---
title: "Zentrik + New Relic"
canonical_url: https://zentrik.ai/docs/integrations/new-relic
markdown_url: https://zentrik.ai/docs/integrations/new-relic.md
category: "Integrations"
learning_track: null
last_reviewed: 2026-09-19
---

# Zentrik + New Relic

Add measured usage and reliability to the evidence behind a product decision.

## Overview

A workspace owner or admin connects New Relic, reviews a report and chooses its product before importing. Each import creates one Discovery signal with the query, reporting window and returned results. Reports are observations, not customer requests.

## Connect New Relic

1. In New Relic, find or create a **User API key** for a user with access to the accounts you want to read. Use a dedicated, restricted user where possible. An ingest license key will not work.
2. Open **Settings → Integrations → New Relic** in Zentrik.
3. Choose the US, EU or Japan region, paste the key, and connect. Zentrik checks account access before saving. The key is not displayed again.

See [New Relic API keys](https://docs.newrelic.com/docs/apis/intro-apis/new-relic-api-keys/). User keys inherit the user's access and may have capabilities beyond reading; Zentrik uses read queries only.

## Preview and import

1. On the connection page, choose a **New Relic account**. A single accessible account is selected for you.
2. Paste an aggregate SELECT query from New Relic. Include filters for the product and environment. For example: `SELECT count(*) FROM PageView WHERE appName = 'Your app'`.
3. Choose one, seven or thirty completed UTC days. Leave out SINCE, UNTIL, LIMIT, TIMESERIES and subqueries; Zentrik supplies the window and row limit.
4. Select **Preview report**. Check the values, account and dates.
5. Give the report a name, choose its product and select **Add to Discovery**.

Zentrik reads the report again before import. If the values changed, preview again. Reimporting the same report for the same product reuses its signal. Analysis runs after import; adding a report does not guarantee that it will produce a new Insight.

This version supports aggregate functions such as count, uniqueCount, average, sum, min, max, percentile, percentage, filter, rate and apdex. It does not schedule imports, map customers, collect individual log records or manage New Relic alerts. Reports reaching 100 rows must be narrowed before import.

## Data and access

- Zentrik reads account IDs/names and the query results you select. Preview results are displayed before you choose to store a report in Discovery.
- Imported reports retain the executed query, account, UTC window and results. The source link opens New Relic; use the stored query and dates to reproduce the report there.
- Choose data that this workspace may share. Source-user permissions do not become per-record access rules in Zentrik.
- Counts depend on the source's instrumentation, sampling and retention. They do not establish customer identity, revenue loss or causality.
- To rotate a key, edit the connection credentials. A failed replacement leaves the existing connection intact.
- Disconnecting removes the saved connection; previously imported reports remain until deleted. Revoke the key in New Relic if nothing else uses it.

## Troubleshooting

### The key is rejected

Check the selected region and use a current User API key, not a license or browser key.

### An account is missing

Ask your New Relic administrator to grant the connected user access to that account, or reconnect with the correct user.

### The query cannot run

Check it in New Relic. Use an aggregate SELECT query and remove time, LIMIT, TIMESERIES and subquery clauses before previewing in Zentrik.

### No rows returned

Check your filters, instrumentation and retention in New Relic. Empty results do not prove that nobody used the product.

## Related guides

- [Zentrik + Pendo](https://zentrik.ai/docs/integrations/pendo)
- [Zentrik + Aha!](https://zentrik.ai/docs/integrations/aha)
- [Zentrik + Zendesk](https://zentrik.ai/docs/integrations/zendesk)