From Investment Banking to Startups and Founding

Introduction

Every recruiting season, a slice of the analyst class decides the buy-side is not what they want and starts looking at startups instead. Almost all of them collapse two completely different decisions into one sentence. Joining a startup is a job search with an unusual pay structure and a solvable diligence problem. Founding one is an open-ended bet on yourself with no salary, no mandate, and no process to follow. The two share a vocabulary, a dress code, and nothing else that matters.

Treating them separately is the whole point of this article, because the honest advice runs in opposite directions. For joining, the answer is mostly mechanical: a small set of roles hire bankers, a specific band of company stages values what you know, and you can evaluate a private company's health from the outside better than most candidates in the applicant pool. For founding, the answer is uncomfortable: banking teaches you about a third of what the job needs, the base rates are worse than the mythology admits, and the analysts who eventually build something good almost always operate somewhere else first.

Money makes the decision harder than it should be. A second-year analyst is looking at a total package most of their college friends will not see for a decade, and the startup offer on the table is often half of it. That gap is real and permanent, and the equity that supposedly makes up for it usually does not, for reasons that are arithmetic rather than pessimism. This article walks through which roles actually hire bankers, what stage to target, how to price an option grant with real numbers, how to diligence a company before you sign, what founding genuinely requires, and how to protect your career if any of it goes wrong. If you are still mapping the full landscape, the broader survey of investment banking exit opportunities puts startups next to the paths that hire far more volume.

Before the detail, the three options side by side on the dimensions that actually drive the decision. Read the columns as three different jobs rather than three rungs on one ladder.

The row people misread is reversibility. Joining a startup at a recognizable company and doing real work is close to career-neutral: you can go back to banking, move to corporate development, or land at another startup with a better title. Founding is not neutral. Two years of a company nobody has heard of reads as a gap to some employers and as evidence of nerve to others, and which one you get depends almost entirely on how you talk about it, covered later.

Startups do not hire "ex-bankers" as a category. They hire for a function and then discover that a banker fits some functions well and others not at all. The realistic list is short, and knowing it saves you months of applying into roles where your resume reads as irrelevant.

Strategic Finance and FP&A

This is the highest-volume landing spot by a wide margin. Strategic finance at a venture-backed company means owning the operating model, building the board deck, running headcount and scenario planning, and being the person who can answer "what happens to runway if we miss the quarter by 15%" without a two-day fire drill. It is the closest thing in the startup world to the analyst job, minus the pitch books and plus actual decision rights.

The honest distinction between strategic finance and traditional FP&A is that FP&A is variance-driven and backward-looking (budget versus actuals, department reporting, close support) while strategic finance is decision-driven and forward-looking. Many companies use the titles interchangeably, so read the job description rather than the header. A banker is a strong fit for either, but the strategic finance version uses more of what you were trained on.

Corporate Development and Chief of Staff

Larger startups, generally those past $100 million in revenue or with an acquisitive strategy, build small corporate development teams that look almost exactly like a corporate M&A group. This is the most direct translation available: sourcing targets, running diligence, building the model, managing bankers and lawyers, and integrating. If this is your target, the mechanics of the move are covered in depth in the guide to going from investment banking to corporate development, and the same skills apply whether the employer is a Fortune 500 or a Series D company.

Chief of staff is the role most misunderstood by candidates. It is not a promotion into leadership and it is not an executive assistant job. It is a general-athlete role attached to a specific executive, usually the CEO or COO, where you take on whatever the company cannot currently staff: a pricing overhaul this quarter, a fundraise the next, an org redesign after that. Bankers do well in it because the job rewards fast synthesis, comfort with senior people, and the willingness to produce a defensible answer under a deadline. It is also the role with the least defined career ladder, which is either its best or worst feature depending on your tolerance for that.

Business Operations and Capital Markets at Fintechs

Business operations is the third common landing spot, particularly at marketplaces and consumer companies. The work is analytical rather than financial: sizing a new market, deciding which cities to launch, running an experiment, building the metrics layer that everyone else uses. It suits bankers who liked the problem-solving and disliked the accounting, and it converts well into general management later.

