Top N rows

User interface

General Principles Add Missing Dates Add text column Add Total Rows Aggregate Append Compare Text Columns Argmax Argmin Concatenate columns Convert Convert date to text Convert text to date Cumulated sum Custom step Delete column(s) Geographically dissolve data Duplicate Duration Evolution Extract date information Extract substring Fill null Filter Formula Geographical hierarchy Get unique groups/values Hierarchical rollup ifthenelse Join Keep column(s) Moving Average Percentage Pivot Rank Rename Replace Geographical simplification Sort Split column Column's Statistics To lowercase To uppercase Top N rows Unpivot Waterfall

Technical documentation

Packages Steps Designing a new step Known limitations Variables UI components Custom code editor Python package

Top N rows

Use this step if you need to get the top N rows of your data based on value column to be ranked. The top can be performed by group if specified (i.e. if you want a top 3 by group, it will return the top 3 rows for every group - see the example below for illustration).

Similar to a SELECT TOP clause in SQL.

This step is supported by the following backends:

Mongo 5.0
Mongo 4.2
Mongo 4.0
Mongo 3.6

Where to find this step?

Widget Filter
Search bar

Options reference

Get top...: specify the numer of top rows to retain
Sort column...: the ranking will be based on this column (so its values must be sortable)
Sort order...: wether the you want the above column to e sorted on ascending (asc) or descending (desc) order
Group by... (optional): if you want to return a top by group, you can select one or several columns that will be used to constitute unique groups (equivalent to the columns that you would specify after the GROUP BY clause in SQL or in the _id field of a $group in Mongo aggregation pipeline)

Example

This configuration results in:

We get the top 3 brands by country.