Metaplane Pricing and Datadog Data Observability Pricing: Cost per Monitored Table
Metaplane's pricing page says Pro is billed per monitored table and never prints the rate. Datadog, which bought Metaplane in April 2025, does print one: its Data Observability list price is $16 per monitored table per month on an annual commitment. Below is what that meter costs at 10 to 5,000 tables, what counts as a table, and how it compares with Monte Carlo and Bigeye. Read on 16 September 2026.
Read-only connection. Datatrail never moves or mutates your data.
Metaplane is free for 10 monitored tables, and its Pro plan is billed per monitored table up to 100 tables at a rate Metaplane does not publish. Datadog, Metaplane's owner, publishes Data Observability Quality Monitoring at $16 per monitored table per month billed annually, $21 month-to-month and $24 on-demand, so 100 monitored tables cost $19,200 a year and 1,000 cost $192,000 at the annual rate. Lineage and metadata syncing are included at no charge. Jobs Monitoring for Databricks, Spark, Airflow and dbt is billed separately per host-hour or job-hour. Datadog does not state that Metaplane Pro is billed at the Quality Monitoring rate, so treat $16 as the published anchor and confirm it on your quote.
Datadog Data Observability pricing: three billing options, two products
Datadog's pricing comparison page lists every product with three columns: billed annually, billed month-to-month and billed on-demand. Data Observability appears as three line items with the Datadog site set to US. The Serverless Databricks tier of Jobs Monitoring is the only one without a figure.
| Product | Billing unit | Billed annually | Month-to-month | On-demand |
|---|---|---|---|---|
| Quality Monitoring | Per monitored table, per month | $16 | $21 | $24 |
| Jobs Monitoring, Clusters (Databricks and Apache Spark) | Per host, per hour | $0.05 | $0.06 | $0.072 |
| Jobs Monitoring, Orchestrators (Airflow and dbt) | Per job, per hour | $0.10 | $0.12 | $0.144 |
| Jobs Monitoring, Serverless (Databricks) | Per job, per hour | Contact sales | Contact sales | Contact sales |
$24 against $16 for a monitored table, $0.072 against $0.05 for a cluster host-hour, $0.144 against $0.10 for a job-hour. Month-to-month sits in between at $21, a 31% premium. The annual commitment is the only way to get the $16 figure.
The Quality Monitoring card reads "Starting At $16 Per Monitored Table, per Month". Datadog does not publish volume brackets for it, unlike Monte Carlo's rate card. Treat $16 as list for a small annual deal and expect a large estate to negotiate below it.
Datadog's FAQ states that lineage, schema and metadata syncing are included at no additional charge, and that column-level lineage covers tables you do not monitor. You pay for the alerts, not the graph.
Datadog Quality Monitoring cost per year, 10 to 5,000 monitored tables
Monitored tables times the monthly rate times twelve. No discount is applied, because none is published, and the table count assumes standard monitors only. Custom SQL monitors add to the count, as the next section shows.
| Monitored tables | Annual, $16 | Month-to-month, $21 | On-demand, $24 |
|---|---|---|---|
| 10 | $1,920 | $2,520 | $2,880 |
| 50 | $9,600 | $12,600 | $14,400 |
| 100 | $19,200 | $25,200 | $28,800 |
| 300 | $57,600 | $75,600 | $86,400 |
| 1,000 | $192,000 | $252,000 | $288,000 |
| 5,000 | $960,000 | $1,260,000 | $1,440,000 |
Per table per year: $192 annual, $252 month-to-month, $288 on-demand, at every size. Datadog prints no bracket that lowers it.
Monte Carlo charges 1.75 credits a day for each of the first 1,000 tables and 0.27 by the fifth thousand. Datadog's list shows one price for table one and table five thousand. At 100 tables the two are within 8% of each other. At 5,000 the Datadog list total is more than twice Monte Carlo Scale.
Datadog measures monitored tables every hour and bills the monthly average. A backfill or migration where you monitor 200 extra tables for ten days adds about a third of a month for those tables, roughly $1,070 at the annual rate, not a full month.
What counts as a monitored table, and where custom SQL changes the bill
This is the part of the price most estimates get wrong, and Datadog defines it clearly on its own FAQ. Unlike Bigeye's "Active Monitored Table" or Monte Carlo's "credit", the unit has worked examples attached.
You can sync as many tables as you want from a warehouse or database. A table becomes billable only when a monitor is added to it, or to a view, materialized view or column in it.
Freshness, row count, nullness and uniqueness on the same table count as one monitored table, however many you add and whether they sit on the table or on its columns.
