Datadog Synthetics Pricing and Datadog Synthetic Monitoring Pricing With Cost Per Uptime Check

Datadog Synthetics pricing is metered per test run. API tests, which is what an uptime check is, cost $5 per 10,000 runs billed annually, $6 month to month and $7.20 on demand. Browser tests cost $12, $15 and $18 per 1,000 runs. Every hit from every location is a run, so one URL checked every minute from one location is 43,200 runs a month, or $21.60 at the annual rate.

That per run model is fair for a handful of critical checks and expensive for a fleet of simple ones. The rest of this page puts real numbers on it: the full rate card from Datadog's own price list, how runs are counted, what uptime monitoring costs at 5 to 650 URLs, where phone call alerts add cost, and where a flat priced monitor wins.

How much does Datadog synthetic monitoring cost?

These are the list prices on datadoghq.com/pricing/list, in USD. Datadog shows three columns for each product: billed annually, billed month to month, and on demand.

Datadog's product page frames the same numbers as "starting at $5 per 10K test runs" for API testing and "starting at $12 per thousand test runs" for browser testing, with the on demand price in the footnote. Large customers with committed spend often negotiate below list, so read these as the ceiling, then check the rate in your own contract.

How are Datadog synthetic test runs counted?

Datadog's pricing FAQ is specific. HTTP, SSL, DNS, TCP and ICMP checks all fall under API test billing, and "each hit being made on your endpoint counts as a single test run." Its own example: a test set to run from 3 locations every minute counts as 180 test runs in an hour. Three more rules from the same FAQ matter for the bill:

• Multistep API tests count every step as its own API test run. A four step login and fetch flow run every minute is 172,800 runs a month.
• Browser tests count one run per 25 steps. A 50 step test is two runs on each execution.
• Private locations cost nothing to set up, but runs from them are billed exactly like runs from managed locations.

The FAQ does not say whether fast retries are billed as extra runs. If you turn retries on, watch the Synthetic Monitoring line in Plan and Usage for a week before you assume they are free.

The quickest way to reason about it is runs per URL per month. In a 30 day month:

The interval and the location count multiply each other. Going from a 5 minute check in one region to a 1 minute check in three regions is a 15 times increase for the same URL.

What Datadog uptime monitoring costs at 5, 25, 100, 250 and 650 URLs

Here is the part most teams only see after the first invoice. Each row assumes simple HTTP API tests checked every minute, list prices and a 30 day month. The TrackDowntime column is our own price for the same number of endpoints checked every 60 seconds from one location, with phone calls and SMS included.

Relax the interval to 5 minutes and Datadog gets much closer: 25 URLs from one location cost $108 a month at the annual rate, 100 cost $432 and 650 cost $2,808. That is a legitimate choice for pages where a few minutes of blindness is acceptable. It is a poor one for checkout, login or an API with an SLA, because a 5 minute interval can leave up to 5 minutes of every outage out of the record.

Now the honest crossover. At one minute from one location, Datadog is cheaper than our $99 Professional plan up to four URLs, and cheaper than Professional billed yearly ($49 a month) up to two. From five URLs upward the flat plan is cheaper, and from about fourteen URLs even our unlimited Enterprise plan at $299 costs less than Datadog's annual list price.

What phone call alerts add to a Datadog bill

Synthetic tests alert through Datadog monitors, which can notify people by email, Slack and other integrations. Phone calls and text messages come from Datadog On-Call, a separate product sold per seat. Its documentation lists push notifications, email, SMS and phone calls as contact methods, notes that phone support varies by country, and says SMS does not reach landlines, VoIP or virtual numbers.

On-Call is listed at $20 per seat a month billed annually, $24 month to month and $29 on demand. A team of three who all want the phone to ring adds $60 to $87 a month on top of the test runs. Seats are counted at the monthly peak and a swapped seat bills twice that month; the Datadog On-Call pricing breakdown has the seat rules, Incident Management at $30 and the $40 bundle. That is reasonable if you also want rotations, escalation policies and handoffs. If you do, and you are shopping beyond Datadog, a dedicated incident management tool that pages the right engineer is the category to compare.

If what you actually need is simpler, every person on a short list getting a call the moment a URL fails, a seat per phone is the expensive way to buy it. TrackDowntime calls and texts every number attached to an endpoint at once, and repeats at 5, 10, 15, 20, 30, 40, 50 and 60 minutes, then at 2, 4, 8, 24, 48 and 96 hours, until the site recovers. There are no seats and no message credits. The mechanics are on our downtime alerts by phone call page.

