Aggregate and Group Data Online — Sum, Mean, Count by Category
Summarize any CSV or Excel file by grouping rows and applying sum, mean, min, max, or count functions. No pivot tables or SQL required.
How to Aggregate Data
A complete guide to configuring your data pipeline.
Step 1: Selecting the Group-By Column
Open the Aggregate tool. In the Group By field, type the column whose unique values will define each group (e.g., Region, Category, or $1). Every distinct value in that column becomes one row in the output.
Step 2: Choosing the Target and Function
Select the numeric column you want to summarize in the Column dropdown, then pick the aggregation function:
- sum: Total of all values per group.
- mean: Average value per group.
- min / max: Lowest or highest value per group.
- count: Number of rows per group — works on any column type.
Step 3: Interpreting the Output
The result is a compact summary table. Each row represents one unique group, and the aggregated value appears in the second column. You can chain this result into a Sort or Filter tool immediately after to rank your groups.
Technical Specifications & Use Cases
Aggregation is the backbone of business intelligence reporting. Summarizing raw transactional data into grouped metrics — revenue by region, average score by cohort, count of events per day — is required before any dashboard or model can consume it.
flowingTable executes groupby operations via pandas.DataFrame.groupby(), which uses a highly optimized hash-map partitioning algorithm. This guarantees that groups are identified and summarized in a single pass over the data, regardless of file size, preventing the memory overflow common in Excel PivotTables when handling tens of thousands of rows.
Frequently Asked Questions
What is the Group By operation and when should I use it instead of a pivot table?
The Group By aggregation produces a simple two-column summary table: one column for the group labels and one column for the aggregated metric. It answers questions like 'What is the total revenue per region?' or 'How many orders does each customer have?'. A pivot table extends this concept into two dimensions simultaneously, spreading one variable across rows and another across columns to create a cross-tabulation matrix. Use Aggregate when you need a simple, flat summary. Use Pivot when you need to compare a metric across two categorical dimensions at the same time.
Can I group by multiple columns simultaneously — for example, by both Region and Product Category?
Yes. To group by multiple columns, enter all of them as a comma-separated list in the Group By field (for example, Region, Category). The engine will treat each unique combination of values across those columns as a single group. The output will contain one row per unique Region-Category pair, with the aggregated metric calculated for all records that match that specific combination. This is equivalent to a SQL GROUP BY clause with multiple fields.
What is the practical difference between the 'sum' and 'count' aggregation functions?
The 'sum' function adds together all the numeric values in your target column within each group, producing a total (for example, the total revenue per region). The 'count' function simply counts how many rows exist in each group, regardless of what the values in the target column are — it works on both numeric and text columns. Use 'sum' when you want to total a quantity. Use 'count' when you want to know the frequency or volume of records in each category, such as the number of transactions per customer or the number of products per supplier.
How does this tool handle aggregation performance on large files compared to Excel PivotTables?
Excel PivotTables build their summary by loading the source data into the application's rendering layer, which is constrained by the available RAM in your workstation and often causes freezing or crashes on files exceeding 100,000 rows. flowingTable's aggregation engine executes the groupby entirely in the server-side Python process using pandas' hash-map partitioning algorithm, which identifies and summarizes all groups in a single linear pass over the data. This architecture processes millions of rows in seconds without involving the browser DOM at all, making it reliably faster for large datasets than client-side spreadsheet software.