The Problem with Traditional Lead Scoring
Your rep opens a new lead in the CRM. Score: 82. She clicks into it and finds a company name, an email, and a note that says "downloaded whitepaper." That's the entire context. She picks up the phone with nothing to say beyond the generic pitch.
Most B2B lead scoring works this way. Points get assigned for demographic fit (job title matches ICP, company size is right) and digital engagement (visited the pricing page, opened three emails, attended a webinar). Add them up and you get a number. The number tells your rep to call, but it doesn't tell them what to say.
SiriusDecisions research found that 67% of B2B leads that score as "qualified" never convert. The scoring model said they were ready. The conversation proved otherwise. The gap between a scored lead and a real buyer is where pipeline goes to die.
What TruSQL Scores Differently
TruSQL™ is Rover Insights' proprietary scoring system. It rates individual leads 0-100, but the inputs are different from anything else on the market: real phone conversations with HR and finance professionals, not web clicks or form fills.
Two halves make up every score. Call lead quality contributes up to 60 points: the stated buying timeline, the person's role in the buying decision, company-size fit against your profile, and the substance of the conversation itself. Digital engagement contributes up to 40, drawn from what that same person reads, attends and clicks over time. Bands are 0-49, 50-69, and 70 and above.
Both halves are visible per record. Your rep doesn't see a single number and wonder what it means. A lead at 76 is 46 from call lead quality plus 30 from digital engagement, so they see exactly which factors pushed the score up or down, plus a written explanation of the reasoning.
One distinction is worth drawing before going further, because these two get conflated constantly. TruSQL is a score, not a lead type. It measures how well a lead fits your ICP and where they sit in the buying journey for your vertical, which means the same lead carries a different TruSQL score for you than for your competitor. The lead type sits on a separate axis: a Context Qualified Lead (CQL) describes how completely Rover knows the person, and that does not change from vendor to vendor.
How the Two Halves Work
Call Lead Quality (up to 60 points)
This is where TruSQL diverges from every other scoring system. The larger half of the score isn't inferred from page visits or content downloads. It's captured directly from the conversation.
Before any scoring happens, you configure your ICP in the platform. Target job titles, management levels, company size ranges, industries, and product needs (Payroll, HRMS, LMS, ATS, PEO, EXP). Company-size fit and the person's role in the buying decision are scored against that profile. A VP of HR at a 500-person manufacturing company looking for payroll software scores differently than an HR coordinator at a 10-person startup.
During each 6-12 minute call, CDR representatives capture structured buying signals: Is the prospect in-market, lifecycle, or not currently buying? What's their timeline: immediately, within 6 months, 6-12 months, or 12-24 months? Do they have budget approved? Have they expressed interest in a demo? What's their current solution, and how would they rate their satisfaction?
A prospect who says "We're evaluating payroll vendors with a Q2 deadline and budget approved" scores very differently from someone who says "We signed a 3-year contract last month and we're happy with it." The substance of the conversation counts too: a prospect who spent 11 minutes explaining their compliance challenges and asking about integration options is a stronger signal than someone who gave one-word answers and ended the call at 4 minutes.
Digital Engagement (up to 40 points)
The smaller half tracks what the same person reads, attends and clicks across Rover's community: articles opened, whitepapers downloaded, webinars registered for and attended, emails engaged with. It accumulates over months rather than being captured in one session.
Keeping it on its own axis is deliberate, and the reason is measured. Across 9,046 people carrying both scores, the correlation between the call half and the digital half is 0.048. Knowing how much somebody clicked tells you close to nothing about what they'll say when you ask, which is exactly why the conversation carries the larger weight and why both layers exist.
What Makes the Score Explainable
Black-box scoring is the norm in B2B. Your lead scored 85. Why? "The model determined high intent." That's not useful when your rep needs to prepare for a call in 10 minutes.
Every TruSQL score includes two things other systems don't: a Lead Score Description (a written explanation of why this lead scored the way it did) and AI-generated Recommended Next Steps (a prioritized action list based on the score and conversation context).
A rep looking at a TruSQL score of 84 might see: "Strong ICP match (VP HR, 800-employee financial services company). Stated Q3 buying timeline with budget approved. Currently using [Competitor] with satisfaction rating of 2/5. Expressed interest in a demo. Call sentiment: Strong Positive (10-minute call with detailed feature discussion). Recommended: Schedule demo within 48 hours focusing on compliance automation and multi-state payroll."
Compare that to: "Lead score: 84. Engagement: High." The explainability isn't a nice-to-have. It's the difference between a rep who walks into a call prepared and one who's guessing.
Score Ranges and What They Mean
TruSQL uses three score bands with color coding in the platform.
- 70-100 (Green, Strong Fit): High ICP match, clear buying signals, positive call sentiment. These leads carry the strongest combination of fit, intent, and engagement. Prioritize immediate follow-up.
- 50-69 (Gray, Moderate Fit): Partial ICP match or mixed buying signals. The prospect may be in a lifecycle stage (approaching contract renewal) or early in the research process. Nurture with targeted content.
- 0-49 (Red, Weak Fit): Low ICP match, no immediate buying signals, or negative call sentiment. These leads may become relevant later. The score explanation tells you why the fit is weak today.
The thresholds aren't arbitrary. They're calibrated against actual conversion data. Leads scoring 70+ don't just feel more qualified. They close at measurably higher rates.
How TruSQL Differs from Predictive Lead Scoring
Predictive scoring platforms (6sense, Lattice, EverString) use machine learning to identify patterns in historical data. They answer: "Which accounts look like your past customers?" The models are trained on closed deals, web behavior, and third-party data.
TruSQL answers a different question: "What did this specific person tell us about their needs, timeline, and buying process?" It's not predictive. It's verified. The data comes from a conversation that already happened, not from a model predicting what might happen.
Both approaches have value. Predictive scoring covers more accounts with less effort. TruSQL provides deeper context on fewer leads. The right choice depends on whether your team needs volume (many accounts with probability scores) or depth (fewer leads with conversation-level intelligence).
The 50+ Data Points Behind Each Score
Every TruSQL-scored lead arrives with the full conversation record. Pain points extracted and prioritized (High/Medium/Low). Feature needs ranked by importance. Current solution details: vendor name, satisfaction rating, contract start and end dates. Decision-maker identity: name, title, email, and their role in the buying process. Buying committee members. Budget availability. Demo interest level.
All captured in a single 6-12 minute phone conversation by a trained CDR representative. Not scraped from the web. Not inferred from content downloads. Stated by the prospect in their own words, then structured and scored by the platform.
Rover Insights delivers these scored leads to your CRM within 48 hours. Your rep opens the lead and sees everything: the score, the breakdown, the rationale, the next steps, and the full conversation context. That's what conversation-based scoring makes possible.