Methodology
Benchmark methodology
This page is the standing methodology for Replede's benchmark publications on contact-form conversion and speed-to-lead. We are publishing the rules before the numbers on purpose: our first benchmark report is forthcoming, and until it ships, Replede makes no benchmark claims of its own.
What we will measure
- Intake conversion rate — visitors who start the intake vs. those who complete a buyer profile.
- Completion depth — how many qualification fields (problem, role, budget range, timeline) a completed intake actually fills.
- Speed-to-lead — elapsed time from intake completion to the profile landing in the CRM and the first rep alert.
- Profile completeness vs. static-form baseline — fields present per lead compared with what the tenant's previous form captured, where the tenant can share that baseline.
Data sources
- Aggregated, anonymized tenant data — production intake metrics from Replede tenants who have explicitly agreed to inclusion. No tenant is identifiable in published figures; no profile contents are ever published.
- Third-party studies — where we reference external research, it is always cited, linked, and labeled as external; we will not blend external figures into our own metrics.
- Never synthetic data presented as observed — modeled or illustrative numbers, where used, are labeled as such.
Sample handling
- Minimum sample before we publish a statistic: at least 10 tenants and at least 1,000 completed intakes for any headline figure; segment cuts (by industry, traffic band) only where the segment independently clears the threshold.
- Medians over means — headline figures are medians with interquartile ranges, so a single outlier tenant cannot move the result.
- Fixed windows, no cherry-picking — each report covers a declared calendar quarter, stated up front; we will not select flattering date ranges.
- Spam and test traffic excluded symmetrically — the same exclusion rules (spam, internal tests, incomplete intakes) apply to every metric, favorable or not, using the same definitions as billing.
Update cadence and versioning
Benchmark reports will be refreshed quarterly. Every report is dated, methodology-versioned, and kept available after newer editions ship — we will not silently rewrite past numbers. Changes to this methodology will be listed on this page with the date of the change.
Corrections
If you believe a published figure is wrong, email hello@replede.com. Verified corrections are made in place with a visible correction note, not quietly. The same policy covers reported third-party pricing on the AI SDR pricing comparison.
Current status
No benchmark report has been published yet. The contact form ROI calculator and the speed-to-lead calculator are estimators driven by your own inputs — they are not benchmarks and make no claims about observed Replede performance. When the first report ships, it will be linked from this page.