पाठशाला Pathshala · दल Dal, The team · Lesson 05 · Start

The solo founder: how to make it work

More than a third of new companies now start with one founder, and they raise a small fraction of the money. The four things a co-founder supplies, and the advisory, hiring and decision structures that supply them instead.

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

Paul Graham’s first mistake that kills startups is having a single founder. Carta’s data says more than a third of new companies now have exactly one. Both are right, and the gap between them is the subject of this lesson: not whether to go alone, which is often not a choice, but what to build around yourself so that going alone does not mean doing it alone.

It is written for the founder who has looked for a co-founder, run the [forty questions](/library/choosing-a-cofounder-forty-questions) on the candidates available, and found nobody worth giving half the company to. That is a legitimate outcome. The wrong co-founder is worse than none. What follows is the structure that makes none workable.

What the numbers say

Carta’s solo founders report, published in December 2025, tracks the share of new startups on its platform with a single founder rising from 23.7 per cent in 2019 to 36.3 per cent in the first half of 2025. The money has not followed at the same rate. In 2024 solo-led companies were 30 per cent of the startups founded that year and received 14.7 per cent of the cash raised in priced equity rounds. The earlier Founder Ownership Report makes the same point from the other side: solo founders were 35 per cent of the companies incorporated on Carta in 2024 and 17 per cent of those that also closed a venture round that year.

Read the two figures together. Going alone has become normal. Raising alone remains roughly half as likely. Whatever a founder builds around themselves has to work for the business first and for the pitch second, but it has to work for both.

The orthodoxy is softer than its reputation. Graham’s essay gives two reasons for the single-founder warning: it suggests the founder could not persuade anyone to join them, and the low points of a startup are so low that few could bear them alone. Y Combinator’s own FAQ says it regularly accepts solo founders, while repeating that one-person startups are tough and that a co-founder makes success more likely. Neither says do not. Both say: know what you are missing.

So be precise about what is missing, because each piece has a different replacement. A co-founder supplies a second skill, usually the one you lack: the engineer to your seller, the operator to your builder. A second pair of hands at the same level of commitment, so that two things happen at once. A check on your judgement, from someone with the standing to say no and the stake to mean it. And someone to carry the low points with, which Graham calls the emotional support and which is the one founders underestimate most until month nine.

A solo founder who tries to replace all four with one hire, or with a vague circle of mentors, replaces none. The structure below takes them one at a time.

The second skill: a founding hire, not a late co-founder

The instinct is to keep looking for a co-founder and offer the role to the first strong candidate who appears in month six. Resist it, for the reason the equity split lesson gives: the person joining at month six to a company with a product and a few customers has not taken the risk the founder took, and a split that pretends otherwise is unfair to the founder, while a split that reflects it leaves the newcomer feeling like an employee with a grand title. Wasserman’s research found that founders who gave away more equity built more valuable companies, but to people who earned it, over time, on a schedule.

The better instrument is a founding hire with a founder-sized grant. Index Ventures’ Rewarding Talent handbook puts a typical senior engineer at seed at around one per cent, and notes that grants of two to three per cent can make sense for solo founders who need a wide range of experienced skills early. That is the range for your first engineer or first operator: two to three per cent, four-year vesting with a one-year cliff, a title that says founding, and a seat at every decision in their area. It is a tenth of what a co-founder would cost and it is earned. The [first engineer lesson](/library/hiring-first-engineer-below-market) covers the hire itself, including how to make the offer when the cash is below market.

The check on judgement: a personal board of three

A co-founder’s most valuable function is to tell you that you are wrong before the market does. Replace it with three advisers, chosen for three different gaps, on a written arrangement with a fixed meeting cadence. Not a list of twelve names on a slide. Three people who have each agreed to a monthly hour and to answering a message within a day.

Pay them in equity, on terms that already exist so that nobody negotiates. The Founder Institute’s FAST agreement is the standard template: at pre-seed, 0.5 per cent for a standard adviser who meets monthly and 1 per cent for an expert adviser who also opens doors and takes on projects; at seed those fall to 0.25 and 0.75 per cent; at Series A to 0.1 and 0.5. Vesting over two years with a three-month cliff, so an adviser who disappears after two meetings has earned nothing. Three advisers at pre-seed therefore cost one and a half to three per cent of the company, against the forty or fifty a co-founder would have taken, and they can be changed when the gaps change.

