Maple

Maple is an OpenTelemetry observability platform for distributed systems that ingests traces, logs, and metrics and stores them in ClickHouse. Teams can examine distributed traces as request spans, correlate them with logs and service data, and search structured logs by severity, service, or attribute. A log can link directly to its trace. Metric tools include exploration, dashboards, service maps, and alerting. Browser session replay captures clicks, routes, console lines, and failed requests, connecting sessions with traces through a shared ID. Alerts can go to Slack, PagerDuty, Discord, Telegram, email, webhooks, and Hazel. A hosted MCP server and documented REST API support automation for searching telemetry, triaging errors, building dashboards, and managing alerts. Maple offers cloud, self-hosted, and local options; Maple Local runs ingestion, an embedded database, query API, and dashboard as one binary on a developer machine. Self-hosted and Local plans are free. Startup costs $39.00 per month and includes 30-day retention; Enterprise starts at $2,000 by contact and offers custom retention.

Maple brings several telemetry types and browser replay into one observability platform, with cloud, self-hosted, and local deployment choices. Review the 30-day Startup retention and fixed cloud region selection before choosing a cloud setup.

Maple suits development teams that need to inspect traces, logs, metrics, and browser sessions together. It may also fit teams that want to run observability software locally or on infrastructure they operate.

Maple is an OpenTelemetry platform for collecting traces, logs, and metrics across distributed systems, with browser replay tied to telemetry. It suits teams that want those signals in one place and value a choice of cloud, self-hosted, or local deployment. Its breadth is appealing, though the Startup plan’s 30-day retention and fixed cloud region deserve close attention.

Overview

Maple stores telemetry in ClickHouse and connects related signals: engineers can move from a structured log to its trace, while traces show request spans alongside logs and service data. That shared view can help teams investigate an issue across services without treating each telemetry type as a separate workflow.

Browser session replay adds a user-facing view of failures. It records clicks, routes, console lines, and failed requests, and links a session to traces through a shared ID. Product events are unlimited and free during beta, but synthetic monitoring is not part of the offering.

Key features

Telemetry and investigation

Maple supports trace exploration, structured log search by severity, service, or attribute, and metric dashboards, service maps, and alerting. This makes it a credible fit for teams standardizing on OpenTelemetry who need both investigation and operational visibility. Alert rules can send notifications to Slack, PagerDuty, Discord, Telegram, email, webhooks, or Hazel.

Integrations and automation

SDKs and guides cover Effect, Node.js, Next.js, Python, Go, Rust, Java, C#, Kotlin, and Laravel; integrations also include Cloudflare, PlanetScale, GitHub, WarpStream, and Hazel. A hosted MCP server and documented REST API let agents and automation search telemetry, triage errors, build dashboards, and manage alert rules. Teams with these workflows can connect Maple to automation; teams needing synthetic checks will need another tool for that job.

Deployment and data control

Maple Cloud offers US and EU regions, with EU telemetry stored and processed in Frankfurt. Region selection is permanent, so the choice needs to match a team’s residency needs from the outset. Self-hosting gives customers control over infrastructure, retention, and residency. The source is published under FSL-1.1, with each release converting to Apache 2.0 after two years. Maple Local packages ingestion, an embedded database, query API, and dashboard into one binary that runs without an account or cloud service.

Pricing

Maple is freemium, with paid plans from $39/mo and a 14-day trial. The free Self-hosted plan lets teams run the source on their own infrastructure, leaving retention and residency to them. Local is also free and runs on one machine as a single binary, without an account, cloud dependency, or rate limits. These options avoid a subscription but put self-hosted infrastructure under the customer’s control, while Local is intended for one developer machine.

Startup costs $39 per month and includes 100 GB each of logs, traces, and metrics, plus 5,000 browser sessions. Additional signal ingestion is $0.30/GB per signal, and additional sessions are $0.002 each. Product events are unlimited and free during beta. The plan has 30-day retention and a private support channel. It fits teams that can work within those quotas and retention window; growing usage can add metered costs.

Enterprise starts at $2,000 with custom pricing. It offers higher volume, custom retention, priority support, and the option to run in the customer’s own cloud account. That makes it the relevant tier for organizations that need more headroom, longer or tailored retention, or that cloud arrangement.

Platforms

Maple supports API, iOS, Linux, macOS, self-hosted, and web environments. Its Local binary is aimed at developer machines, while self-hosting supports deployments on customer-operated infrastructure.

Who it's for

Maple is best suited to engineering teams that want traces, logs, metrics, and browser replay connected in one OpenTelemetry-oriented platform, especially when they also want a self-hosted or local option. Its cloud Startup plan is less suited to teams that need retention beyond 30 days, a changeable region, or predictable costs above the included quotas.

Pros and cons

• Pro: Traces, searchable logs, metrics, and browser sessions can be connected, giving teams multiple angles for investigating service and browser problems.
• Pro: Cloud, customer-operated self-hosting, and a no-account local binary cover distinct deployment needs.
• Pro: Startup includes substantial per-signal ingestion and browser-session quotas, with product events free and unlimited during beta.
• Con: Startup retention is 30 days; custom retention is an Enterprise feature.
• Con: Maple Cloud’s selected region cannot be changed later, which makes initial residency planning consequential.
• Con: Startup charges per signal for ingestion above 100 GB and per session above 5,000, so usage can raise the monthly bill.
• Con: Synthetic monitoring is not included.

Alternatives

Compare observability platforms if you want to survey the category before choosing a product.

• Dynatrace Kubernetes Monitoring is a paid option with a free trial and hourly pod-based Kubernetes monitoring pricing.
• Honeycomb is another freemium choice; its free tier includes up to 20M events and 100M metric data points per month, plus two triggers.
• KloudMate offers a free tier with capped monthly logs, traces, metric samples, and seven-day retention.
• Middleware has a free tier that includes data ingestion, RUM sessions, synthetic checks, and browser test runs.
• SigNoz is an alternative with a free self-hosted community edition.
• New Relic Session Replay is a freemium alternative whose free plan includes 100 GB of monthly ingest and one full platform user.
• Axiom offers a free tier with 500 GB monthly ingestion and 30-day retention.
• Grafana Cloud Application Observability is a freemium alternative with a free allowance of 2,232 host hours.

Verdict

Choose Maple if your team wants connected OpenTelemetry signals and browser replay, with the flexibility to run in the cloud, on your own infrastructure, or locally. Its unified investigation tools and deployment choices are the strongest reasons to choose it. Look elsewhere if you need synthetic monitoring, longer retention without an Enterprise plan, or the ability to change a cloud region after setup.