Every unique custom SQL monitor adds one monitored table on top. Datadog's own example: freshness plus row count plus three custom SQL monitors on one table is four monitored tables.
A group-by monitor that evaluates 100 groups on a table is charged as one monitored table. Segmenting a row count by region or customer does not multiply the bill.
Here is how that plays out on a realistic dbt project. Say a team monitors 200 production models with freshness and row count, and writes 40 custom SQL checks on the revenue, billing and customer models where business rules matter. That is 240 monitored tables, not 200: $46,080 a year at $16, of which the custom SQL layer is $7,680. The custom checks are 17% of the monitors and 17% of the bill, which is the right way to think about them. Every business-rule test you write in the observability tool instead of in dbt is roughly $192 a year at list.
That is also the line item worth arguing about. Metaplane's own Pro plan caps custom SQL monitors at 5, so a team moving from Metaplane Pro onto Datadog's meter, or onto Metaplane Enterprise, often discovers that the checks they could not write before are the ones that grow the bill now.
Metaplane pricing plans: Free, Pro and Enterprise
Metaplane still sells under its own brand, now "Metaplane by Datadog", with a 14-day trial of all features before you choose a plan. The comparison table on its pricing page, read by rendering the page on 16 September 2026:
| Plan | Price | Monitored tables | Custom SQL monitors | What it adds |
|---|---|---|---|---|
| Free | $0, free forever | 10 | 3 | Volume, schema, freshness, uniqueness, nullness, distribution, custom SQL; Slack, email, Teams |
| Pro | Usage-based, "pay per monitored table", no rate printed | 100 | 5 | Adds column-level lineage, partition and rolling-window monitors, PagerDuty; Data CI/CD and spend monitoring are add-ons |
| Enterprise | Custom | Unlimited | 10 | Adds custom integrations, SSO (Okta, AD, SAML), AWS or Azure PrivateLink, API and webhooks |
The comparison table sets Pro at 100 monitored tables and Enterprise at unlimited. A warehouse with a few hundred dbt models cannot run on Pro, so the self-serve plan and the custom quote meet at a size that is ordinary for a mid-market data team.
Plan cards say Free includes 4 users and Pro 12 users with unlimited viewers. The comparison table on the same page says 1 and 5. Ask which applies before you plan on the larger number.
The page invites Snowflake customers to pay with existing Snowflake credits. For a team with an unspent Snowflake commitment, that can matter more than the per-table rate, because it moves the spend off a new vendor line.
Third-party sites quote Metaplane at "$10" or "$208.33 per month". Neither figure appears on metaplane.dev or datadoghq.com, and we do not repeat them as prices. The $16 Datadog figure is the only per-table dollar amount published by the company that owns Metaplane.
Datadog Jobs Monitoring pricing for Databricks, Spark, Airflow and dbt
Quality Monitoring watches the tables. Jobs Monitoring watches the pipelines that write them, and Datadog bills it independently of how many tables or integrations you connect. The Clusters tier covers Databricks and Apache Spark jobs on Kubernetes, Amazon EMR and Google Dataproc at $0.05 per host per hour on an annual commitment. The Orchestrators tier covers Airflow DAGs and tasks and dbt jobs and models at $0.10 per job per hour. Serverless Databricks jobs are "Contact Us for Pricing".
Two worked examples at the annual rate, as arithmetic on the published unit. A 20-host Spark cluster running ten hours a day for 30 days is 6,000 host-hours, which is $300 for the month ($432 on-demand). Fifty dbt and Airflow jobs that each run 30 minutes once a day come to 750 job-hours a month, or $75 if Datadog counts observed run time. Datadog does not publish how partial hours round, so check that on the quote.
The fine print carries three allotments worth knowing. Each Jobs Monitoring host-hour includes 0.205 GB of ingested spans and 1,370 indexed spans, and traces beyond that bill as APM span usage. Spark on Kubernetes includes 10 Jobs Monitoring containers per host. And log collection, which is optional, bills through Datadog Log Management. Hosts running Jobs Monitoring are not also billed as Infrastructure hosts, except for Spark on Kubernetes, where any node running the Datadog agent counts as an Infrastructure host.
