Where competitors fall short

DatadogOwn-team experience

$8,500 a month, with value under scrutiny

“We're paying more for the insurance than the damage.”

Reports that Datadog caught two otherwise-missed production incidents in a quarter, and questions the cost relative to the impact. Describes reducing monitoring on less critical systems. A commenter flags that the post’s cost-per-incident calculation mixes monthly spend with quarterly incidents.

Relevance: a concrete cost concern and an active effort to reduce observability spend. This does not establish an open purchasing process.

DatadogOwn-team experience

Lower costs, and a production issue Datadog missed

“A production heap leak that Datadog's AI saw nothing about.”

Reports cutting monthly observability costs from $5,700 to $285–$300 six months earlier. Says the team found the heap leak using TraceQL and Loki log correlation.

Dissatisfaction with cost and diagnostic coverage after changes already made. A commenter notes that the stated savings do not account for engineering time spent maintaining the self-hosted stack.

Grafana CloudClient experience

A $15–20k monthly bill reduced to around $500

“After setting things up properly, the bill dropped to around $500.”

Describes helping a client whose default Grafana Cloud Helm configuration led to substantial telemetry charges. Criticizes collecting large amounts of data that teams never use.

A specific client example, not an identified buyer. Trayo’s post search lists Techstars, while its thread profile lists Partner at Pelotech; affiliation is inconsistent.

DatadogComment

Easy to use. Hard to justify the bill.

“I love datadog for the ease of use. I hate datadog for the billing.”

Criticizes separate charges for ingestion, retention and features, saying the bill rises sharply once the tool delivers value.

Explicit billing frustration from a relevant platform engineering leader; no stated buying timeline.

DatadogComment

Switched after Datadog exceeded the AWS bill

“the data dog bill surpassed the whole AWS bill!”

Says one production service generated a Datadog bill larger than the whole AWS bill, prompting a switch. Also praises Datadog’s detail and quality.

Firsthand account of a completed switch; the affected organization or client is not identified.

DatadogComment

Licensed features make the cost hurt

“If you're using it for logs, great, when you're using it for obscure licensed things, it hurts.”

Contrasts useful logging with painful costs for licensed features in a discussion about Datadog pricing.

Relevant SRE role and direct criticism, but no specific spend or switching plan.

DatadogCustomer anecdote

A tagging change reportedly tripled a customer’s bill

Recounts a conversation from the author’s time at PagerDuty: a customer restricted developer access after a tagging change caused a threefold bill increase.

Unnamed customer, not the author’s current company. The thread also mentions alternative vendors; treat as a lead for research, not a verified buyer.

Datadog / SplunkGeneral criticism

Monitoring overhead blamed for latency

Argues that log serialization and monitoring agents can consume CPU and contribute to garbage collection and latency problems.

No specific customer deployment is identified. The technical figures are unverified author claims.

DatadogSecondhand commentary

How invoices can outgrow infrastructure spend

Summarizes a Reddit discussion about metric cardinality, log indexing, span indexing and host-tier pricing driving unexpected costs.

A summary of other people’s discussion, not evidence of the author’s own frustration or buying intent.

Datadog / Grafana CloudPricing analysis

One metric label can change the economics

Explains how adding high-cardinality labels can multiply the number of billable series without a corresponding increase in customer usage.

General analysis rather than a named customer incident. The author’s company is below the saved ICP’s 50-employee minimum.

DatadogQuoted opinion

Criticism of charging for repeated data

Shares Charity Majors’ criticism of Datadog’s data storage and charging model.

A quoted opinion from an observability competitor, not a firsthand complaint by this author. Kept as context only.