Choose the three for the gaps the forty questions revealed in you. A founder who sells well and cannot build wants a technical adviser who has shipped at the stage you are at, not a chief technology officer from a large company. A founder who builds and cannot sell wants someone who has carried a quota in your market. The third is almost always someone who has raised in India in the last three years and will read your numbers once a month without flattery. In Bengaluru, Delhi NCR and Mumbai these people exist and are findable; outside those cities the third adviser is often remote, and that is fine.

A solo founder does not lack a co-founder. They lack four specific things a co-founder would have done, and each of the four can be bought, hired or scheduled.

The decision structure: write it down before you do it

Two founders argue, and the argument is the quality control. One founder has to manufacture the argument. The cheapest way is a decision memo: one page, written before any decision that commits more than a month of runway or cannot be reversed in a week. What is being decided, the options considered, the one chosen, why, what would have to be true for it to be wrong, and the date on which it will be reviewed. Then send it to one adviser before acting, with a deadline for objections.

The memo does three things a co-founder does. Writing it exposes the reasoning you would otherwise skip. Sending it creates the no that nobody else has standing to say. And the review date forces the look back that solo founders avoid, because there is nobody to ask how it went. Keep the memos in one folder. A founder with twenty of them has a record of their own judgement that no pitch deck can fake, and the honest ones become the best answer to an investor’s question about why you are alone.

Add a weekly review with one other person. Thirty minutes, the same day each week, with an adviser, the founding hire or a peer founder. Three numbers from the week, one decision taken, one thing not done. The content matters less than the appointment, which is the thing a co-founder provides without anyone scheduling it.

The low points: peers, not mentors

Graham’s second reason is the one with no commercial fix. A co-founder is in the same boat; an adviser is on the shore. The nearest substitute is other founders at the same stage, met regularly, with the explicit understanding that the meeting is for saying the things that cannot be said to staff, investors or family. Founder peer groups exist in every Indian metro, some through incubators and accelerator alumni networks, some informal; a group of four to six that meets fortnightly and keeps confidences is worth more than any number of networking events. If none exists where you are, start one with two people. It takes a month.

Treat your own health as a line item, because a solo founder has no redundancy. A co-founded company survives a founder being ill for a month. A solo-founded one may not, so the sleep, the exercise and the one day a week without the laptop are not indulgences; they are the maintenance schedule on the only critical component. The [founder track](/library) has its own lessons on this; the point here is only that a solo founder carries the risk alone and should price it.

Answering the investor’s question

Every solo founder is asked why. The weak answers are that you have not found the right person yet, or that you prefer to work alone. The strong answer is the structure in this lesson, already running, described in a sentence: a founding engineer on three per cent who has been here eight months, three advisers on FAST terms I meet monthly, a peer group I have sat in for a year, and a folder of decision memos you are welcome to read. An investor who hears that is no longer assessing whether you can bear being alone. They are assessing the company, which is what you want.

Then raise a little more than a team would, because Carta’s numbers say the rounds are harder to close and the founder’s own time is the scarcest resource in the company. The [runway lesson](/library/runway-how-many-months-you-really-have) explains why the six-month floor for starting a raise is already longer in India; a solo founder running the process alone should treat it as eight.

The structure, as a calendar

Month one: recruit the three advisers and sign the FAST agreements; start the decision-memo folder with the first one, which is the decision to go alone. Months one to four: hire the founding engineer or operator on a two to three per cent grant with vesting. From week one: the weekly thirty-minute review with one other person, same day every week. From month two: a founder peer group, fortnightly. Monthly: one hour with each adviser, with the month’s memos sent in advance. Quarterly: re-read the memos and write one paragraph on what the pattern says about your judgement. And every six months, honestly: is there now a person who should be a co-founder, and would the forty questions say yes? Being solo is a decision, reviewed, not a condition.

Nothing here is legal or investment advice. The adviser percentages are a published template, not a rule, and the equity for any hire or adviser belongs in a written agreement with vesting, drafted by someone who has done it under Indian law.