Metaplane and Datadog pricing vs Monte Carlo and Bigeye, per year
Three observability vendors now publish a number you can multiply by a table count: Datadog directly, Monte Carlo through its order forms and rate card, and Bigeye through two marketplace packages. Every figure is list, dated, and read by this site from the vendor's own document.
| Vendor | Published price | 100 tables | 300 tables | 1,000 tables | Source, read |
|---|---|---|---|---|---|
| Datadog Quality Monitoring | $16 per monitored table per month, billed annually | $19,200 | $57,600 | $192,000 | datadoghq.com, 16 Sep 2026 |
| Monte Carlo, Scale | $0.28 per credit, 1.75 credits per table per day | $17,885 | $53,655 | $178,850 | Order form and docs, 15 Sep 2026 |
| Monte Carlo, Start | $0.18 per credit, up to 1,000 monitors | $11,498 | $34,493 | $114,975 | Order form and docs, 15 Sep 2026 |
| Bigeye | $45,000 for 100 tables, $75,000 for 300 | $45,000 | $75,000 | No published package | AWS and Microsoft Marketplace, 15 Sep 2026 |
| Metaplane Pro | Per monitored table, rate not printed | Not printed | Above the 100-table cap | Above the cap | metaplane.dev, 16 Sep 2026 |
| Datatrail | Flat monthly plan, no per-table meter | $99 to $799 a month | $99 to $799 a month | $99 to $799 a month | datatrail.ai/pricing |
$19,200 against $17,885 at 100 tables, $57,600 against $53,655 at 300. Both are well under Bigeye's $45,000 and $75,000 packages. The choice at this size is about where your alerts already live, not about price.
Datadog's list stays at $192 a table a year. Monte Carlo's Scale blended rate falls to $86 at 5,000 tables and $47 at 10,000. If you are pricing a very large estate, Datadog's "starting at" language is the lever, because the published number does not bend on its own.
The full breakdowns: Monte Carlo data pricing for the credit rate card, Bigeye pricing for the two packages, Metaplane vs Monte Carlo for the head-to-head, and data observability pricing for every vendor in one table.
When a flat plan beats $16 a table
Metaplane is well liked for a reason. It is quick to connect, its Free plan is a real free tier, and if your infrastructure alerts already run through Datadog, putting data incidents in the same on-call rotation is a genuine advantage that no independent tool can match. For a team in that position, $16 a table is a fair price for the convenience.
The per-table meter hurts a different team. They are looking at pricing because a renamed column broke three Looker dashboards, or because nobody could say what a table fed before it was dropped. Those are lineage and change-impact problems. Datadog gives you the lineage graph at no charge, but the thing that tells you a change is dangerous, an alert, is metered per table, and the 300-model dbt project that needs it most is $57,600 a year at list.
Datatrail is priced the other way round. It connects to Snowflake, Databricks, BigQuery, Redshift or Postgres with a read-only role and parses query history and dbt artifacts into column-level lineage. Open any column and impact analysis lists every model, exposure and dashboard that reads it, before the change merges. Freshness monitoring and schema change alerts run on the same graph, ranked by what each failure breaks downstream. Plans are flat at $99, $299 or $799 a month on the pricing page, with no count of tables to forecast.
It is a narrower product than Datadog, with fewer monitor types and no infrastructure monitoring. If you want the whole platform, the honest comparison is on Metaplane alternatives, and the wider field is in data observability tools.
Metaplane and Datadog pricing questions buyers ask
How much does Metaplane cost?
Metaplane is free for up to 10 monitored tables. Pro is billed per monitored table up to a cap of 100 tables, and metaplane.dev does not print the rate. The nearest published price is from Metaplane's owner: Datadog lists Data Observability Quality Monitoring at $16 per monitored table per month billed annually, $21 month-to-month and $24 on-demand, which is $19,200 a year for 100 tables on the annual rate.
How much does Datadog Data Observability cost?
Datadog prices Data Observability as two products. Quality Monitoring is $16 per monitored table per month billed annually, $21 billed month-to-month and $24 on-demand. Jobs Monitoring is $0.05 per host per hour for Databricks and Spark clusters and $0.10 per job per hour for Airflow and dbt, both billed annually, with Databricks serverless on request. Lineage and metadata syncing carry no separate charge.
What counts as a monitored table in Datadog?
A table only counts once you add a monitor to it. Syncing tables for lineage and metadata is free. Every standard monitor on a table, whether freshness, row count, nullness or uniqueness, counts as one monitored table together, however many you add. Each unique custom SQL monitor adds one more. So a table with freshness, row count and three custom SQL monitors bills as four monitored tables.
Is Datadog Quality Monitoring the same product as Metaplane?
Datadog acquired Metaplane on 23 April 2025, and metaplane.dev still sells Metaplane by Datadog with its own Free, Pro and Enterprise plans. Datadog's Quality Monitoring uses the same meter (monitored tables), the same monitor types and column-level lineage from query history. Datadog's pricing page does not say Metaplane Pro is billed at the Quality Monitoring rate, so ask which contract you are signing before you use $16 as your number.
