If you're considering a 30-min ROI call, these are the questions we get asked first. Read them; if you still have questions left, those are the ones worth the call.
Last updated 2026-05-20 · Replaces ~80% of pre-sale call questions in one read.
Read-only, always. We never write to, modify, or delete anything in your tracker, repository, or CI system. Jira API token needs the inherited project read scope (no extra grants) — this is the only primary tracker we support. GitHub: classic personal access token with repo + read:org + workflow read. GitLab: personal access token with read_api. Azure DevOps Repos & Pipelines (secondary code platform only): PAT with Code (Read), Build (Read), Release (Read) — Work Items scope is NOT required. Full token-scope documentation goes out before any pilot starts.
First projections in <5 minutes. After connecting your tracker, we pull your last 90 days of issue, commit, and pipeline history; ROI projections from that historical data appear immediately. Your measured savings (the immutable Quality Cost Ledger) starts populating after you capture your first baseline plus ~2 weeks of correlated activity. Full team-scoped per-team P&L: another week if you have multiple teams and want repo-to-team mapping.
Yes, with caveats. The Correlation Engine relies on a few fields you almost certainly already have: issue type, status transitions, created/resolved timestamps. Bug-vs-feature classification falls back to issue-type labels you can map at onboarding (a 30-second config step). Missing fields are filled in with smart defaults (industry-benchmark ratios) that the engine flags as projected rather than measured. The cleaner your data, the higher the confidence score on every entry; messy data still produces usable output, just with lower confidence weighting.
They measure engineering throughput. We measure engineering economics. Velocity dashboards tell you how fast you ship; QualityProfit tells your CFO what shipping cost and what was prevented. The Quality Cost Ledger is the differentiator no one else has: an immutable, baseline-tagged audit trail that finance can quote in board meetings. If you only need "are we shipping faster?" pick one of the velocity tools. If you need "can I defend the QA budget to the board?", that's us.
Read-only across the board. No outbound data leaves your environment except license validation (a sub-1KB ping per day to api.qualityprofit.com). All metrics are computed in your VPC; we self-host nothing on your behalf. Your security review needs to validate: (1) the read-only token scopes above, (2) the outbound license-validation domain, (3) the Docker image source (ghcr.io/shiroo81). Most reviews complete in <2 days.
Every euro is traceable. The Correlation Engine assembles "Quality Impact Chains": a chain is one or more events across Jira, Git, and CI, linked by referential keys (issue keys, commit hashes), temporal proximity, or heuristic signals. Each event carries a confidence score. The chain's total impact = sum of (event cost × confidence score). Costs anchor in your team economics (team size, developer rate) and industry-published benchmarks (DORA, Boehm, Capers Jones, NIST, CISQ). One worked example: a single commit → build failure → bug → release delay chain books €28,840 with 0.78–0.95 per-event confidence. The whitepaper walks through the formulas and benchmark sources end-to-end.
Yes — but minimally. QualityProfit ships as a self-hosted Docker stack (backend + frontend) that runs inside your VPC. The only outbound dependency is license validation. There's no SaaS multi-tenant database, no telemetry, no analytics pixels. Most IT reviewers approve us in one short call after seeing the Docker Compose file and the firewall rules. The hard-mode IT teams (regulated industries) take ~1 week with our standard SOC-style answer pack.
Yes — the Live Demo runs against a representative dataset. No login, no tokens. Walk every tab, every screen, every ROI breakdown. Once you've seen what we measure and you want it pointed at your own data, the 30-min ROI call is the natural next step.
Honestly? The math is hard to defend. The ROI floor we see in production is "one defect avoided per year covers the subscription." Below ~60 engineers, defect volume is usually small enough that the math gets uncomfortable and the audit-trail value gets thin (you already know which team shipped the bug — there are only one or two). We'd rather not waste your time or ours. Come back when you're past 60 engineers, or when your CFO starts asking specific questions about QA spend. We'll still be here.
Going forward, no. The Quality Cost Ledger is now an immutable audit trail — baseline resets save the old baseline to history and preserve every existing entry. Each entry carries the baseline it was calibrated against, so the ledger naturally spans multiple "eras" of your operation. If you reset your baseline before mid-May 2026, the entries booked under the original baseline were lost in that earlier behavior. From your next reset forward, the trail is permanent.
Still have questions?
If something here didn't answer it, that's exactly what the 30-min call is for. We'll show you the math against your own data — or tell you you're not a fit, which is also useful.