Create metrics and KPIs in the Mammoth Metrics panel

Create a metric from a column header

Use this for a quick Sum, Average, Standard Deviation, Min or Max of one column.

1. Hover a number column's header in the data grid.
2. Click the Create Metric icon that appears (number columns only).
3. In the Create Metric popup, check the Metric name (filled in, for example "Sum of quantity_on_hand", and editable).
4. Pick a Function: Sum, Average, Standard Deviation, Min or Max.
5. Set Formatting: currency, percentage, decimals and short form.
6. Click Create Metric. A "Metric added" message confirms it.

To add a title, subtitle, description, expression or conditions, click More options instead. The full metric editor opens with these settings filled in.

Open the Metrics panel

Two controls open the same panel:

• the Metrics icon in the action bar at the top of the data editor, next to the Highlights icon (it shows an active state while the panel is open);
• the rotated "Metrics" tab on the right edge of the View, which shows how many metrics the View has.

With no metrics, the panel shows Create a metric, "Use this space to create some metrics from numerical data" and a + New Metric button. Once it has cards, add more with the blue + button at the bottom of the panel.

In preview mode, the panel says metrics are unavailable and, when the View has metrics, offers Exit Preview.

Create a metric with AI, SQL or a formula

Click + New Metric. The modal opens in SQL mode, with an AI expression generator at the top. A switch changes between SQL and Formula. The book icon beside the title opens the documentation.

Describe the metric and let AI write the SQL

1. Type what you want in the box labelled "Describe what you want to generate", for example "Create a metric for the total quantity sold".
2. Click Generate, or press Enter. The AI writes the SQL expression.
3. Review it in the SQL Expression editor and adjust it if needed.
4. Optionally add a condition to limit which rows count. If your description includes a filter, Mammoth can move it into the condition.

Build the expression yourself in Formula mode

Click Formula (next to SQL) and build the expression from functions and column references. You can add conditions here too. The editor shows "Looks good" when the syntax is valid.

Column names in formulas go in double quotes: SUM("deal_value"), not SUM(deal_value).

Formula examples:

For a calculation that needs different conditions in one expression, such as a defect rate or a win rate, use SQL mode and describe it in plain words.

Preview, label and submit

The right side of the modal previews the metric card with its current result, above a grid of the underlying rows.

Click Submit. It stays disabled, with a tooltip saying why, until the metric has a title and a valid expression, and while the SQL is still generating. If another metric already has the title, Mammoth adds a number (for example "Total sales 1").

Format a metric as currency, a percentage or short form

The same format options appear in the column header popup and the full modal.

Use currency for revenue, percentage for rates and ratios, and short form for large counts.

Use a project parameter in a metric condition

A metric's condition can use a project or workspace parameter, so one value (a threshold, a date, a name) is set in one place. See Parameters and snippets.

1. In the New Metric modal, click Add Condition.
2. Pick the column and the operator.
3. For the value, choose Parameter (Bind to a project parameter) instead of Value (A fixed value).
4. Click Select a parameter and pick one. Only parameters whose type fits the column are offered; if none fit, the list shows "No parameters available".
5. Click Apply Condition, then Submit.

Example: SUM("sales") with the condition "discount greater than the parameter max_discount" adds up sales above the current discount threshold.

The expression itself has no Insert Parameter button, in SQL or Formula mode. To use a parameter inside a calculation, use a Math step in the pipeline. See Calculate with formulas.

The metric uses the parameter's value when you save it. After you change the parameter, the metric picks up the new value the next time the View's pipeline runs. If the parameter is deleted, the metric keeps the last value as a fixed value.

Copy, edit, duplicate, lock or delete a metric card

A card shows its title (with a lock icon when locked), the value, the Subtitle and the Description. A metric built from a plain-English description shows it in a row you can click to see the generated query. A metric with a condition shows it as a filter summary. If a metric can't be calculated, the card shows Error; hover it to see why.

Hover a card to see five action icons. They are disabled while the pipeline has an error.

Drag a card by its header to reorder the cards.

Watch metrics change as you filter

Metrics recalculate when you filter with Explore Cards. For example, before filtering: Total Revenue $5.2M, Average Deal $12,340, Order Count 421, Win Rate 67%. After double-clicking "West" on the Region card: $1.8M, $14,560, 124 and 72%.

Compare a segment with the total using lock

Lock a metric while no filters are on to keep its all-data value.

1. Lock "Total Revenue" at $5.2M.
2. Filter Region to West.
3. Create another Revenue metric and leave it unlocked. It shows $1.8M.
4. Compare: West is about 35% of total revenue.

The same works for any segment: a locked Total Sales of $3.2M beside an unlocked $1.1M filtered to East shows East is 34% of the business.

To follow a trend, filter month by month (Jan, Feb, Mar) and watch the values change.

Example metric sets

• Sales: SUM("deal_value"), AVG("deal_value"), COUNT() with status is Won, and a win rate in SQL mode.
• Manufacturing quality: COUNT(), COUNT() with defect_severity is not None, a defect rate in SQL mode, and AVG("yield_percentage").
• Compliance: COUNT(), COUNT() with amount > 10000, SUM("amount") with amount > 10000, and a compliance rate in SQL mode.

A good start for any dataset is 3–5 metrics: a total count, a key sum, a main average and one or two business ratios.

Metrics compared with Explore Cards and View Insights

Use Explore Cards to find patterns, then create metrics to track the numbers behind them.

Troubleshooting metrics

Related

• Explore Cards: the filters metrics follow.
• View Insights: let AI summarize the View.
• Parameters and snippets: set a value once and use it in metric conditions.
• Tables, KPI cards, insights and text: show numbers on a dashboard.
• Schedules and triggers: refresh data on a schedule so metrics stay current.
• Use Mammoth on a phone: read metrics on a phone.