Instead of Datadog Database Monitoring
Rethink your Datadog Database Monitoring bill. Own your query performance review.
We build your version around the services you run, bring supported Datadog performance exports across and run the review process for you. Start with saved evidence, then scope the live connections your business needs.
Datadog Database Monitoring starts at US$70 per database host each month, billed annually: US$840 a year per host. Infrastructure Monitoring is separate.
- check_circleYour data in a database you own
- check_circleOpen source, no lock-in
- check_circleBuilt and run by Enterprise DNA
The monitoring bill
Paying by database host adds up as your estate grows
Datadog sells monitoring and diagnostics, including collection of live measurements. The comparison starts with the review work your team actually uses and the operational coverage it must keep.
A recurring charge for each host
The published starting rate is US$70 per database host per month when billed annually. More monitored hosts mean more billable units.
A different rate on demand
Datadog lists US$84 per host per month on demand. The lower starting rate carries annual billing.
Infrastructure Monitoring alongside it
Database Monitoring uses the Datadog collection service and an Infrastructure Monitoring subscription. Check that part of the invoice too.
Performance trends and history
The subscription records how tracked database requests perform over time. The free version reviews supported saved exports in your own database.
Investigation and diagnostic views
Datadog includes detailed execution plans and relationships between measurements. Enterprise DNA scopes the diagnostics your custom version needs before any switch.
A place for your decisions
Beside the measurements, your own version holds the person responsible, the investigation deadline and the evidence used to close each tuning task.
Your performance review
From a slow request to an owned investigation
Performance
Read the measurements with their limits
Compare saved request performance without treating missing measurements as a good result.
- check Latest saved results for each request
- check Average duration weighted by execution counts
- check Changes between equal-length review windows
- check Missing samples shown before anyone quotes a result
Follow-up
Put an owner beside the slowdown
Connect performance evidence to the work your business agreed to do.
- check A named owner and business label for each request
- check Investigation tasks with due dates
- check Overdue work listed alongside evidence gaps
- check Closure notes that explain what was checked
Review
Prepare the weekly evidence pack
Bring measurements, decisions and record checks together in reports carrying your business name.
- check A read-only weekly performance report
- check A tuning evidence brief for each request
- check A source access and retention review record
- check Private weekly drafts prepared from saved results
Ownership
Keep the evidence when your tools change
Supported exports become your records, with their source and review history attached.
- check Original supported export retained with its source reference
- check Repeated imports skip evidence already present
- check Changed or overlapping evidence stops for review
- check Your own fields, labels and review rules added as needed
Make it yours
Design your own Datadog Database Monitoring replacement
Everything starts from the open-source base. Then it gets your name, the parts you run, your rules and your data from Datadog Database Monitoring. Try it here, and we build the real one.
2. What it runs
3. Make it fit
4. Your colour
Answered from your own database
Good morning. Here is what needs you today.
We build it from the open-source base, bring your Datadog Database Monitoring data across and run it for you.
What it looks like
Ask it like you would ask a colleague
Real answers from the demo data that ships with Query Performance for Claude Code. It runs in Claude Code inside the Claude app, so it feels like any chat you already use.
Query Performance for Claude Code
Connected to your own database
Which saved requests became slower between the two review windows?
orders-prod
- Query_name
- Order lookup
- Previous_ms
- 250.000
- Mean_ms
- 750.000
- Change_pct
- 200.0
orders-prod
- Query_name
- Catalogue search
- Previous_ms
- 1000.000
- Mean_ms
- 500.000
- Change_pct
- -50.0
billing-prod
- Query_name
- Billing summary
| Host | Query_name | Previous_ms | Mean_ms | Change_pct |
|---|---|---|---|---|
| orders-prod | Order lookup | 250.000 | 750.000 | 200.0 |
| orders-prod | Catalogue search | 1000.000 | 500.000 | -50.0 |
| billing-prod | Billing summary |
Ask it anything
Ten questions the free version answers
These are working questions from the fictional demo, not a claim that Datadog cannot answer them.
Switching
How to switch from Datadog Database Monitoring
A move involves exporting an aligned window of saved query evidence, checking the import, then keeping Datadog running until you confirm the replacement covers your review needs.
- 1
Export an aligned evidence window
Export a native Datadog v1 Query Timeseries response with PostgreSQL query execution-count and nanosecond execution-time measurements for a complete, aligned time window. Include the host, database, user and query signature groupings. Use the exact count and time expressions specified for this import. Datadog recommends its newer v2 endpoint, but this importer supports the still-documented v1 response only.
