Plans & pricing
Your team ships the code. Our team builds the replay suites, maintains the mocks, and triages the failures. Pay for the services covered and your usage.
From running it yourself to handing us the suite. Every plan uses Speedscale’s traffic replay platform.
Your team runs it
For teams that want to build and run their own production-realistic tests.
We get you running
For teams that want an experienced engineer to help build the first suites.
Onboarding + usage
Quoted setup fee and committed creditsWe own the testing work
For teams that want test coverage delivered and maintained for them.
Per service + usage
Quoted service fee and committed creditsSaaS: $0.01 per MB of new ingest, totaled monthly. BYOC: billed at $0.01 per memory-GiB-hour under an annual capacity license. See the calculation. Assisted onboarding and Managed service fees are quoted separately.
Plan your usage
Choose where your traffic lives. See what we measure and how your estimate changes as you grow.
Use the total billable data sent to Speedscale Cloud during a month. 1 GB = 1,000 MB.
$0.01 per MB ingested = $10 per GB. Usage accumulates during the month. Recording windows and traffic filters help control the volume.
For cluster-wide installs, use total node memory capacity. For namespaced installs, use effective pod memory requests across authorized namespaces. 1 GiB = 1,073,741,824 bytes; 1 GB = 1,000,000,000 bytes.
For each UTC day, add each cluster’s highest reported metered RAM. The largest daily sum is the monthly peak. Cluster-wide installs count every node reported by Kubernetes, including nodes without captured applications.
Estimated SaaS usage charge
$100.00 / month
$1,200.00 / year at the same monthly usage
Enter a valid amount of zero or more within the input’s range.
10 GB × 1,000 MB/GB × $0.01/MB = $100.00 / month
Usage only. Assisted onboarding and Managed service fees are quoted separately. Commitments and overage terms follow your agreement.
Discuss your estimateRequest and response payloads, headers, metadata, enriched context, and enabled replay orchestration traffic count toward ingest. Keeping a recording does not re-ingest it each month. Reusing it in a replay adds test artifacts and orchestration data rather than charging for the original recording again.
For cluster-wide installs, adding nodes, choosing larger nodes, or adding clusters can raise your monthly peak. Namespaced installs measure effective pod memory requests across authorized namespaces. Both measure capacity or reservations rather than live memory consumption. A short scale-up can set the peak for the month.
Billed at $0.01 per memory-GiB-hour under an annual capacity license. The estimate uses 8,760 hours per year, or 730 hours per month on average. Running a workload for fewer hours does not prorate this capacity estimate. Your cloud provider bills infrastructure and storage separately.
SaaS example
10 × 1,000 × $0.01 = $100. At the same volume for 12 months, usage totals $1,200. Increase ingest to 100 GB per month and the estimate becomes $1,000 per month, or $12,000 per year.
BYOC example
100 GiB × 8,760 hours × $0.01 = $8,760 per year, equivalent to $730 per month. At 200 GiB, the estimate doubles to $17,520 per year. These are annualized capacity estimates; your quote defines the license commitment and capacity changes.
For the meter, if two clusters peak at 160 GiB and 32 GiB on the same UTC day, the daily sum is 192 GiB. If every other day is lower, that is the monthly peak, even after scaling down.
Managed pricing combines ownership of your test suites with usage measured by SaaS ingest or BYOC RAM.
A base fee per covered service for suite creation, maintenance, triage, and agreed SLAs.
A prepaid annual pool for the usage in your agreement, sized using your proof of concept. Covered usage draws down this balance.
Usage beyond your commitment, billed monthly at the overage rate agreed in your quote.
Prepaid credits pay for covered usage; they are not an extra charge on top of that usage. Metering, rates, caps, and rollover terms are set in your agreement. An alert is a notification; a cap or pause must be agreed separately. Failed tests and bugs found are never billing units.
See what your team takes on, and what Speedscale takes off your plate.
