Does it run in the background while I work?

No. The analysis runs while you wait, on a sample of the file rather than every row. It is fast enough not to need a queue, and we would rather say that plainly than imply a background job that does not exist.

What happens if my file scores Low?

Nothing is blocked. A Low band tells you to expect a poor connect rate and to judge reps against that, which is more useful than discovering the same thing across three weeks of calls. The decision stays yours.

Does it change my data?

The analysis writes nothing. It reads a sample and reports. Normalisation and duplicate handling happen at import, as a separate step you approve.

Intake Intelligence

Know what is in the list
before you import it.

A bought list looks the same whether it is gold or garbage until a rep has burned a week on it. Intake reads a sample first, valid phones, duplicates, missing fields, B2B or not, and hands back one verdict: High, Medium, or Low.

Start free trial See lead scoring

leads-march.csv · 2,140 rows

sampled

M

Medium readiness

Usable, with gaps to plan around

Valid phones82%

Duplicates of existing11%

Missing company6%

Reads as B2BYes

Deterministic checks on a sample. Illustrative figures.

01The live product

Drop a file in and read it before it lands.

This is the actual import step. The check runs on a sample while you wait, and the verdict shows before a single row is committed.

Import Leads3

Lead Ingestion

Import from CSV, Google Sheets, or add manually. Indian phone normalisation and dedup built in.

1Set the batch details

Batch nameLead sourceInitial stageHow old is this list?

2Upload your file

Drag & drop your CSV here

or browse files · CSV, max 10 MB

Other ways:

Required: name, phone. Optional: email, company, designation, city, state, pincode, budget, interest level (High/Medium/Low), last contact days, notes. Column names are matched automatically, so “Mobile No”, “Full Name” and the like all work.

Regrade all leads

Re-run scoring on every lead with the latest grade thresholds

Import history

Batch	Date	Status	Added	Hot	Value	Dupes
Import · 02 Aug 2026 · Website form leadkaun-import.csv	02 Aug, 07:57 am	Complete	28	1	₹12.8Cr	2
Import · 19 Jun 2026 · Google Ads test-leads.csv	19 Jun, 06:41 pm	Complete	0	—	—	35
Import · 19 Jun 2026 · Google Ads test-leads.csv	19 Jun, 06:40 pm	Processing	0	—	—	10
Import · 14 Apr 2026 · Trade fair leads_v17.csv	14 Apr, 03:11 pm	Complete	8	—	₹25.1L	—

Quick answer

What does Intake Intelligence check before I import a lead list?

Four things, on a sample of the file: whether phones are valid Indian mobiles, how many rows duplicate leads you already have, which fields are missing, and whether the list is genuinely B2B. It is deterministic, no model, runs while you wait, writes nothing, and returns a single readiness band, High, Medium or Low.

02The checks

Four questions, answered before you commit.

The expensive moment is the hour after a bad import, once the rows are in and the calls have started. These checks happen before that.

Contactability

Valid phones, not populated ones

A filled phone column tells you nothing. Every number is checked against real Indian mobile formats before a rep dials it.

Duplicates

Rows you already have

Matched on normalised phone, so a lead someone is already working does not arrive twice under a different spelling.

Shape

B2B or B2C, honestly

Company names are read to tell whether this is a business list or a consumer one, whatever it was sold as.

Readiness

One verdict, not a score

High, Medium or Low. A band is a decision; a number is an argument.

03The verdict

A band is a decision. A number is an argument.

A readiness of “71.4” invites a debate about whether 71 is good. High, Medium or Low does not, it tells you whether to import, import with caution, or push back on whoever sold you the list. Nothing is ever blocked; the decision stays yours.

H

High

Clean phones, few duplicates, clearly B2B. Import and work it.

M

Medium

Usable, with gaps. Expect a lower connect rate and judge reps against it.

L

Low

Poor formats or a consumer list sold as B2B. Know it before three weeks of calls.

04How it works

No AI. On purpose.

Deterministic arithmetic

Phone shape, email validity, duplicate matching, field coverage, business-versus-consumer signals. Around eighty per cent of understanding a dataset needs no model at all, and a rule you can read is easier to trust than one you cannot.

Runs while you wait

On a sample, not every row, fast enough to skip a queue. We would rather say that plainly than imply a background job that does not exist. The analysis writes nothing; normalisation happens later, as a step you approve.

05In their words

A bad list costs three weeks before anyone admits it was bad.

Before

Forty portal enquiries a day, and the one we called first was whoever had messaged most recently.

“We were never short of leads. We were short of an order to work them in. The first morning the queue told us to call a site visit from nine days ago before any of the fresh ones, and that settled an argument we had been having for two years.”

Mandar Deshpande

Director of Sales, Vaastavik Realty

Pune · Real estate

06FAQ

Import questions.

07Works with

Intake is step zero. Here's what comes after.

Lead Scoring

Once a clean list is in, scoring grades every lead so reps know who to call first.

Priority Queue

Imported leads land straight into one ranked queue per rep.

Score Evolution

Same instinct as intake: show the working, so a number can be trusted.

Product mechanics, published in full at /methodology

Deterministic

Rules you can read, not a model

While you wait

No queue, no background job

Non-destructive

Reads a sample, writes nothing

One band

High / Medium / Low, a decision

08 · Ready when you are

Check the next list before it costs you a week.

Upload a file and see its readiness band in seconds, before a single row is imported or a single rep dials.

Start free trial See pricing

Free ₹0 · no card · same-day setup