Inspectable inputs
See which values are observed, prior-backed, or missing. Ask to inspect a source date, a coverage limitation, and the effect of a disputed assumption. Review the input checklist before connecting systems.
Capability
Cyber Risk Quantification Software
Compare cyber investments by modeled annual loss, action cost, and budget. Inspect the evidence, missing inputs, and assumptions behind each recommendation.
Prepared by ValtyUpdated
Evaluate with a real question
What to look for in cyber risk quantification software.
Ask the vendor to carry one funding decision from evidence to a comparison, then show how approval and subsequent verification are recorded. In Valty’s supported evaluation, agree the company and available capabilities before using your own data.
Comparable choices
Compare no optional action, required work, and priced alternatives under one budget and one loss objective. Check whether an option changes when the objective moves from mean annual loss to P95.
Governed follow-through
Inspect the approval owner, authorized scope, supported execution path, exceptions, and required verification. Confirm the exact integration and workflow in scope; a completed task is not proof of effective remediation.
A reviewable output
Ask for the assumptions, model version, loss units, alternatives, unresolved gaps, and next review. Use the board report template to assess whether the output supports an actual decision.
Valty is offered in a design-partner evaluation. Authentic application captures demonstrate individual interfaces with illustrative data; the synthetic decision memo is a separate, reproducible example. Neither is permissioned customer-outcome evidence or a guarantee of loss reduction.
Design partner
Carry the funding decision into approved work.
Review the proposed action, the person who can authorize it, and the evidence required to verify the result. Funding, execution status, and demonstrated risk reduction remain distinct.
Where the boundary sits
Modeled reduction is not observed savings or a guarantee. Exact metrics, integrations, uncertainty treatment, and execution capabilities depend on the agreed evaluation scope and available product configuration. The sample is not independently calibrated customer evidence.
Best next step for PE operating partner, CFO, board: choose “Request a platform demo” to review the workflow, evidence boundary, and fit for your environment.
Fictional worked example
Inspect the assumptions behind one funding decision.
This excerpt of the sample memo shows one fictional company's annual gross-loss distribution. Its mean, median, and P95 describe different modeled outcomes; they do not establish confidence in the input assumptions.
Fictional sample · 2026-09-05
Example Manufacturing Co.
P95 annual loss · no insurance recovery modeled
Baseline from the fictional sample memo. Synthetic assumptions, separate from the actual application captures and not a customer forecast.
About 5% of modeled years exceed P95. It is an outcome percentile, not a worst case or a confidence interval. These synthetic inputs have not been externally calibrated.
Fictional baseline assumptions
Reproduce the estimate.
- Annual event frequency
- 0.8 events/year
- Mean loss per event
- $1,000,000
- Loss variability
- 1.2 log-scale standard deviation
- Simulation
- 50,000 modeled years
These are synthetic inputs, not measured company data. Changing frequency or severity changes the loss distribution and can change which investment is preferred.
Inspect the full assumptions and alternatives →Inspectable CRQ
See what is measured, what is prior, and why the dollar moved.
Every entitled figure exposes its evidence mix, the next source that can tighten it, and the exact model snapshot behind the output. Reviewers can challenge a node instead of accepting a black-box score.
Coverage
Measured / prior / missing stays visible
The coverage meter names the evidence mix and the next connector that would replace a missing or prior-backed node.
Challenge
Pin a formula node and see the dollar delta
Every formula-tree node is labeled. Pinning a measured node to prior updates the displayed dollar and records a provenance row.
Replay
Recompute a published p95 from its snapshot
The evidence snapshot and model version replay the printed figure without requiring a live connector.
Sufficiency
Thin data widens the band and creates an uplift plan
Low input sufficiency changes the uncertainty treatment and ranks the next source to connect instead of hiding the gap.
Illustrative coverage mix
33% measured · 33% prior · 34% missingMFA enforcement
Okta API
EDR response time
Connect EDR
Cloud path state
Connect CSPM
Next evidence move: connect the source attached to the highest-impact missing node.
Availability
Inspect the public method Design-partner workflow. Access, connected evidence, and the recorded model version determine what can be reviewed. Dollar outputs remain decision-support estimates.
One company. One funding question.
See the decision workflow in a platform demo.
In 30 minutes, review the loss model, compare funding choices, and inspect the approval and evidence trail using illustrative data. Agree your company scope, supported workflows, and evaluation terms before connecting your own evidence.