The most underrated option is capital markets at a fintech. Lending platforms, buy-now-pay-later companies, embedded finance providers, and specialty finance startups all need people who can negotiate warehouse facilities, structure forward-flow agreements with credit funds, manage rating agency relationships, and build securitization models. That is leveraged finance and structured products work performed inside an operating company, and a banker from a debt capital markets, financial sponsors, or FIG group is not merely qualified but genuinely scarce. Compensation in these seats runs at the top of the startup range because the alternative hire is expensive.

Startup hiring managers still open with technicals before they get to the interesting part: work through valuation, accounting, and behavioral questions in one place, start practicing interview questions for free and find the weak spots while the stakes are low.

Choosing the stage is a bigger decision than choosing the company, and it is the one bankers get wrong most often. The instinct is to go early because early sounds like more upside. The arithmetic and the job description both argue against it.

Why Seed-Stage Is Usually the Wrong Target

A seed-stage company has raised a median of roughly $2 million and employs maybe eight people, none of whom is doing anything a banker was trained to do. There is no operating model to own because there is no operating history. There is no board deck worth building because the board is two people who text each other. Nobody needs a scenario analysis; they need someone to write code, sell to the first ten customers, or answer support tickets.

That is the real problem with joining at seed. Your equity percentage is larger, but the work you would be hired for does not exist yet, so you end up doing generalist tasks a smart 24-year-old from any background could do, at a company with roughly a one-in-three chance of raising a Series A. You are taking maximum risk for the version of the job where your training is worth the least.

There is a second problem: seed-stage companies rarely have the cash to pay you. The trade is not "less cash for more equity" so much as "much less cash for a lottery ticket in a company that has not yet demonstrated anything." If that is the trade you want, founding is a better version of it, because at least you own a founder's stake rather than an employee's.

Runway

The number of months a company can keep operating before it runs out of cash, calculated as current cash balance divided by net monthly burn. Runway is the single most important number at a private company because it sets the deadline for everything else: a business with nine months of runway is already fundraising whether or not it has told employees. Investors generally want a company to raise when it has twelve to eighteen months left, since a round takes three to six months to close and negotiating leverage collapses below six months.

Series B to Series D Is Where Banking Skills Price Best

The band where a banker is genuinely valuable starts at Series B. By then the company has a real revenue base, a finance function that needs building rather than inventing, a board that expects proper reporting, and enough complexity that scenario planning changes decisions. It also has enough cash to pay a competitive salary and a real equity grant that has not been diluted into irrelevance.

Series C and D companies are further along the same curve: more structure, more cash, smaller equity grants, lower risk of zero. Late-stage and pre-IPO companies start to look like corporate jobs with liquid-ish equity, which is a fine outcome but not the reason most people leave banking. The practical rule is to pick the earliest stage at which the job you want actually exists, and for a banker that is almost never earlier than Series B. If the fundraising vocabulary is still fuzzy, the differences between growth equity, private equity, and venture capital explain who is writing the checks at each stage and what they expect in return.

How Title and Scope Translate

Title inflation runs in both directions and confuses everyone. A third-year banking analyst is usually hired as a Senior Analyst or Manager in strategic finance, occasionally as Director at a company under 100 people that is using titles as compensation. An associate with two or three years of post-analyst experience lands at Manager or Senior Manager, sometimes Head of Strategic Finance at a smaller company.

What matters far more than the title is scope, and there are three questions that reveal it. Who owns the operating model today, and will that be you? Who presents to the board, and will you be in the room? How many people report to you, now and in twelve months? A Director title with no model ownership and no board exposure is worth less than a Manager title with both, and the second one compounds into a CFO track while the first does not.

This is where bankers, of all people, stop doing math. The offer arrives with a percentage attached, the percentage sounds meaningful, and the candidate mentally converts it into the company's last valuation. Almost every step of that conversion is wrong.

What You Are Actually Being Granted

An option grant is the right to buy a fixed number of shares at a fixed price. It is not a share of the company today, it is not worth the headline valuation, and it produces nothing at all unless the company is eventually sold or goes public at a price above what the preferred investors are owed first.

Strike Price

The fixed per-share price at which a stock option can be exercised, set at the fair market value of the common stock on the grant date and determined by an independent 409A valuation. The strike price is almost always well below the price investors pay for preferred shares, because common stock lacks the liquidation preferences and protective provisions attached to preferred. An option only has value if the company's common shares are eventually worth more than the strike price, and the gain is the difference between the exit price per share and the strike.

