Tables present your data in a familiar format, much like an Excel spreadsheet.
In this view, your search identifies specific attributes and columns to display as a table. Trends AI automatically aggregates these results based on the level of detail you define in your search. For instance, if you type amount, you will see the total sum as a single value, but adding the keyword monthly breaks that total down by month. You can interact directly with column headers to rename, sort, or filter your data. You can also customize the layout by dragging and dropping columns into a new order or adjusting their widths to fit your needs.
Rearrange column order
You can easily adjust the order of your columns once you have finished adding your search terms. To rearrange the column order:
While viewing your answer as a table, select the column header you would like to move.
Drag it across to its new position.
You can also manage the layout through the edit table: configure menu. Simply select the layout icon, then drag and drop the attribute or measure to your desired position. The sequence in the configure menu directly matches the column order shown in your table.
Hide a column
You can hide a column from your view without changing the underlying search logic. For example, if you search for Campus Name, Amount, and Fund Name sorted by Campus Name, you can hide the Fund Name while keeping the specific sorting order intact.
To hide a column from a table, follow these steps:
Enter your search in the search bar and press Go.
Select the table view of the results, and click the chart configuration icon (gear icon).
From the Visible Columns section, select the column you want to hide, and drag it to the Hidden Columns section.
Click Apply.
Resize column widths
You can fine-tune the width of your columns at any time. Any changes you make are preserved when you pin the table to a Board. To resize the column widths:
While viewing your answer as a table, hover over a column border in the column header row.
Click and drag the border to create your preferred column width.
Clip or wrap text
To keep your tables clean, you can choose to clip or wrap long text within cells or headers. You can apply these settings to the entire table or to specific columns.
Clipping shows only the beginning of the text, hiding the rest unless you expand the column width.
Wrapping increases the height of the cell to display the full text across multiple lines.
To clip or wrap text for the entire table, select the edit table configuration icon (gear icon). Select Settings, and under text wrapping, choose wrap or clip.
To adjust text wrapping for an individual column, hover over the column name, select the more options menu icon (3 dots), and choose your preferred setting.
Number formatting
You can customize how numbers appear in any table column or summary based on a measure. This includes changing the category (number, percentage, or currency), the units (auto, none, thousand, million, billion, or trillion), and how negative values are displayed.
Changing the formatting for a column does not automatically update its column summary. You must format column summaries separately.
To change the number formatting:
Select the edit table layout icon to the upper right of your table to open the Edit table panel.
Select the specific column or summary you wish to format.
The Edit panel will appear. For summaries, specifically select Summary. You can also access this by hovering over a column name and selecting the more icon (3 dots):
Under number formatting, use the dropdown menus to adjust the category, units, or negative value style.
Specify a category: number, percentage, or currency. Selecting currency allows you to choose the specific type (USD, EUR, etc.). If left blank, Trends AI will select the most appropriate category for your data.
Specify units: Choose none for standard decimal precision or a specific unit like millions for rounded labels. Trends AI can also automatically select the best units for you.
Depending on your selection, you can also toggle the thousand separator and define decimal places.
Conditional formatting
You can apply conditional formatting to measures and attributes within your table cells and summaries to highlight important data trends. For more details, please see Apply conditional formatting. These features, including formatting for table summaries, are part of the new answer experience.
Sort columns
To sort your table, simply click the column title. You can sort by multiple columns simultaneously by holding down the SHIFT key while clicking headers, which is very helpful for organizing date-heavy data. For example, if you are looking at giving amount by week and quarter, sorting by quarter first and then week ensures everything remains in chronological order.
If you press SHIFT and then select the weekly column header, the weeks are in order, by quarter:
You can also define these sorting rules directly in the search bar using commands like sort by date quarterly.
This sorting method works across all tables in Trends AI, including the Data page.
Content density
To fit more information on your screen, you can adjust the content density of your tables.
Select the Display icon.
Under Content density, choose between regular or compact.
Table theme
You can change the visual style of your table to improve readability.
Select the Display icon.
Under Table theme, choose outline, row (removes vertical lines), or zebra (alternates row colors).
Column summaries
For numeric columns, you can enable summaries to view totals at a glance. These summaries support the same number and conditional formatting as individual cells.
Enable or disable column summaries
Select the Column icon to edit the table panel.
Toggle column summary for the entire table, or manage them for specific columns in the column's individual Edit panel.
Column summaries are generally limited to tables with fewer than 15,000 rows, unless you are using a direct connection to an external database. If your results exceed this limit, the summary will display a dash and suggest using a filter.