- 2
Test the import first
Have the source registered and ask for a test import before loading records. Provide the actual export window and filters, along with the distinct query count you expect. The test checks the series before writing records. Reconcile its count with the export, and resolve any differences before asking for the final import.
- 3
Review the import checks
Review the test results for validation errors or rejected series, and check the coverage review for missing samples. Confirm that the export contains the supported PostgreSQL measurements and groupings. The importer rejects conflicting or overlapping windows, and repeating an import skips existing observations. Investigate gaps before relying on the loaded results for performance comparisons.
- 4
Run both systems side by side
Keep Datadog Database Monitoring running while you use the imported evidence for your saved-query reviews. Compare the results and review process over a short period, and check that the new system supports the work you need to do. Identify any requirements it does not cover, such as continuous collection or live diagnostics, before planning to turn Datadog off.
- 5
Switch off Datadog when ready
Switch off Datadog Database Monitoring only after you have checked the imported evidence and review process, and covered any remaining operational requirements elsewhere. If you still rely on continuous collection, live diagnostics or other needs outside saved-evidence review, keep Datadog in place until those needs are covered. The replacement focuses on reviewing saved performance evidence, not continuous monitoring.
Where the real effort goes
Most of the work is mapping your own fields and rules, then choosing the history window and records to bring across. The importer supports PostgreSQL count and time measurements from Datadog v1 responses. Other engines, export formats, query samples, explain plans and alerts need separate mapping or implementation. When you have Enterprise DNA run it for you, they handle this mapping and import work.
Not ready for a call?
Get your Datadog Database Monitoring switch plan by email
We will send you the step-by-step plan for moving your business off Datadog Database Monitoring, what the move really involves, and how we would do it for you.
Sent. Check your inbox.
Your Datadog Database Monitoring switch plan is on its way.
Rather talk it through? Book a call arrow_forwardThree ways to own it
Do it yourself, have it customised, or have it run for you
Open source
Do it yourself
Free, open source
- 1 Clone the repo, run npm install and npm run demo
- 2 Open it in Claude Code, Codex, OpenCode or Cursor and try /regression-review
- 3 Follow the export guide, register your source and test the import before writing
- 4 Set your business name and run npm run view and npm run docs
Built around your business
We customise it
One setup fee, then a retainer for ongoing work
- check Your database estate, tuning rules and review responsibilities
- check Your supported Datadog exports mapped and checked
- check Live connections and a web view scoped around your requirements
- check A different technical foundation where your business needs one
Omni by Enterprise DNA
We run it for you
One setup fee, then a retainer
- check Installed with access, backups and retention responsibilities agreed
- check Weekly performance reviews and evidence packs prepared
- check Connections and review rules maintained as your estate changes
- check Run through Omni by Enterprise DNA beside the rest of your business
Questions
Instead of Datadog Database Monitoring, answered
Does this replace all of Datadog Database Monitoring? expand_more
The free version reviews saved performance evidence and manages tuning follow-up. Datadog also provides continuous collection, live diagnostics and relationships between application measurements. We scope the connections and screens your version needs, and keep existing monitoring until that coverage is in place.
What will it cost compared with Datadog? expand_more
Datadog lists a starting rate of US$70 per database host per month with annual billing, or US$84 on demand. The free version has no licence charge. Hosting, your coding assistant subscription and operating time remain separate. We review your actual invoice and required coverage before recommending a change.
Can we bring our existing performance history? expand_more
The free version imports a defined pair of request-count and execution-time measurements from supported Datadog exports. We check the source, time window and record counts first. Other engines, export formats, execution plans and alerts need separate mapping or implementation. Existing evidence is preserved when an import is repeated.
How do we avoid misleading performance numbers? expand_more
Request duration is weighted by execution counts. Missing pairs withhold the average, and percentage changes require equal-length windows. Each source and request stays separate. The reports cover the loaded evidence, so a recorded slowdown is a reason to investigate rather than proof of its cause.
Who looks after access and the weekly review? expand_more
You do in the free version, or we do through Omni by Enterprise DNA. We agree who can access the records, how backups work and when retention is reviewed. Named owners and due dates keep tuning work visible. The checks support your review and do not certify compliance.
Bring your Datadog invoice to a 30-minute call
Spend 30 minutes with Sam. Bring the database host count, a recent invoice and one performance review your team struggles to finish. We will identify the work your own version should cover and the monitoring you still need.
Query Performance for Claude Code is free and open source, published by Enterprise DNA.
Done paying for Datadog Database Monitoring?
Get your version built