The second mechanism is vesting, which decides how much of the grant you keep if you leave.

Vesting Cliff

A minimum service period that must be completed before any portion of an equity grant becomes yours. The standard startup structure is four-year vesting with a one-year cliff: you receive nothing if you leave before twelve months, 25% of the grant vests on your first anniversary, and the rest vests monthly or quarterly over the remaining three years. The cliff exists to protect the company from granting equity to short-tenure hires, and it means an eleven-month stint at a startup produces exactly zero equity value.

A Worked Example: What the Grant Is Really Worth

Numbers make this concrete. Assume a Series B company with 40 million fully diluted shares that just raised at a $200 million post-money valuation, which prices preferred shares at $5.00 each. You are hired into strategic finance with a grant of 0.15% of the company, or 60,000 options, at a strike price of $2.00 set by the most recent 409A valuation. Vesting is four years with a one-year cliff. The company has raised $60 million to date and will raise more.

Now run three outcomes. In the good case, the company grows well and is acquired in year five for $600 million. By then, two more rounds have taken the fully diluted count to 60 million shares, and the preferred investors convert to common because their pro-rata share exceeds what their preferences would pay them. Common is worth $10.00 per share. Your 60,000 shares produce $600,000 gross, less $120,000 to exercise at the $2.00 strike, for a pre-tax gain of $480,000. After combined federal and state tax at roughly 35%, call it $312,000 net, five years after you started.

In the middling case, the company is acquired for $250 million after a hard slog. Total liquidation preferences across all rounds stand at $150 million, and because the preferred investors' pro-rata share of $250 million is less than $150 million, they take the preference instead of converting. The preferred hold roughly 33 million of the 60 million shares, so their pro-rata claim is about $137 million, below the $150 million stack. That leaves $100 million for common and options spread across roughly 27 million shares, or $3.70 per share. Your gain is 60,000 multiplied by the difference between $3.70 and your $2.00 strike, which is $102,000 pre-tax. The company is acquired in year six after a hard slog: six years of work, one exit, roughly $66,000 after tax.

In the disappointing case, which is the most common of the three, the company raises a $50 million flat round, never finds a second gear, and sells for $120 million in year four. The preference stack, now $110 million, leaves a residual of $10 million across the common and option pool, well under your $2.00 strike. Common receives nothing. Your options are worth exactly zero, and you would never have exercised them.

Where the Value Leaks Out

Four leaks explain most of the difference between what a grant looks like on paper and what it pays. Dilution shrinks your percentage every time the company raises, and a grant expressed as a percentage without a share count and a total share number is not a real number at all. Liquidation preferences sit ahead of you and, in a mediocre outcome, absorb the entire purchase price, which is why understanding how cap tables and liquidation preferences work matters more than any valuation you argue about.

The third leak is the post-termination exercise window, the classic 90 days most companies still use. If you leave after two and a half years with vested options and a $60,000 exercise bill for stock you cannot sell, in a company that may fail, many people rationally let the options expire. The fourth is time: the median venture-backed exit takes the better part of a decade, and your grant is fully vested in four years, so the last several years are spent waiting on a decision someone else makes.

Here is the genuine advantage bankers have over every other candidate: you can evaluate a private company as an analyst rather than as a hopeful applicant. Most candidates assess a startup on the office, the product, and how much they liked the founder. You can assess it on capital structure, burn, and the distance to the next financing, and you can do most of it before the first interview.

Funding History and Investor Quality

Start with the public record. Press coverage, the company's own announcements, and state-level filings give you round sizes, dates, and lead investors. Three patterns matter. First, time between rounds: eighteen to twenty-four months is healthy, and four years of silence after a Series B means either the company is profitable, which they will happily tell you, or it could not raise. Second, who led each round: a new lead investor at each stage is a market validating the company, while an inside round led entirely by existing investors is often a bridge dressed up as a milestone. Third, round size relative to stage: an unusually small round suggests a compromise between what the company wanted and what the market would give.

Context matters too, and current context is extreme. The PitchBook-NVCA Venture Monitor for the second quarter of 2026 recorded roughly $413 billion of US venture deal value in the first half of the year, more than any prior full year on record, with the overwhelming majority of those dollars flowing into artificial intelligence companies and into rounds of $100 million or larger. Headline strength and broad health are not the same thing. A company outside the favored categories is raising into a market that looks generous in aggregate and is anything but at its own stage.

