Executive summary

Lead quality must be measurable, actionable and tied to operational workflows. Use a small, instrumented pilot to quantify contact performance, conversion funnel impact and noise factors before scaling. Turn raw lead attributes into a composite quality score, feed that score into CRM and dialer logic, and monitor daily for drift.

Core metrics to track (what to measure)

Capture these metrics at lead and campaign level. Record values per lead so you can aggregate by source, sub-source and time window.

• Delivery latency – time from lead generation to first contact attempt (median and 90th percentile).
• Contact rate – connects divided by unique leads dialed; report daily and cumulative.
• Right-party contact (RPC) rate – percentage of connects that reach the intended person rather than voicemail or third-party.
• Attempt distribution – average and median call attempts per lead plus percent of leads exhausted at each attempt threshold.
• Invalid/undeliverable rate – phones flagged by real-time validation or returned as invalid on first attempt.
• Duplicate rate – share of leads with matching contact keys (phone, email) within your suppression window.
• Time-to-conversion – median time from first contact to paid or qualified event; use percentiles to detect long tails.
• Lead-to-conversion – conversion fraction for the target action (registration, funded account, appointment) over a fixed window (7, 14, 30 days).
• Revenue per lead (RPL) – incremental revenue attributable to leads divided by leads delivered, when available.

How to score lead quality (practical model)

Convert raw metrics into a single, repeatable score for routing and pricing conversations. Use weighted sub-scores so the system is transparent and adjustable.

1. Define sub-scores: deliverability (0-30), contact efficiency (0-25), conversion potential (0-30), data hygiene (0-15).
2. Map metric bands to points. Example: Contact rate >25% = 25 points; 15-25% = 15 points; <15% = 5 points.
3. Normalize and sum sub-scores. A final score of 80+ signals high-priority routing; 50-79 qualifies for standard treatment; <50 triggers quarantine and manual review.
4. Adjust weights where business priorities differ (e.g., prioritize conversion potential over contact efficiency for VIP acquisition).

Pilot design and sample sizing

Run a controlled pilot per campaign segment and source cohort. Practical guidance:

• Minimum pilot size: 1,000-5,000 leads per source to get stable contact and invalid-rate estimates; use larger samples for low-base-rate conversions.
• Split traffic evenly between control and test routing to isolate supply behavior from dialer settings.
• Run pilots for at least 7-14 days to capture weekday/weekend variation and batch-timing effects.

Operational integration (CRM and dialer)

Embed quality signals into operational systems so agents and automation act on them:

• Write the composite quality score into CRM lead fields and use it for prioritization and SLA rules.
• Use score bands to set dialer pacing and retry logic: higher scores get faster routing and more aggressive attempts.
• Tag leads with invalid or duplicate flags to exclude them from expensive outreach paths.
• Instrument call dispositions to feed back into the scoring model and to close the loop on label quality indicators (wrong number, not interested, callback requested).

Monitoring, alerts and remediation

Automate checks and thresholds to spot supply degradation quickly.

• Daily dashboards: delivery latency, contact rate, invalid rate and conversion per source.
• Alert triggers: invalid-rate spike >5 percentage points day-over-day, median delivery latency >12 hours, duplicate rate >3%.
• Remediation steps: pause internal routing for recent deliveries, isolate recent batches for investigation, perform targeted data checks, and run a manual review on a random 100-lead sample to validate labels and dispositions.

Examples of actionable thresholds (rule-of-thumb)

Use these as starting points; tune to your vertical and product funnel.

• Invalid/undeliverable: <5% preferred, >10% unacceptable for scale.
• Contact rate: >20% on fresh outbound; <10% warrants investigation of data freshness or dialer timing.
• Duplicate rate: <2% target; >5% indicates poor source deduping.
• Median time-to-contact: <2 hours for live offer capture; longer times reduce contact efficiency dramatically.

Operational checklist before scaling

• Instrument per-lead tracking fields in CRM.
• Run a pilot that meets minimum sample size and timeframe guidance.
• Define composite scoring rules and routing logic.
• Configure daily dashboards and automated alerts.
• Establish an internal remediation workflow and escalation path for spikes and drift.

FAQ

How quickly should leads be delivered to maximize contact rate?

Shorter delivery latency improves contact outcomes. Aim for median delivery under 2 hours for most acquisition funnels; measure the impact by comparing contact rates by latency cohort.

What sample size shows a reliable conversion estimate?

Use at least 1,000 leads to estimate contact and invalid rates with reasonable stability; increase sample size for low conversion endpoints or when you need to detect small percentage differences between sources.

How often should scoring weights be recalibrated?

Recalibrate monthly during the first three months of a new campaign, then move to quarterly if performance stabilizes. Trigger an out-of-cycle review on any sustained metric drift.

Should high-scoring leads get preferential agent routing?

Yes. Route higher scores to more experienced agents and faster dialer queues. Monitor whether preferential routing improves downstream conversion enough to justify the resource allocation.

Which lead attributes are most predictive?

Predictive value varies by offer. Start with contact validity, delivery latency and prior conversion signals, then use a small machine-learning model or uplift tests to find additional predictors.