Optimize DQL cost with Davis CoPilot for Workflows

Latest Dynatrace
Tutorial
4-min read
Published Nov 07, 2025
Preview

With Davis CoPilot for Workflows, you can automate problem summarization and ask Davis CoPilot to suggest remediation steps that can be sent to your email.

Overview

With this guide:

You get introduced to cost optimization for Dynatrace Query Language (DQL) using Davis CoPilot for workflows.
You learn how to automate insights and recommendations to reduce query execution costs.

Target Audience

This guide is written for:

Observability engineers
Platform owners
FinOps practitioners
SREs

Scenario

Automatically detect high-cost DQL queries.
Receive optimization suggestions and cost-saving actions via email or chat.

To help teams manage observability costs more effectively, this use case demonstrates how to automate the identification and optimization of expensive Dynatrace Query Language (DQL) executions.

A simple workflow, scheduled to run once daily, detects the top 20 most costly queries from the past 24 hours. By integrating Davis CoPilot into this workflow, each query is automatically analyzed for optimization opportunities.

Davis CoPilot provides tailored recommendations to reduce query costs and sends these insights directly to the respective query authors via email. This helps you to proactively control costs and improve query efficiency without manual intervention.

Before You Begin

Prerequisites

To use Davis CoPilot in Workflows, ensure that you have:

Conversational recommender (ALLOW davis-copilot:conversations:execute;) permission.
Installed Davis CoPilot for Workflows.
Access to DQL usage metrics and billing data.

Steps

1. Configure a workflow trigger

Go to Workflows.
Select + Workflow to create a new workflow.
From triggers, select Fixed time trigger.
Configure the fields:

Set the Run at to your desired time.
Set the Rule to Every day.

2. Get the expensive queries

Select + Add task.
In the search field, enter DQL query, or select Execute DQL Query from the list of the Workflow actions.
Rename the task to dql_query.

In the DQL query field, add the following query:

fetch dt.system.query_executions, from:now() - 24h
| filter status == "SUCCEEDED"
| summarize executionCount = count(), sum = sum(scanned_bytes.on_demand), user = collectDistinct(user.email), app = collectDistinct(client.application_context), by: {query_string}
| sort sum desc
| limit 20

3. Ask Davis Copilot to find the improvements for every query

Select + Add task.
In the search field, enter Davis CoPilot, or select Prompt Davis CoPilot from the list of the Workflow actions.
Configure Davis CoPilot:

In the Prompt field, enter the following prompt:

I've supplied you with result of a DQL Grail query. This result has information about top 20 expensive executed by users in last 24 hours.

Create a json array with the following info:

- query: that is the original query that is given in the result
- email: email of the user who executed the query
- improvement: tell me the reasons why query is expensive and how can user improve it
- context: any relevant context where the query is executed and so on

Make sure that there is no other text beside the json array and no backticks or anything

In the Additional context field, enter the following: {{result('dql_query').records}}
Enable Auto-trim.
Set Document retrieval to Dynatrace.
Rename the task to davis_copilot.

4. Send Davis CoPilot results to user's emails

Select + Add task.
In the search field, enter Send email, or select Send email from the list of the Workflow actions. For more information about the Email workflow actions, see Email Connector.
Enter the workflow task name.
Configure the fields:

In Configure email > Recipients, set the To field to {{_.list.email}}. This is the email of the user.
In the Content, Subject field, enter the following text: Expensive query executed.

Set the Message field to the following:

Hi {{_.list.email}},

You've executed an expensive query that could be optimized to reduce costs. Below are the details to help you improve it:

---

### Query Details:

- **Original Query**:
  `{{_.list.query}}`

- **Suggested Improvements**:
  {{_.list.improvement}}

- **Context**:
  {{_.list.context}}

---

Taking these steps can help improve performance and reduce expenses.

Thanks,
Your Admin

Go to the Options tab and configure the fields as following:

Enable Loop task.
Enter list in the Item variable name field and enter {{result('davis_copilot').text}} in the List field.

5. Finalize workflow configuration

Select Deploy to deploy the workflow.
Select Run to test the workflow.