Writing & Analysis
Essays on what I'm building and learning
Long-form thinking on private markets, healthcare venture, corp-dev, and building without inherited networks. Newsletter: What the Data Room Doesn't Say, on Substack.
Subscribe →Start here
For students & founders
New to my work? Read the featured essay, then work with me if you're a leader, founder, or student facing a decision.
After years inside corporate development watching deals get priced on gut and precedent, I built the thesis for Value Alpha: institutional valuation infrastructure for the 99% of M&A that happens below $50M, starting in the US and now Poland. This is the full argument, why the tooling is broken, what fixing it requires, and why it's a defining infrastructure opportunity in finance and AI.
Healthcare startups don't die from small markets. They die on a Tuesday.
Why I back the operation, not the market size, and why complexity is the best moat in healthcare venture.
In private markets, the house has already read everyone's cards.
Public markets reward analysis. Private markets reward access. I built two companies on closing that gap.
Access is a subsidy I never qualified for. It made me a better pricer of risk.
A warm network lowers the cost of being wrong. Someone always pays for that, and it is usually whoever is outside the circle.
I built in American finance without a single warm intro. It became my edge.
No rolodex, no benefit of the doubt. Why that became the most useful analytical tool I have.
What I gave a nonprofit, and what it gave back.
Four months building an investment reserve policy for the Lower Manhattan Cultural Council, and a lesson in what business school is for.
The cofounder is the most expensive line on your cap table.
Choosing a cofounder and building an AI-era team, read like a valuation: roles, equity, decision rights, and the failure modes.
The AI-Native Valuation Benchmark
How AI-native companies are actually priced when the comps break: the AI-ready premium, the AI-exposure discount, and the comps gap.
How to value a pre-revenue startup when there are no comps.
A pre-revenue valuation is a negotiated number, not a calculation. How to triangulate a defensible range and defend it.
What corporate development really pays for when it buys your company.
The buy-side view of getting acquired, from the seat across the table. What strategics pay for, and why structure beats the headline.
How to know you have product-market fit.
PMF is not a vibe. How to measure it with retention and pull, the false signals, and what the value cliff is really worth.
How to handle a down round.
The headline is the least important number. How anti-dilution, structure, and the alternatives decide what a down round really costs.
How to read a cap table.
Ownership percentage is the least useful number on the page. How fully diluted shares, the option pool, and the waterfall decide what you keep.
Raising on SAFEs: the dilution you signed and forgot.
Each SAFE looks small. The stack is not. How post-money SAFEs convert, why the dilution compounds silently, and when to price the round.
Your 409A is not your valuation.
A company has two prices at all times. What a 409A actually does, why common is worth less than preferred, and which number answers which question.
How VCs actually decide whether to fund you.
You are pitching a machine, not a person. The funnel, the portfolio math, and the ownership target behind every pitch meeting, with real numbers.
How much should you actually raise?
The ask is a valuation decision in disguise. Milestone-back sizing, the dilution budget, and the bar the money sets.
How I price a company in 48 hours.
A valuation is an opinion you can defend. The full process, hour by hour: intake, normalized numbers, a size-first comp set, triangulation, and a range.
The five term sheet clauses that move money.
The valuation is the price of the round. The clauses are the price of the exit. Five clauses ranked by dollars moved, with worked math and the counter for each.
The option pool: the dilution that is only yours.
Every other line of dilution is shared. A pre-money pool is carved out of you alone. What the shuffle costs in dollars, and the five moves that cut it.
The strategic buyer premium is 5%, not 50%.
Every seller is told to hold out for a strategic. Across 185 transactions the gap is half a turn, and in the middle market right now it runs backwards.
Ecommerce clears 2.42x SDE across 4,584 deals.
The deepest cell in the index is also its cheapest. Why the multiples quoted around online businesses disagree, and the one conversion that reconciles them.
Micro-SaaS sells for 2.86x SDE, not 4x ARR.
Small software is quoted in revenue multiples and sold on earnings. What 523 micro-SaaS transactions cleared at, and how to convert between the two.
What a letter of intent actually locks in.
Exclusivity and confidentiality bind on signature. The price does not. What diligence does to the number, and how to price before you sign.
From October 1, the SBA stops taking the seller's word for it.
A quality of earnings report is now mandatory above $3M, reconciling four sets of books into one earnings figure. Why the rule exists, in one company's books.
SBA rewrote the acquisition rules six days before they took effect.
A prohibition reversed, a $350,000 valuation carve-out added, and a third file posted with no notice. The five changes, and how to verify each one yourself.
The problem I took to every class is now somebody else’s capstone.
Three years ago I carried one question into every class at Columbia. This term a Columbia capstone team is working on the same question, and I am on the other side of the table.
SBA lending shrank 28 percent, and the loans that vanished were the small ones.
Fiscal 2026 approvals fell 28 percent by count and 13 percent by dollars. The gap between those two numbers tells you which deals stopped getting financed.
Businesses sell for 94 percent of asking price. The ones that matter never sold.
The most quoted number in small business M&A can only count deals that closed. What the missing listings do to it, and why nobody is paid to publish the other number.
The SBA $3 million line is not measured on the price you agreed.
Both SBA thresholds run off Business Purchase Price, which excludes owner-occupied real estate at appraised value. What that changes, and the three routes out of the report.
AI revenue keeps 50 to 60 cents on the dollar, and buyers price it that way.
Bessemer puts AI gross margins at 50 to 60 percent against 80 to 90 for SaaS. At the same multiple of gross profit, a buyer pays about a third less.
What a search fund actually pays for a company.
Stanford puts the median search fund purchase at 6.2x EBITDA. That is the going rate for the size band, and every extra turn is paid in equity.
Why size sets the valuation multiple.
Crossing roughly $5M of EBITDA roughly doubles the multiple, 2.9x to 6.4x. Why the jump happens, and what selling a year early costs.
What is SDE, and which add-backs actually survive?
About a third of a typical SDE figure is not profit, it is argument. Which add-backs a buyer pays for, which get struck, and what each line is worth.
How an independent sponsor prices a deal.
They commit to a price before the capital is committed, so the number gets underwritten twice. What that changes, and what half a turn costs the promote.
Is a business worth 3x profit?
Every published source puts the multiple near 2.7x SDE. Across 7,601 transactions the middle half ran 2.0x to 3.4x, and the rule of thumb hides the whole negotiation.
Quality of earnings: the report that moves the price.
A QoE is not an audit. It is a price-adjustment document, and the adjustment gets multiplied. What it checks, what it costs, and who pays.
Nobody opens the product.
On an online or software business the product is the asset, so what it is worth and whether it is real are the same question. Almost nobody answers both.
Convertible notes vs SAFEs: the differences that still matter.
SAFEs took 93 percent of pre-seed. When the round goes well the two produce the same cap table, so every difference that survives shows up on the bad days.
Due diligence: what actually gets checked.
Not a test you pass. A search for reasons to pay less. The four files an investor opens, the findings that recur, and what each one costs.
Why most founder equity splits quietly destroy value.
The first price your company sets, decided in an afternoon on backward-looking inputs. What the fast 50/50 costs, measured, and the four fixes that are free at formation.
What I learned about M&A pricing after running 1,000+ valuations
I came to American M&A through the back door. A thousand valuations later, here is what the data taught me about how private companies are really priced.
I left billion-dollar deals to price ten-million-dollar businesses
What a $26 billion merger taught me about the 99% of M&A that runs on guesswork, and why I left to price the businesses no one builds tooling for.
on Substack Subscribe →