Which Datadog Synthetics tests are worth their cost

Not every test should leave Datadog. The per run model is well priced for tests that give you something only Datadog can:

• Browser tests on money paths. A 10 step login or checkout test every 15 minutes from 2 locations is 5,760 runs, about $69 a month at the annual rate. That is cheap insurance for a flow that returns 200 and still breaks.
• Multistep API tests that pass a token from one request to the next and assert on the body.
• Private location tests against internal services a public checker cannot reach.
• Tests tied to APM, where a failure opens straight into the trace that caused it.

The tests that are usually overpriced are single step HTTP checks whose only assertion is "status is 200." Those are often the majority of a synthetic test list: homepages, marketing sites, client sites, API base URLs, health endpoints. They produce a yes or no answer, and on a per run meter they produce it at full price every minute from every region.

How to cut Datadog synthetics costs without losing coverage

1. Export your test list and sort by runs. In Plan and Usage, find the tests with the most runs. Almost always they are one minute, multi location HTTP checks.
2. Match interval to consequence. A marketing page does not need three regions every minute. Five minutes from one location is 15 times cheaper.
3. Drop redundant locations. Two regions catch most regional false positives. Five rarely catch anything two do not.
4. Move plain status checks to a flat plan. Keep browser, multistep and private location tests in Datadog, and move the single step checks to a monitor priced per endpoint. We explain the side by side approach on our Datadog uptime monitoring alternative page.
5. Recheck the bill after a full month. Test runs are billed on usage, so the saving shows up on the next invoice, not the day you pause a test.

What you give up on a flat priced monitor

We would rather you move the right tests than all of them. TrackDowntime checks every 60 seconds from a single location. It has no browser tests, no multistep tests, no keyword or body assertions, no private locations and no status pages. It checks status code, response time, timeouts, DNS and SSL expiry, alerts by phone call, SMS, email, Slack and a signed webhook, and stores every incident with its start, end and measured duration, 90 days on Professional and with no limit on Enterprise, exportable to CSV.

If two or three URLs are all you watch, Datadog at the annual rate is cheaper than our Professional plan and you should stay. If you have dozens or hundreds of status checks on a per run meter, our plans and pricing page shows cost per endpoint from 5 to 650, and the wider uptime monitoring pricing comparison puts Datadog's model beside UptimeRobot, Pingdom, Better Stack and Uptime.com. For the head to head with Pingdom's flat tiers, including the 15 minute interval where Datadog comes out cheaper, see Pingdom vs Datadog Synthetics pricing. New Relic works the other way round for plain uptime: its ping monitors are not metered at all, and the bill is the user seats, which our New Relic synthetics pricing breakdown prices at 1 to 5 people.

Frequently asked questions about Datadog Synthetics pricing

How much does Datadog synthetic monitoring cost?

Datadog lists synthetic API tests at $5 per 10,000 test runs billed annually, $6 month to month and $7.20 on demand, and browser tests at $12, $15 and $18 per 1,000 runs. One URL checked every minute from one location is 43,200 runs a month, which comes to $21.60 at the annual rate and $31.10 on demand.

How are Datadog synthetic test runs counted?

Each hit on your endpoint from each location is one test run. Datadog's own example is a test run from 3 locations every minute, which counts as 180 test runs an hour. Each step of a multistep API test counts as a separate run, and browser tests count one run for every 25 steps.

Does Datadog charge for private locations?

No, there is no charge to set up a private location. Test runs executed from a private location are billed exactly like runs from Datadog's managed locations, so a test inside your network costs the same per run as one from a public region.

How much does a Datadog browser test cost?

Browser tests cost $12 per 1,000 runs billed annually, $15 month to month and $18 on demand, with each 25 steps counting as one run. A 10 step test every 15 minutes from 2 locations is 5,760 runs a month, about $69 at the annual rate. The same test every 5 minutes from 3 locations is about $311.

Does Datadog Synthetics include phone call alerts?

Not by itself. Synthetic monitors notify through email, Slack and other integrations. Phone calls and SMS come from Datadog On-Call, listed at $20 per seat a month billed annually, $24 month to month and $29 on demand. Every person who should be paged by phone needs a seat.

Is Datadog Synthetics cheaper than Pingdom for uptime checks?

Usually not for plain uptime at one minute. Pingdom's price list shows $149 a month for 100 uptime checks with unlimited users, while 100 Datadog API tests every minute from one location are $2,160 a month at the annual list rate. Datadog gets closer at 5 minute intervals, and it does far more beyond uptime. Our Pingdom pricing breakdown has every tier.

If most of your synthetic bill is status checks that only need to answer "is it up," create your TrackDowntime account, add those URLs with every number that should ring, and run it beside Datadog for a week before you pause anything.