पाठशाला Pathshala · वृद्धि Vṛddhi, Growth · Lesson 04 · Start

Founder-led sales: the first hundred calls

The first salesperson a company hires should be a founder. A list of a hundred, a script with one question, a CRM with five stages and a weekly target, and the three numbers that fall out by call fifty.

Pathshala, The Founder Library · 11 October 2026 · 8 min read

Two engineers in Hyderabad have a working product and eight pilot users who have never been asked to pay. Their plan is to hire a salesperson. The salesperson will arrive, be given no list, no script, no price and no idea who buys, will fail inside a quarter, and will be replaced by another. The company will conclude that sales is hard. Sales is not the hard part. Not having done it is.

This lesson is the process a founder runs for the first hundred sales conversations, built from three sources that agree with each other: Tyler Bosmeny’s Y Combinator talk How to Sell, Steli Efti of Close in his Stripe Atlas AMA, and Pete Kazanjy’s Founding Sales, the handbook whose first half is titled, accurately, going from zero to about thirty customers yourself. It adds the Indian specifics: who decides, how to find them, what closing actually involves here, and the terms worth insisting on.

Why the founder sells first

Bosmeny’s argument, from building Clever’s sales from nothing, is that founders should sell before any rep is hired because they know the pitch, carry the conviction, understand the industry and can absorb the pivots that the first hundred conversations will force; and because buyers, especially early ones, prefer dealing with the person whose name is on the company. Efti adds the reassurance technical founders need to hear: sales is results-driven communication, and anyone who has persuaded a co-founder to quit a job has sales experience. Kazanjy’s point is the practical one. The first sales hire can only be trained in a process that exists, and the only way a process comes to exist is for a founder to run it until it repeats.

There is an Indian reason too. In a small or mid-sized Indian business the buyer is the owner, and owners buy from owners. A twenty-four-year-old with a quota does not get the second meeting in a Ludhiana factory. The founder does, because the owner across the table recognises someone who also signs the cheques. That advantage is free and it is wasted when the founder delegates the phone in month three.

The list before the script

Write the buyer in one sentence before anything else, in the form the [customer interview](/library/the-customer-interview-done-properly) lesson insists on: the company type, the size band, the city or cities, and the person who signs. “Owner-run garment exporters in Tiruppur with fifty to three hundred workers; the owner or the owner’s son decides.” That sentence is a hypothesis and will be rewritten at call twenty-five. It still has to be written now, because without it the list is anyone with a phone.

Then the list: a hundred named prospects who fit the sentence, each with a number or a profile. In India the sources are richer than founders expect. Industry associations publish member directories. Trade fairs publish exhibitor lists. GST registration data, searchable by category, tells you which businesses in a pin code exist and are large enough to register. LinkedIn, with more than a hundred million Indian members by Microsoft’s April 2023 count, carries the decision-maker in any company above fifty people. For owner-run businesses the number is on the shop board, on the IndiaMART listing and on the invoice. Build the hundred in a weekend. Then the week begins.

The script, and the data on what works

A script is ninety seconds and three parts. Who you are, in one line. Why you are calling this person, which has to be true of them specifically: their category, their city, something a peer told you, a job they are hiring for. Then one question about what they do today about the problem, asked and followed by silence. Gong’s analysis of 90,380 cold calls found a baseline of about 1.5 per cent of cold calls producing a booked meeting; stating the reason for the call raised the odds 2.1 times, successful calls ran nearly twice as long as unsuccessful ones, and in the successful ones the prospect’s longest uninterrupted stretch of talking was about three and a half seconds against eight in the failures, which means the caller was asking short questions and letting the buyer answer them. Bosmeny’s version is the ratio: the best sales calls are about seventy per cent listening and thirty per cent talking.

The question that does the most work is the interview question: what do you do about this today, and what does it cost you. The answer is the real competitor, usually a person or a spreadsheet or nothing, and it is the price anchor, because a product that saves a ₹25,000-a-month clerk half their time has a price. Do not pitch features until the buyer has named the problem in their own words; then pitch the one feature that answers it and stop. End every call by asking for a next step with a date on it. A demo on Thursday. A paid trial from the first of the month. A purchase order by Friday. “Send me a deck” is a polite no; answer it by asking what in the deck would change their decision.