Burn, Runway, and the Distance to the Next Round

Ask directly. A well-run company will answer, and refusal is itself a data point. The three questions are net monthly burn, current cash balance, and the metric the next round depends on. Divide the second by the first and you have runway; compare it to the third and you know whether the company is on schedule or already behind.

The follow-up question separates good diligence from polite conversation: what has to be true in twelve months for the next round to be raisable at a higher price? A CFO or founder who can answer that in specific numbers, such as a revenue level, a retention threshold, or a gross margin target, is running the business the way an investor would. One who answers with narrative is telling you the plan does not survive contact with a spreadsheet. Because early companies are usually unprofitable by design, being fluent in how to value a company with no profits lets you judge whether the growth story is actually worth what the last round paid for it.

Cap Table Structure

The last layer is structure, and it is where a banker's training pays off most directly. What you are looking for is whether the common stock, which is what you are being paid in, has a realistic path to value.

A company that raised $60 million at reasonable terms and is worth $250 million has healthy common stock. A company that raised the same $60 million with a 2x senior preference and a ratchet from a hard round has common stock that is closer to a warrant with a very high exercise price, no matter what the last headline valuation said.

Switch decisions now. If the plan is to start something rather than join something, the first useful exercise is separating what your training actually gave you from what you assume it did.

Three things transfer cleanly. Financial discipline is the most underrated: you know what a unit economic is, you will not confuse bookings with revenue or revenue with cash, you can build a model that ties, and you will notice a burn problem two quarters before a first-time founder does. A striking number of startups fail on cash management rather than product, and you are structurally unlikely to be one of them.

Deal process transfers next. Fundraising is a sell-side process: you are running a competitive auction for equity in an asset with an information asymmetry problem. You know how to build a target list, run parallel conversations, create timeline pressure, prepare a data room, and survive diligence. Most technical founders learn this the expensive way. So does negotiation: term sheets, employment agreements, commercial contracts, and eventually an acquisition offer are all documents you have read hundreds of.

Investor communication is the third. You have spent years translating messy reality into a board-ready narrative for senior people with no patience. Writing a monthly investor update that is honest, quantitative, and short is a skill most founders take three years to develop, and it materially affects whether investors support you in a hard quarter.

The gaps are larger than the transfers, and each one is a common cause of death.

Product and customers is the biggest. Nothing in banking teaches you to identify a problem worth solving, talk to fifty potential users without leading them, or decide what to cut. Banking teaches you to answer a question a client already framed. Founding requires framing it, and the framing is most of the value.

Distribution is a close second. Bankers win clients through institutional relationships and inbound mandates. Startups acquire customers through channels you build yourself, and a very good product with no distribution routinely loses to a mediocre product with great distribution.

Ambiguity tolerance and building without a mandate are the psychological gaps. Banking is a high-workload, low-ambiguity job: the hours are brutal but someone can always tell you the deliverable and the deadline. Founding is the reverse. Nobody assigns the work, nobody checks it, the feedback loop is months long, and the right answer is often unknowable in advance. Analysts who thrived on the structure of banking, and many did without realizing structure was what they liked, find this disorienting.

The data is not encouraging and should not be hidden. Bureau of Labor Statistics business employment dynamics data on establishment survival shows roughly half of new US employer establishments still operating five years after opening, with the sharpest drop in the first year, when roughly one in five closes. That is the figure for all businesses, and venture-backed startups face a harsher version of it because "still operating" is a much lower bar than "produced a return." A venture fund's portfolio is built on the assumption that most companies return nothing.

Layer on the concentration of capital described earlier and the picture sharpens. When most venture dollars flow to a small number of large rounds in a favored sector, a company outside that sector needs to be substantially better to raise at all. Founding into that market with no operating experience, no distribution advantage, and no technical co-founder is not brave. It is a low-probability bet made on incomplete information, which is exactly the kind of decision your training should make you skeptical of.

None of this argues that nobody should found a company. It argues for honesty about the odds, and for structuring the attempt so that failure costs you a year rather than a decade. That usually means building the idea while employed, finding a co-founder with the skills you lack, and setting an explicit decision point where you either commit or stop.

