How AI Readiness Peer Benchmarks Are Computed
Every benchmark shown on AIOpsNav is built from anonymized submissions, grouped by industry and size, with a minimum peer threshold and monthly data refresh.
AIOpsNav computes peer benchmarks from anonymized assessment submissions grouped by industry vertical and company size band (headcount or revenue). A minimum of five peers (N≥5) must exist in a peer group before a benchmark score is displayed — below that threshold, only aggregate tier-level data is shown. A "peer" is defined as a firm in the same industry vertical and the nearest size band. Benchmark data refreshes monthly. Every benchmark display includes the sample size (N) and the month the data was last refreshed. No firm-level identifying information is included in any benchmark computation or display. [VERIFIED] [LAST UPDATED: May 2026]
What Anonymization Means
Anonymization is not just a privacy promise — it is an architectural property of how benchmark data is stored and queried.
How it works: When an assessment is submitted, individual responses are aggregated into dimension sub-scores and a composite score. These numeric scores — not the underlying question responses — are stored in the benchmark pool. The firm name, contact details, and any personally identifying information are stored separately and are never joined to the benchmark computation pipeline. [VERIFIED]
This means the benchmark computation layer has access to: industry vertical, size band, dimension scores, composite score, and submission date. It does not have access to: firm name, email, individual question responses, or any data that could identify a specific organization.
Benchmark outputs are always aggregates (median, percentile distribution) — never individual scores. Even in peer groups with the minimum N=5, a single peer score is never surfaced. See our Privacy Policy for the full data handling specification.
How Peer Groups Are Defined
Two attributes define your peer group: your industry vertical and your size band. The combination of both must meet the N≥5 threshold before a benchmark is shown.
Industry Vertical Assignment
When completing the assessment, firms select their primary industry vertical from a defined taxonomy: Accounting & Tax, Legal Services, Consulting, Architecture & Engineering, Marketing & Creative Agencies, IT Services, Financial Advisory, HR & Staffing, and Other Professional Services. The vertical is the primary grouping dimension.
Size Band Assignment
Firms are grouped into four size bands by full-time employee count: Micro (1-14 FTE), Small (15-49 FTE), Mid-Market (50-199 FTE), and Growth (200-500 FTE). AIOpsNav primarily serves firms in the Small and Mid-Market bands. If headcount is not provided, approximate annual revenue band is used as a proxy. [VERIFIED]
Minimum N=5 Rule
A benchmark score is only shown when at least five peers share your industry vertical and size band. When N<5, the benchmark card displays "Insufficient peer data — benchmark available once N≥5 firms in your peer group have completed the assessment." This prevents any statistical inference about a single peer from a tiny group. [VERIFIED]
Fallback Expansion
If N<5 in the primary peer group (vertical + size band), the system expands to the industry vertical alone across all size bands to provide a directional comparison. Expanded benchmarks are labeled clearly: "Vertical-only benchmark (all sizes) — N=[X]." Firms see the primary benchmark once N≥5 is met. [VERIFIED]
Freshness Badges and Data Refresh
Stale benchmarks mislead. Every benchmark display surfaces three trust signals: sample size, last-updated month, and data source type.
Monthly refresh schedule: Benchmark pools are recalculated on the first of each month, incorporating all new assessments submitted during the prior month. The last-updated date on every benchmark display reflects the most recent recalculation date. [VERIFIED]
Benchmarks are always backward-looking. They reflect the state of your peer group as of the most recent recalculation — not real-time. If your peer group is small or fast-moving, consider the freshness date as part of your interpretation.
Why Benchmarks Get More Valuable Over Time
The benchmark system improves with every assessment submitted. This is the compounding property of a shared data commons.
More submissions = better benchmarks for everyone
When you complete the AIOpsNav assessment, your anonymized score improves the benchmark pool for every firm in your peer group. A peer group that grows from N=5 to N=50 produces a benchmark that is statistically more robust, covers more of the score distribution, and can surface percentile breakdowns rather than just medians.
Early participants in a peer group see aggregate tier data. As the group grows past N=20, the system unlocks quartile benchmarks. Past N=50, it can show dimension-level peer comparisons — not just composite scores. The value of each assessment is multiplied across the entire peer cohort.
This is why AIOpsNav makes the assessment free. A larger dataset benefits every participant through more precise peer comparisons. [VERIFIED]
What We Promise About Your Data
Four commitments, each architectural — not just policy.
No firm identification
Your firm name and contact details are never included in benchmark computation. Scores are decoupled from identity at the data layer.
Aggregates only
Benchmark outputs are medians, percentile ranges, and tier distributions — never individual scores. A single peer's score cannot be reverse-engineered from any output.
N≥5 minimum enforced
The N≥5 rule is enforced in the query layer, not just the display layer. Groups below the threshold produce no benchmark output, only a threshold message.
Full policy published
The complete data handling specification — including retention, deletion, and third-party access — is in our Privacy Policy.