> ## Documentation Index
> Fetch the complete documentation index at: https://docs.metricanic.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Find what drives campaign performance

> Move from campaign profit to countries, devices, and offers, then use individual records to explain unexpected results.

Start a report with a decision you need to make. “Which campaign lost money yesterday?” needs campaign totals. “Where did this campaign lose money?” needs a breakdown. “Why is one purchase missing?” needs an individual event record.

Metricanic provides aggregate reports for comparing performance and raw views for following individual visits, clicks, events, and postback attempts. Use each for the question it can answer.

## Investigate a campaign that looks profitable overall

Suppose a campaign reports 1,000 visits, USD 200 revenue, and USD 150 cost. Its USD 50 profit looks positive, but a country breakdown shows two different results:

| Country        | Visits | Revenue | Cost    | Profit  |
| -------------- | ------ | ------- | ------- | ------- |
| Canada         | 400    | USD 140 | USD 60  | USD 80  |
| United Kingdom | 600    | USD 60  | USD 90  | USD -30 |
| Total          | 1,000  | USD 200 | USD 150 | USD 50  |

The next useful question is why the United Kingdom segment loses money. A device or offer breakdown can show whether the result is broad or concentrated. Changing the report does not change traffic routing.

## Build the comparison

1. Open **Campaigns** and choose the date range and reporting timezone.
2. Apply the selection and display **Visits**, **Conversions**, **Revenue**, **Cost**, **Profit**, and **ROI**.
3. Select the campaign and open its **Report** so the following analysis stays scoped to that campaign.
4. Group by **Country**.
5. Add **Device Type** and **Offer** when you need another level of detail.
6. Apply the grouping and sort by the metric that answers your question.
7. Compare the losing segment with the campaign total before taking action elsewhere in the tracker.

Keep the same dates and filters while changing the breakdown. In the example, the country rows should still explain the campaign's USD 50 profit. If the scope changed, they are no longer the same comparison.

## Choose flat rows or a drill view

The report interface supports up to three grouping dimensions. Their order determines the breakdown.

**Flat** puts each combination on a separate row, such as `Canada / Mobile / Offer A`. Use it to compare combinations directly and sort the visible result.

**Drill** starts with the first dimension and lets you expand a row into the next level. Use it when you want to inspect selected segments without opening every country, device, and offer combination at once. Lower levels load as you explore them.

A chart helps you see whether the selected metrics changed across the reporting period. Use the table to inspect the precise amounts and the active grouping to explain which segment contributed.

## Choose columns for the stage you are checking

A landing page comparison often starts with **Visits**, **Clicks**, and **CTR**. An offer comparison needs **Conversions**, **Revenue**, **Cost**, and a rate with the right denominator. A funnel with several stages needs **Event 1**, **Event 2**, or other configured event columns.

The [metric dictionary](/reports/metrics) explains those denominators. In particular, Metricanic **Clicks** are tracked offer clicks, so **CR** answers a different question from a traffic source's ad-click conversion rate.

Use the column settings to adjust supported numeric precision when small amounts round to zero. Precision changes the display only. For a measure that is not built in, create a [custom metric](/reports/custom-metrics).

## Switch to records when a total needs explaining

Open the **Events** navigation group and choose the appropriate view:

| View            | Use it to answer                                                            |
| --------------- | --------------------------------------------------------------------------- |
| **Live Visits** | Did the visitor reach Metricanic tracking, and which campaign was recorded? |
| **Live Clicks** | Did the tracked offer-click step occur?                                     |
| **Events**      | Was the incoming action processed, with which event slot and amount?        |
| **Postbacks**   | Was an outgoing request attempted, and what did the receiver return?        |
| **Errors**      | Was a tracking failure recorded for this flow?                              |

Raw views use the record's own timestamp. Aggregate performance is attributed to the original visit date. A purchase received today can therefore belong to yesterday's campaign report. Read [Dates and filters](/reports/filters-and-dates) before treating that difference as missing data.

## Keep the result useful after this session

Apply [tags](/reports/tags) to campaign segments that need follow-up. Tags record your assessment of a dimension value, while [automated rules](/automation/rules) can maintain markers from conditions.

For a spreadsheet or external analysis, use the authenticated [reports and events API](/api-reference/reports-and-events) and fetch every required page. The current report interface does not provide a CSV export action. A copied screen or one API page should not be treated as a complete dataset.

If a report is empty, remove filters, check the date and timezone, include the relevant entity status, and inspect a known visit. If totals disagree with another platform, follow [Numbers do not match](/troubleshooting/discrepancies).