The most useful finding for anyone weighing this comes from the research on founder age. The National Bureau of Economic Research study of US Census data by Azoulay, Jones, Kim, and Miranda on age and high-growth entrepreneurship found the mean founder age among the fastest-growing one in a thousand new ventures is 45.0 years, and that prior experience in the specific industry is a strong predictor of success. The twenty-three-year-old founder is a media archetype, not a base rate.

The mechanism is not age itself but what age proxies for: domain knowledge, a network of people who will take a call, credibility with early customers, and an accumulated sense of which problems are real. An analyst two years out of college has a strong financial toolkit and almost none of the rest.

The practical sequence, then, is to join a startup first in one of the roles above, spend two to four years watching how a company is actually built, and start something from inside that domain with a network, a thesis, and evidence. You will also learn what you are good at, which is the thing the direct jump denies you. The people who make this work do not skip the operating step; they compress it.

Founder conversations still turn technical the moment money is involved: download our comprehensive 160-page PDF, access the IB Interview Guide covering the valuation, accounting, and modeling frameworks that follow you out of banking and into every board room after it.

Three practical issues decide more outcomes than strategy does, and none of them appear in the recruiting conversation.

How to Explain the Move If It Fails

Assume it might, and prepare the story now. The reframe that works treats the move as a deliberate decision with a thesis and a result, not as an adventure that ended. Name why you joined or founded, what you owned, what you learned, and what happened, in that order and without defensiveness. Concrete ownership beats narrative: "I built the operating model and ran the Series C data room" is a sentence any employer understands, while "I helped grow the business" is not.

Two audiences respond differently. Corporate development teams, growth funds, and other startups usually treat startup experience as a plus, because it is directly relevant work. Traditional buy-side recruiting is less forgiving, which is one more reason to keep any startup detour inside a domain you could argue for later. If your longer-term destination is an investing seat, the paths described in the guide to moving from investment banking to venture capital tend to value operating experience more than buyout recruiting does.

The Visa Problem for International Candidates

For non-US candidates on an F-1 or H-1B, this is often the constraint that settles the decision, and it deserves specialist legal advice rather than internet summaries. The structural problem is that H-1B sponsorship requires an employer with the administrative capacity and willingness to sponsor, plus a lottery outcome you do not control. Large banks sponsor as a matter of routine and have dedicated immigration counsel. A 60-person Series B company frequently has neither, and a transfer of sponsorship is not automatic.

Founding is harder still. The H-1B category presumes an employer-employee relationship, which is difficult to establish when you control the company, and the cost and litigation environment around the program has been unsettled through 2025 and 2026. The alternatives founders actually use, such as the O-1A for individuals with extraordinary ability or the international entrepreneur parole pathway, have their own high evidentiary bars and timelines.

Some candidates should stay where they are, and saying so plainly is more useful than encouragement.

Do not do this if the primary motivation is escaping banking hours. Startup hours at the stages worth joining are lighter than banking but not light, and founding hours are worse than banking with none of the compensation. Leaving a job you dislike is a reason to change jobs, not a reason to change asset class.

Do not do this if you need the income. Family obligations, meaningful debt, or a visa that depends on continuous employment all make the volatility genuinely dangerous rather than merely uncomfortable. The right move is usually to stay two more years, build a cash buffer, and reassess from a position where the downside is survivable.

Do not do this if the equity is the point. If your model of the outcome requires the good case to hit and the good case is the reason you are going, you are underwriting a low-probability outcome as a base case, and the worked example above shows what the base case actually pays.

And do not found a company because you cannot decide what else to do. Founding is the highest-cost way to postpone a career decision. If you cannot name the specific problem, the specific customer, and why you in particular should be the one solving it, the honest answer is that you have a desire to be a founder rather than a company to build.

The best version of this move is unglamorous. You pick a domain you find genuinely interesting, join a company at a stage where your training is worth something, negotiate a grant you have actually priced, and spend three years learning how a business is built rather than how a transaction is executed. If founding is the eventual goal, that is also the fastest legitimate route to it, because you arrive with a network, a thesis, and a realistic sense of what you can do. And if it does not work out, you have a specific, describable set of accomplishments and a market that will hire you, which is more than most people who leave banking on a feeling can say.