Is there a minimum number of monitored tables on Datadog?
No. Datadog's pricing FAQ says there is no minimum and you can start with one monitored table. Usage is measured every hour and you are charged on the average number of monitored tables over the month, so turning monitors on for a two-week migration costs roughly half a month for those tables, not a full month.
Does Datadog charge extra for data lineage?
No. Datadog states that lineage, schema and metadata syncing are included at no additional charge, and that column-level lineage is available for tables you do not monitor. You pay only for monitored tables and any Jobs Monitoring usage. That makes the lineage graph free to browse, but putting a monitor on a table costs $16 to $24 a month for that table.
Is Datadog Data Observability cheaper than Monte Carlo?
At list and on annual billing, Datadog is slightly more expensive per table than Monte Carlo's Scale tier and well above Start. For 1,000 tables Datadog is $192,000 a year against $178,850 on Monte Carlo Scale and $114,975 on Start. The gap widens with scale, because Monte Carlo's rate card falls in brackets after 1,000 tables and Datadog publishes one flat per-table price.
How much more is month-to-month or on-demand billing?
Month-to-month Quality Monitoring is $21 per table, 31% more than the $16 annual rate, and on-demand is $24, exactly 50% more. For 300 monitored tables that is $57,600 a year on an annual commitment, $75,600 month-to-month and $86,400 on-demand. The same 50% on-demand premium applies to Jobs Monitoring.
Is Datadog Quality Monitoring available in GovCloud?
No. Datadog's pricing FAQ says Quality Monitoring is available in its AP, EU and US data centers and is not currently available in GovCloud (US). Jobs Monitoring has the same restriction, and both products are marked as unavailable on the US-FED site in the pricing navigation. Federal teams need a different tool.
What is a cheaper alternative to Metaplane for lineage?
If lineage and change impact are what you need, a flat-priced lineage-first tool avoids the per-table meter entirely. Datatrail connects read-only to Snowflake, Databricks, BigQuery, Redshift or Postgres and includes column-level lineage, impact analysis, freshness monitoring and schema-change alerts on plans from $99 to $799 a month, with no table count. At 100 Datadog-monitored tables the annual rate is already $19,200.
Where these numbers came from
Both pricing pages render their prices in the browser, so a plain fetch returns navigation and no numbers. We rendered them with headless Chrome on 16 September 2026 and read the text. On Datadog's comparison page the site selector was set to US (US1, US3, US5).
DD=https://www.datadoghq.com/pricing
google-chrome --headless=new --dump-dom \
--virtual-time-budget=25000 "$DD/list/"
# Data Observability: Quality Monitoring
# Per monitored table, per month $16 $21 $24The monitored-table definition, the hourly averaging, the lineage inclusion and the GovCloud exclusion are quoted from the FAQ on the Quality Monitoring pricing card at datadoghq.com/pricing/?product=quality-monitoring. Supported sources come from Datadog's Quality Monitoring documentation: Snowflake, Databricks, BigQuery, Redshift, Iceberg on AWS Glue and PostgreSQL, with lineage from Airbyte, Fivetran and Hightouch and downstream impact in Looker, Tableau, Sigma, Metabase and Power BI. Metaplane's plan limits come from the comparison table on metaplane.dev/pricing, rendered the same way.
Three limits, stated plainly. Datadog does not say that Metaplane by Datadog contracts are billed at the Quality Monitoring rate, and Metaplane's page still reads "pay per monitored table" with no figure, so the $16 is the owner's published price for the equivalent meter, not a Metaplane Pro quote. Every total is list with no discount, and "starting at" signals that larger commitments negotiate lower. And prices change: these are correct as read on 16 September 2026. Monte Carlo and Bigeye comparison figures come from this site's own readings on 15 September 2026.
Related pages
What Metaplane does well, the 100-table Pro ceiling, and the realistic shortlist when you move off it.
Both published per-table prices side by side, with the capability differences that decide it.
$0.18 to $0.28 per credit from Monte Carlo's own order forms, multiplied out per table.
$45,000 for 100 tables and $75,000 for 300 on two cloud marketplaces.
Every vendor in the category that publishes a verifiable price anywhere, in one table.
The full field compared on lineage depth, monitoring, deployment and price.
See your lineage before you sign up for a per-table meter
Connect your warehouse with a read-only role and get a column-level map of what feeds what, down to the dashboards. If lineage and impact analysis are what you were pricing, you can have them this afternoon on a flat plan.