Column summary aggregation
You can change how a summary is calculated by clicking its dropdown menu. For measures, you can choose from total, average, standard deviation, variance, min, and max. For attributes, you can select unique count or total count.
Table aggregate summary
The Table Aggregate summary appears when your search includes an aggregate function via the search bar or a formula. It recalculates the specific function for the entire table.
In the example below, the summary uses an Average amount to divide the total gifts in dollars by fund name. Trends AI performs this calculation across the whole table for accuracy, rather than simply averaging the pre-calculated averages of each row.
If a formula contains both a conditional clause and a group_aggregate expression, Trends AI identifies the aggregation type from the first non-constant output clause to determine the summary type.
Table summary aggregate matrix
Function | Total | Average | Min | Max | Variance | Standard deviation | Table aggregate | Total count | Unique count |
Column measure | Yes | Yes | Yes | Yes | Yes | Yes | No | No | No |
Column attribute | No | No | No | No | No | No | No | Yes | Yes |
average(x) | No | No | Yes | Yes | No | No | Yes | No | No |
count(x) | Yes | Yes | Yes | Yes | Yes | Yes | No | No | No |
max(x) | No | Yes | Yes | Yes | Yes | Yes | No | No | No |
median(x) | No | No | Yes | Yes | No | No | No | No | No |
min(x) | No | Yes | Yes | Yes | Yes | Yes | No | No | No |
percentile(x) | No | No | Yes | Yes | No | No | No | No | No |
stddev(x) | No | No | Yes | Yes | No | No | Yes | No | No |
sum(x) | Yes | Yes | Yes | Yes | Yes | Yes | No | No | No |
unique count(x) | Yes | Yes | Yes | Yes | Yes | Yes | No | No | No |
variance(x) | No | No | Yes | Yes | No | No | Yes | No | No |
cumulative_* | No | No | Yes | Yes | No | No | No | No | No |
moving_* | No | No | Yes | Yes | No | No | No | No | No |
rank | No | No | No | No | No | No | Yes | No | No |
rank_percentile | No | No | Yes | Yes | No | No | No | No | No |
group_* (with aggregate) | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No |
group_* (no aggregate) | No | No | No | No | No | No | No | Yes | Yes |
group_aggregate (outer agg) | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No |
group_aggregate (no outer agg) | No | No | No | No | No | No | No | Yes | Yes |
group_aggregate (agg in search) | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No |
sum - last/first_value | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No |
count - last/first_value | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No |
unique count - last/first_value | Yes | Yes | Yes | Yes | Yes | Yes | No | No | No |
min - last/first_value | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No |
max - last/first_value | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No |
average - last/first_value | No | No | Yes | Yes | No | No | Yes | No | No |
stddev - last/first_value | No | No | Yes | Yes | No | No | Yes | No | No |
variance - last/first_value | No | No | Yes | Yes | No | No | Yes | No | No |
Table aggregate
Table aggregates are used when aggregations are non-composable or when an expression combines multiple aggregate functions, such as AVG(x) + SUM(x).
Non-composable aggregations
These aggregations can produce errors if applied to data that is already aggregated. They depend on specific row counts and require precise handling. Examples include:
AVG(x): An average of averages is not mathematically identical to an overall average.
VARIANCE(x) or STDDEV(x): These require every data point for accuracy.
Complex expressions: Formulas like
AVG(x) + SUM(x)lose precision when re-aggregated.
Composable aggregations
These can be safely applied across different levels of aggregation without losing correctness because they do not rely on the original row count. Examples include:
SUM(x): A sum of multiple sums remains a valid total.
COUNT(x): Counting existing counts provides the correct total number.
MAX(x) or MIN(x): The extreme values remain correct regardless of the aggregation level.
Condition | Applicable |
Single composable function at root level | No |
Single/multiple standard functions ONLY (with specific exceptions) | Yes |
Single/multiple cumulative functions ONLY | No |
Single/multiple group functions ONLY | Yes |
Single/multiple moving functions ONLY | No |
Constant functions (e.g., SUM(1)) | No |
Group function supported cases
Table aggregates apply when using a group function, provided there are no non-aggregate subexpressions involved.
Supported:
group_average(revenue, color)min(group_aggregate(average(revenue), querygroups() + {color}, {}))group_average(revenue, city, region) / group_average(revenue, region)
Not supported:
revenue - group_average(revenue, color)(contains a non-aggregate subexpression).
Limitations
Table aggregates are not permitted if the expression includes uniquecount.












