Lead Scoring, end to end
Lead scoring decides which lead your rep calls next — and getting it right is the difference between a busy team and a closing team. This is the complete map of how Leadkaun grades every lead A–F in real time, the concepts behind it, and the guides to set it up.
Quick answer
What is lead scoring and how does it work?
Lead scoring ranks each lead by how likely it is to convert. Leadkaun grades every lead A–F in real time across three independent scores — Fit (ICP match), Intent (live engagement, which decays as leads go silent), and Quality (data reliability) — with transparent weights, so reps work the highest-probability leads first.
- Three independent scores — Fit, Intent, Quality — combine into one A–F grade.
- Intent decays automatically when a lead goes quiet, so stale Grade A leads fall on their own.
- Transparent weights beat a black-box AI score — reps trust a grade they can see.
- A graded lead is in the queue within the first hour of setup.
What lead scoring actually is
Lead scoring is the discipline of ranking every incoming lead by how likely it is to convert, so a rep works the best-odds opportunity first instead of whatever landed most recently. That's the whole job. It is not a vanity metric or a report your manager reads on Fridays — it is the answer to the one question a rep asks forty times a day: out of everyone sitting in my list right now, who do I call next?
A useful score has to hold three separate ideas apart, because they move at different speeds. Is this the kind of customer we sell to? Are they showing buying behaviour right now? And is the record even usable — a real phone number, a real inquiry, a real company? Collapse those into one number and you lose the ability to explain any grade. Keep them separate and a lead becomes legible: you can see at a glance whether it's weak because it's a poor fit, because it's gone cold, or because half the fields are blank.
That is the model Leadkaun uses — three independent 0–100 scores, Fit, Intent and Quality, resolved into one grade from A to F. The grade drives the action: A means call now, F means junk, and everything between has a defined next step. A score nobody acts on is just decoration.
Why it matters more for Indian B2B
The case for scoring is not academic; it is a leak you are almost certainly paying for. Harvard Business Review's audit of over 2,000 firms found the average company took about 42 hours to respond to a web lead, and 23% never responded at all — even though contacting a lead within the first hour made it roughly 7× more likely to qualify. Separately, InsideSales research reported in Forbes found only about 27% of inbound web leads are ever contacted by a rep. Most demand is never worked. The teams that win are simply the ones that reach the right lead first, and keep reaching them.
In India that gap is wider because the volume is higher and messier. Leads pour in from IndiaMART, JustDial, TradeIndia, Sulekha, website forms and walk-ins — often the same buyer across three of them in one afternoon. An edtech counsellor or real-estate broker can be staring at 150 fresh enquiries by lunch with no honest way to tell the ₹40L-ready buyer from the student comparing every institute in the city. Without scoring, the loudest or newest lead gets the call, and the best one goes cold in the pile.
There's also a channel reality no benchmark can ignore: WhatsApp is where Indian buyers reply. A Kantar study cited by Meta found 91% of online adults in India message a business at least once a week. A lead that ignores your call but replies on WhatsApp within an hour is showing intent — and any scoring model built for India has to treat that reply as a first-class signal, not an afterthought.
The Fit × Intent × Quality model, and how good scoring works
Fit answers 'is this our kind of customer?' It is scored against your ICP — the industries, states, business types, buyer roles and budget bands that define who you sell to. In Leadkaun that's a 0–100 total: industry match is worth up to 30 points, geography 20, business type 20, decision-maker role 15 and budget signal 15. Fit is deliberately slow-moving — it changes only when new firmographic information arrives, not because a lead went quiet.
Intent answers 'are they hot right now?' It starts from a source baseline — a referral opens higher than a cold list — and then rises on real engagement: a WhatsApp reply, an 'asked about pricing' tag, an 'answered and interested' call, a negotiation. Crucially, intent decays. After a window tied to your sales cycle (four weeks by default), a silent lead loses 3 points a day, so a Grade A from a fortnight ago slides to a B on its own while genuinely fresh leads rise above it. Re-engagement spikes it straight back up. This is the single mechanic most homegrown scoring sheets miss — they never let a stale lead fall.