Scroll horizontally to compare all three plans on smaller screens.
| Responsibilities & commitments | Platform | Assisted | Managed |
|---|---|---|---|
| Who owns the replay suites? | Your team | Shared | Speedscale |
| Suite creation and CI setup | Self-serve | Guided onboarding | Done for you |
| Ongoing suite and mock maintenance | Your team | Shared | Speedscale |
| Failure triage | Your team | Shared | Speedscale |
| Readiness reports with AI code quality | Your team prepares | Shared | Delivered by Speedscale |
| Platform availability SLA | Included | Included | Included |
| Support response and onboarding SLAs | Not included | Agreed at kickoff | Agreed at kickoff |
| Triage, maintenance, and test quality SLAs | Not included | Not included | Agreed for your scope |
AI writes more code. You still need to know how it behaves. Your Managed readiness report tells you whether each release is ready for production, with coverage, findings, and consumption in one place.
We establish a baseline during your proof of concept, then agree on the metrics that matter for your services.
Discuss your proof of conceptManaged commitments cover the work we control. We set targets, measurement, and service-credit remedies together before kickoff.
A defined turnaround for investigating failed runs, with priority for release blockers.
Response times matched to issue severity and the support coverage you need.
An agreed milestone for your first replay suite running in CI, once access is in place.
An agreed false-positive threshold, so your team can trust the failures we report.
Business outcomes such as escaped defects, release frequency, and rollback rates are reported against your baseline. Exact SLA targets and capped credit remedies are defined in your agreement.
SaaS usage costs $0.01 per MB ingested, equivalent to $10 per GB. We total ingest by month. BYOC (bring your own cloud) is billed at $0.01 per memory-GiB-hour under an annual capacity license. Annual cost is GiB of capacity × 8,760 hours × $0.01; the monthly equivalent is one twelfth of that. Assisted adds separately quoted onboarding. Managed adds a quoted fee for each covered service. Commitments, capacity changes, overage rates, and billing terms follow your quote.
The rate is per MB of new billable ingest sent to Speedscale Cloud. Usage is totaled monthly. It includes request and response payloads, headers, metadata, enriched context, and enabled replay orchestration traffic. Keeping the same recording does not ingest it again each month. Replay test artifacts can add ingest. Estimates use 1 GB = 1,000 MB, and an annual estimate assumes the same ingest every month.
Cluster-wide installs measure total Kubernetes node memory capacity; every node reported by Kubernetes counts. Namespaced installs measure effective pod memory requests across authorized namespaces. Each UTC day, we sum the highest reported metered RAM for each cluster; the largest daily sum is the monthly peak. Multiple forwarders in the same named cluster do not multiply its capacity. A brief scale-up can raise the monthly peak. RAM is expressed in GiB: 1 GiB = 1,073,741,824 bytes. The annual capacity license uses that capacity basis; the hourly rate does not mean the meter tracks workload running hours. Your quote defines capacity commitments and adjustments.
For each service in your agreed scope, Speedscale builds and maintains replay suites and dependency mocks, runs validation in your release workflow, triages failures, and reports coverage and results. We agree on the service list, API surface, environments, and responsibilities during onboarding.
Committed credits prepay the usage covered by your agreement. Covered usage draws down that balance; it is not charged again on top of the credits. Your agreement defines which usage is covered, the commitment, and overage terms. SaaS ingest is measured monthly; BYOC uses cluster capacity under an annual license at $0.01 per memory-GiB-hour. We use proof-of-concept data to size the commitment. We never charge per failed test or per bug found.
Committed plans include an alert at 80% of the credit pool. Additional usage is billed monthly at the overage rate in your agreement. Spend caps or an auto-pause option help control costs; pausing validation also pauses testing coverage. Unused committed credits can roll over for one term, as specified in your agreement.
During onboarding, we agree on a way to identify AI-assisted changes, such as pull request labels, commit trailers, or data from your coding tools. Managed readiness reports then break out failures and unintended API behavior changes in AI-assisted code alongside other changes. The comparison depends on that agreed tagging method.
SLAs cover agreed work that Speedscale controls, such as triage turnaround, support response, onboarding milestones, and test signal quality. Exact targets, measurement, support hours, and capped service-credit remedies are defined in your agreement. Escaped defects, release frequency, and rollback rates are tracked as outcomes against your baseline; they are not contractual guarantees.
Yes. proxymock remains the free option for local traffic recording, mocking, and replay. Start locally and move to a Speedscale plan when you need cloud workflows, guided onboarding, or managed test coverage.
Bring your critical APIs and release workflow. We’ll scope the coverage, measure usage, and put a plan together.