Skip to content

Interpreting Analytics

Learn how to read ValuePact charts correctly, avoid common misinterpretations, and communicate confidence levels to stakeholders. This guide is essential for anyone who presents analytics to decision-makers.

Who this is for

End User Admin Executive

Prerequisites

How to read charts

Confidence scores

Confidence scores are derived from claim validation, formula evaluation, and benchmark coverage. Use this table when communicating to stakeholders:

Score Meaning Action Safe for external sharing?
85% or higher High confidence Safe to share with external stakeholders Yes
60–84% Medium confidence Add more evidence or validate assumptions before sharing With caveats
30–59% Low confidence Requires significant additional validation No
Below 30% No confidence Do not use in a business case without rework No

Percentiles in benchmarking

  • p10 — 10% of peers are below this value. You are outperforming 90%.
  • p50 (median) — Half of peers are above, half below. This is the middle of the market.
  • p90 — 90% of peers are below this value. Only 10% are higher.

Your percentile tells you where you stand, not whether the value is good. A 95th percentile cost reduction is excellent; a 95th percentile time-to-value may indicate a problem.

Variance decomposition

Variance is additive across drivers. A driver with a large positive variance can mask another with a large negative variance. Always inspect the per-driver table before drawing conclusions about overall performance.

Variance type Visual indicator Interpretation
Positive Green bar Actual exceeded forecast (favorable for revenue, unfavorable for cost)
Negative Red bar Actual fell short of forecast (unfavorable for revenue, favorable for cost)
Near zero Gray bar On target

Common misinterpretations

  • Correlation is not causation. A driver that moves with outcomes may not be the cause. Check evidence provenance in the Governance → Evidence workspace.
  • Aggregate ROI hides risk. A portfolio ROI of 2x may include one initiative at 10x and nine at zero. Use the portfolio dashboard to inspect distribution.
  • Forecast accuracy degrades over time. A model that was 95% accurate last quarter may be 70% accurate next quarter. Re-validate assumptions regularly.
  • Benchmarks are directional, not targets. Peer medians describe what others achieve, not what you should achieve. Your strategy may justify outperforming or underperforming the median.
  • Anomalies are not always errors. A flagged point may represent a genuine one-time event (for example, a major contract win). Investigate before discarding.

Communicating to different audiences

Focus on realization rate, portfolio ROI, and confidence scores. Use the Executive Dashboard export. Avoid driver-level detail unless asked.

Present the CFO View with cost-benefit tables, payback periods, and risk remediation items. Include scenario bounds.

Share the Technical View with evidence provenance, implementation details, and remediation items. Include claim traceability.

Use the Team Dashboard and Individual Dashboard to show task progress, milestones, and blockers.

Tip: Always pair metrics with narratives

A chart shows what happened. The Narrative tab in Value Studio explains why it happened. Use both in stakeholder presentations. A slide with only numbers invites misinterpretation.

Permissions required

Role Permission Scope
User View / Export Assigned initiatives
Admin Configure confidence thresholds / Set narrative templates Tenant-wide
Executive View / Export Portfolio

Limits and guardrails

Limit Confidence thresholds are editable only by admins in Workspace Settings. Limit Chart exports are limited to 50 per day per user. Limit Narrative templates are limited to 10 per tenant.

Troubleshooting

Issue: Confidence score differs between dashboards

Cause: Different dashboards use different confidence inputs (claim validation vs. formula evaluation vs. benchmark coverage). Resolution: Open the Governance → Evidence tab to inspect the underlying truth objects and their maturity levels. The most restrictive score is the safest to report.

Issue: Stakeholders question the forecast

Cause: The forecast was presented without scenario bounds or sensitivity analysis. Resolution: Include pessimistic, base, and optimistic cases from the ROI Calculator in your presentation. Show the sensitivity of NPV to key variables.

Issue: Benchmark percentile confuses stakeholders

Cause: Stakeholders interpret percentile as a grade rather than a rank. Resolution: Explain that percentile shows relative position, not absolute quality. Pair with the peer range and sample size for context.

Issue: Variance chart is hard to read in presentations

Cause: Too many drivers are displayed, or colors are not accessible. Resolution: Filter to the top 5 drivers by absolute variance. Use the exported PNG instead of a screenshot for better resolution.

Escalation path

For training requests or help with stakeholder presentations, contact your customer success manager or open a support ticket with severity P4.