The CRM and the weekly target

A CRM is a sheet with discipline, and the discipline matters more than the software. Five stages, defined in writing so that two founders would put the same call in the same column: contacted (reached a human), conversation (fifteen minutes with the decision-maker about the problem), qualified (fits the sentence, has the problem, has budget authority, has a reason to move this quarter), proposal (a price and terms in writing), closed (won, lost, or the most common and least recorded outcome, no decision). One row per prospect. The date of the last contact and the date of the next, always filled in.

Then a weekly number. Twenty conversations a week, meaning twenty fifteen-minute exchanges with decision-makers, gets to a hundred in five weeks and is achievable alongside building the product if the hours are blocked. It is also a lot of dialling. Efti’s minimums for cold calls are that around fifteen per cent should be answered and around fifteen per cent of those should lead to something, so a hundred conversations may sit on top of several hundred dials and messages, which is why the next lesson, on [outreach that gets replies](/library/cold-outreach-that-gets-replies-in-india), exists. Count the conversations, not the dials, but track the dials, because the ratio between them is the first number you will want to improve.

A founder who has not made a hundred sales calls does not have a sales problem. They have a sales process they have not yet run.

Closing, in India

Bosmeny’s closing advice is three refusals. Do not quibble over minor redlines in a contract; sign, start, and fix the relationship by delivering. Do not build a custom feature to win one customer, and if you must, make them pay for it under a conditional agreement. Do not give free trials, because a free trial produces no commitment and no information; prefer an annual contract with a thirty- or ninety-day opt-out, which gets the customer using the product with a real stake and gets you a real answer. Efti is blunter on pilots: charge for them, and do not count the money as revenue until the customer has stayed six months.

The Indian additions are procedural and they decide whether a verbal yes becomes cash. A purchase order or a signed quotation, because in most Indian companies the finance team pays against a PO and nothing else. A GST invoice with the buyer’s GSTIN correct, because an invoice they cannot claim input credit on is an invoice they will sit on. An advance, as large a share as the buyer will accept, on anything that involves setup or onboarding, because the gap between an Indian SMB agreeing to pay and paying is where early-stage companies run out of cash; the [runway lesson](/library/runway-how-many-months-you-really-have) is specific that receivables are not cash. And a named person on the buyer’s side responsible for the rollout, because a deal the owner signed and nobody inside owns will be churned at renewal with the words “we never really used it”.

What a hundred calls tells you

By call twenty-five the buyer sentence is wrong in a specific way and should be rewritten using the people who actually took the call and actually leaned in. By fifty there are three conversion rates: conversation to qualified, qualified to proposal, proposal to closed, each rough and each real, and they are the input to the [funnel arithmetic](/library/sales-funnel-and-conversion-rates-to-expect) that says how many calls the next hundred customers will need. By fifty the price should have been raised once, because a price nobody has flinched at has not been tested; Efti’s healthy distribution is a fifth finding it expensive, a fifth finding it cheap and the rest weighing it. By a hundred there is a list of every objection heard, with the answer that worked, and that list is the product roadmap and the first salesperson’s training manual in one document.

Then, and only then, the hiring decision. Kazanjy’s test is whether the founder can teach the process; if it cannot be written down as a playbook a stranger could follow, it is not ready to be handed to one. Efti’s rule for the hire itself is to take junior reps in groups of two or three so that performance can be compared and the process rather than the individual is what gets judged, and First Round’s GTM partner Meka Asonye, who built sales at Stripe and Mixpanel, makes the same points in his conversation with First Round Review: founders should be able to diagnose what went wrong in their own sales before hiring, and the first hire is commonly levelled wrong. A founder who hands over the phone at call one hundred with a playbook, three rates and an objection list has given the new hire a job. One who hands it over at call eight has given them a mystery.

The weekly ritual

Friday, thirty minutes, both founders. Read the sheet. Count conversations against the target of twenty and say the number out loud. Count asks: how many calls ended with a dated next step, because a conversation without an ask was a chat. Read every objection logged this week and change one line of the script in response, one, so that next week’s calls test the change. Move every row with no next date to lost or no decision; a pipeline full of “following up” is a pipeline of polite noes. Then write next week’s twenty names at the top of the sheet before the weekend. Five Fridays of this is a hundred calls. A hundred calls is a sales process. A sales process is what the salesperson you were about to hire needed, and now has.

The method is the sources’; the hundred calls are yours. Bosmeny’s talk and Efti’s AMA are both below and both short. Make the list this weekend.