Quality answers 'is this record even usable?' — a valid Indian phone number, an email, a company name, a clear inquiry, a reliable source. It is a guard, not a booster: anything scoring under 20 is forced to Grade F and dropped from the queue, so junk and half-filled rows never eat a rep's morning. The three scores then resolve through a threshold matrix, not an averaged blend — a lead has to clear a real bar on each dimension to earn its grade, which is why a high-fit lead with dead data still can't sneak into the A pile.
Why transparent scoring beats a black-box AI score
It is tempting to hand scoring to an opaque AI model that emits a single number. The problem is trust. When a rep is told 'this lead is a 74' with no visible reason, they do one of two things: obey it blindly, or ignore it and call whoever they had a good chat with last week. Neither is scoring — one is superstition, the other is guesswork. A grade only changes behaviour if the person acting on it can see why it is what it is.
Transparent scoring means the rep can open a Grade B lead and read the story: strong fit, but intent has decayed since the last reply and the email is missing. That is coachable. A manager can audit the same matrix — no hidden per-account drift, no model that quietly re-weights itself. In Leadkaun the scoring weights are fixed and identical for every account precisely so the grade stays explainable. You configure your ICP — who your best customers are — not the underlying weights. That is the right dial to give a sales team: the definition of a good customer, not the mechanics of the maths.
Common mistakes to avoid
The first mistake is a static score that never decays. If a lead earns Grade A on Monday and holds it for three weeks of silence, your queue is lying to you — reps chase corpses while new demand waits. Time has to be a first-class input; silence should cost points.
The second is conflating fit and intent. A perfect-fit enterprise account that has shown zero interest is not the same as a scrappy small buyer messaging you every hour, yet a single blended number treats them as interchangeable. Keeping the two scores separate is what lets a rep decide whether to nurture patiently or call in the next ten minutes.
The third is scoring on rotten data. If half your leads import with no source, no budget and a mangled phone number, no model can save them — quality scoring has to quarantine junk before it pollutes the ranking. The fourth is scoring nobody acts on: a grade that doesn't translate into an ordered call list, a follow-up, and a visible consequence for ignoring it is just a number in a database.
How Leadkaun approaches it — alongside your CRM
Leadkaun is a Sales Behaviour OS: a lead-intelligence layer that sits on top of the system you already run and does the one thing most CRMs do weakly — tell each rep who to call next and why. It doesn't replace your CRM; it's the ranking-and-behaviour layer your CRM is missing. You import leads by CSV (Indian amount formats like ₹1,50,000, 2.5L and 1Cr parse cleanly), and every lead is graded A–F the moment it lands.
From there the grade drives the working screen. The Priority Queue re-ranks each rep's list top-down by a blended, intent-weighted score, so the hottest lead is always on top — not the newest. The Missed Opportunity Engine watches for stale leads past their grade's window (a Grade A gets 24 hours, a B gets 48) and surfaces the ₹ at risk in real rupees, so a sales head sees 'this much revenue is going cold' rather than 'you have 12 overdue tasks'. WhatsApp and call outcomes feed the intent score through manual 3-tap logging — the rep opens their own WhatsApp, has the conversation, and taps the outcome; there's no automated messaging, just honest signal capture.
None of this asks you to rip anything out. Your CRM stays the system of record; Leadkaun is the layer that decides the order of the day and keeps the grades honest without a manager babysitting them.
Getting started and what to measure
Start by defining your ICP honestly — the industries, states, business types, decision-maker roles and budget bands of the customers you actually close, plus your typical sales cycle, which sets how fast intent decays. This is the one input that makes fit scoring meaningful, and Leadkaun can suggest it back from your recent leads once enough have been analysed. Then import a batch and watch where your A and B grades come from.
The metrics that tell you scoring is working are behavioural, not cosmetic. Is first response landing on high-grade leads faster? Is the share of Grade A leads left un-contacted past their window shrinking? Are reps' calls concentrating at the top of the queue instead of scattering? Over time, watch which sources produce grades that actually convert — you may find a portal that sends volume but few winners, which is a renewal conversation you can now win with a number instead of a hunch. Good scoring doesn't just rank leads; it retrains the team to spend its scarce selling hours on the leads that pay.
Everything on lead scoring, in one place.
Questions on lead scoring.
Your reps open their queue tomorrow.
Setup the same day. 14-day free trial. No credit card required.