

Exclude unnecessary columns

Many unnecessary columns are often used in the Questions. All this data will be processed, even if the data that will not be displayed and data that is not necessary for the result you want to achieve. Therefore, we recommend selecting only the necessary columns.

Simply click the select none button to deselect all columns. Then, you can select the desired columns.

The smallest table first & INNER join when joining

The order of tables that will be joined matters when using OUTER joins. We recommend using the table with the least amount of columns first.

The biggest difference between an INNER JOIN and an OUTER JOIN is that the inner join will keep only the information from both tables related to each other (in the resulting table). An OUTER Join, on the other hand, will also keep information that is not related to the other table in the resulting table. So OUTER join will get you more records compared to INNER join, which means it requires additional processing of data, and thus OUTER JOIN won't be faster than INNER join.

INNER join and OUTER join will give you results differently, so you can’t replace an INNER join with an OUTER join, as it will change your business logic itself

Too many cards are used in dashboards

When dashboards are slow, you might consider using fewer cards displayed in your dashboard. Metabase recommends having at most 10 cards, and the fewer the number the cards, the faster data will be displayed.

Another thing to consider is using date filters in Questions or dashboards when the performance still needs to be faster. Feel free to contact the Customer Support team if the performance could be better after the optimization steps.

Events (Similar to custom alerts in GA)

Events are messages you can configure to provide extra information to the user reading dashboards. This might be useful to use when having maintenance, a migration, or technical problems on a specific date. To create an event, click on the calendar button in a visualization.

Now, a pop-up appears to fill in the information.

After submitting, a line containing the provided information appears in visualizations having the specific date on the x-axis.

Did this answer your question?