Use your own rolling 90-day median as the primary benchmark. External ranges are context, not a target, and should never override unit economics or lead quality.
Use this resource when
Decision matrix
| Metric | Diagnostic band | Interpretation |
|---|---|---|
| Lead → SQL | Below 10% | Usually a targeting, offer, form or follow-up problem |
| Lead → SQL | 10–25% | Workable for many broad B2B programs; segment by source |
| Lead → SQL | Above 25% | Strong qualification or narrow demand; check volume and definition |
| SQL → Opportunity | Below 20% | Review qualification consistency and sales acceptance |
| Pipeline ÷ spend | Below 3× | Often too little gross pipeline for long-cycle risk |
| Pipeline ÷ spend | 3–8× | Planning band only; validate win rate and margin |
Read the ranges as diagnostic bands
These are practitioner planning bands, not a universal market dataset. They are designed to tell you where to investigate. Geography, average deal value, sales cycle, form definition and channel mix can move every number materially.
Use the benchmark hierarchy
Start with cost per opportunity and pipeline per dollar. Then diagnose cost per SQL, lead-to-SQL and CPL. CPM and CTR explain delivery; they do not decide commercial success.
Build your internal benchmark
Use at least 90 days, segment by channel and offer, and calculate median rather than relying only on a blended average. Record the number of leads behind every rate so small samples do not look precise.
- Same stage definitions
- Same attribution policy
- Median plus 25th/75th percentile
- Lead and opportunity counts shown beside rates
Know when the benchmark misleads
A high lead-to-SQL rate can come from overly narrow targeting that cannot scale. A low CPL can come from a weak gated asset. A strong pipeline multiple can still fail if win rate or gross margin is low.
Practical questions
Questions that change the decision
Are these industry averages?
No. They are transparent diagnostic bands from operating experience, explicitly not represented as a universal industry study.
What is the best benchmark for a new account?
Start with unit economics and a conservative funnel model. Replace external assumptions with actual cohorts as soon as data accumulates.
How often should targets change?
Review quarterly or after a major offer, market, tracking or sales-process change—not after one volatile week.