Related Ideas
eyko Ideas
A Lead Scoring Optimization Playbook evaluates how accurately your existing model predicts conversion, identifies which signals actually drive opportunity creation, and recommends weight changes that close the gap.
The Challenge
Most lead scoring models are configured at implementation and rarely revisited. Behaviour that predicted intent two years ago may not predict it now. New signals emerge that the original model never captured. The model keeps running, the scores keep flowing, and the predictive accuracy quietly erodes.
When AEs reject MQLs that the model rated highly, it usually means the model is overweighting a behaviour that no longer correlates with intent. Most teams treat the rejection as a sales-marketing alignment problem when the root cause is a model that needs recalibration.
Some behaviours are dramatically more predictive than others. A pricing page visit is materially more intent-laden than an eBook download. Most models weight these similarly because the original setup spread points across many behaviours rather than concentrating on the few that actually predict outcomes.
How eyko Solves It
A Lead Scoring Optimization Playbook reads historical lead and opportunity outcomes alongside the behavioural events that fed each score. It evaluates current model accuracy, identifies which signals are over- or under-weighted, and recommends weight changes that align the model with actual conversion patterns.
The current lead scoring model predicts conversion at 34% accuracy, below the 50% threshold for a healthy model. A pricing-page visit is 4.2 times more predictive of conversion than an eBook download, but both currently receive the same weight. 28% of MQLs are rejected by sales within 48 hours. A recalibrated model projects 52% accuracy and an 18% reduction in sales rejection rate.
| Metric | Current | Benchmark | Status |
|---|---|---|---|
| Primary indicator | Flagged | Target | Action needed |
| Secondary indicator | Monitoring | Within range | On track |
| Trend direction | Declining | Stable | Review required |
Related Ideas
FAQ
Everything you need to know about Signal Predictiveness (Top vs Bottom).
Lead Scoring Optimization is an AI-driven analysis that evaluates how accurately your existing lead scoring model predicts opportunity creation and revenue. The Playbook identifies which behavioural signals are over- or under-weighted, recommends weight changes that align the model with actual conversion data, and projects the resulting accuracy and sales rejection rate.
The Playbook reads from your marketing automation (lead activity, score history, behavioural events), CRM (opportunity creation, conversion outcomes, sales rejection events), and any third-party intent or firmographic data feeding the score. The richer the per-lead activity history, the more precisely the Playbook can attribute conversion to specific signals.
The Playbook compares predicted conversion (from the current model) to actual conversion outcomes across a 12-month lookback window. Accuracy is measured at the threshold the team uses to define MQL: how many of the leads that crossed the MQL line actually converted, and how many leads that converted were missed by the model. Together these define precision and recall, summarised as a single accuracy score against a 50% healthy threshold.
Yes. The Playbook runs a feature-importance analysis across historical conversions to produce a per-behaviour predictiveness score. Behaviours with high predictiveness receive recommended weight increases. Behaviours with low or no predictiveness receive weight reductions. The output is a side-by-side comparison of the current model and the recommended model, with the projected accuracy lift attached.
Join the enterprises replacing weeks of manual analysis with a single prompt. See what eyko Playbooks can do with your data.
