My posts, measured.

David Friedberg (the All-in guy) says the AI jobs panic is a crock. Daniel Priestley says human labor is trending toward worthless. They're both right. The gap between them is where the leverage gets made. Friedberg's point: nobody's cutting headcount with AI. They're building 100x more. The bottleneck stopped being cost and became how fast you can act on opportunity. Priestley's point: if you sell time, you're competing with software that does the same work cheaper and never sleeps. That work gets priced toward zero. Both true at once. Here's how they fit: The 100x leverage is real. It just doesn't get shared. It flows to whoever owns the system and skips right past whoever works inside one. Same technology. Opposite outcomes. One question decides your side: Are you building the systems, or running the tasks they replace? One caps out at the hours in your day. The other compounds while you sleep. The people who get this are building now, while it still feels optional. That window doesn't stay open. So I wrote the exact build. A field guide to making your own acquisition screen with Claude Code, from free public data, instead of renting a $25k database seat. Free, no email: https://lnkd.in/ejNVWmUE

I failed a history class my sophomore year of college. I thought that was it. That my life was over, that I'd never get a job, that I'd be sent packing back from whence I came. What I couldn't see then: the failing grade was never the improbable part. Me sitting in that classroom at all was. We couldn't afford a US education. My mother raised me alone, and we did ok by Zimbabwean standards, but a US tuition bill was not a number we had. I got into the one school that offered a 99% grant. She paid a few thousand a year for something that should have cost six figures. A few years later, a 3.1 GPA walked into investment banking. Into rooms built for 3.9s. Every gate I went through had a number on it that said no. The grade. The tuition. The GPA. I build scoring systems for a living now. I rank companies. I score targets. I decide who makes the list, and who gets outreach. And I never forget that a score is a snapshot, not a verdict. The one worth backing is often the one the model would have cut. I'd know. I was the one the model would have cut. So now the first thing I ask my models isn't who ranks highest. It's: is there a way this one could work? Is there an x factor here that makes the rest of the scorecard irrelevant? The best part of the story is usually the part the data can't see.

what kills proprietary deal flow — bad copy or bad infrastructure? infrastructure. it's not close. you can write like f*&king Hemingway and be the most credible name in the inbox. but if you can't trigger a notification on the other end, none of it lands. if you can't get in the door, there's nothing to show them. and targeting is part of that infrastructure. get the right person in front of you and a template reads like it was written for them — because in the way that counts, it was. the rest is the plumbing most people skip. private IPs, not the shared pools behind instantly and smartlead. emails varied enough — down to the byte size of the subject and body — that no two look the same. a handful per account a day, staged. and the channels that notify faster than cold email: a linkedin connection request, a DM, whatsapp, even physical mail. copy gets you taken seriously once you're in. infrastructure gets you in. full breakdown in the video.

I got this book for my son. But it’s having a bigger impact on me.

The ballsiest way to make a million dollars as an acquisition entrepreneur is probably underwater. Four words: off-bottom oyster farming. You'll need high testosterone for this one. Which is perfect, because oysters deliver more of the crucial building block for T than any food on earth: zinc. One serving = 300% of your daily dose. Let me pencil you the numbers. → An acre of off-bottom water grows 100,000 oysters a year → Unbranded, they clear $0.50 at the farm gate. With a name menus repeat: $1.00 → Cost to grow one: $0.30–0.35 + other ops costs $0.20-0.30 → = $0.35 of EBITDA per oyster The name costs nothing. It's your bay. So a branded acre = $100K revenue, $65K gross margin, $35K EBITDA. For $1M of EBITDA you need ~30 acres. Build cost runs $60K/acre. So: $1.8M in once (+ maintenance), $1M out. Every year. And this isn't some dying trade. Farmed oysters are a $327M US market (USDA Census of Aquaculture) that now out-earns wild harvest AND imports — output nearly doubled over the past decade, pulled up by raw bars going year-round. So why hasn't anyone built this? And why does it take big brass balls? Because an oyster takes 18–36 months to grow. Everything you sell in 2028, you finance today. Two years of inventory sitting in open water, through storms and pestilence, before a dollar comes back. That's why America has 900 oyster farms averaging $360K of revenue. The market isn't fragmented because it's a bad business. It's fragmented because of a working capital wall. And walls like that don't stop markets. They just decide the winner: whoever arrives with a balance sheet. Another problem. You can't just permit 30 new acres. In the premium states, new leases take years and die slow deaths in waterfront politics. Permitted acres are scarce. So the fast path isn't growing acres. It's buying them. Aging founders on 5-acre leases, mature inventory already in the water, no succession plan. You're not buying companies. You're buying water, permits, and a two-year head start someone else already paid for. And it gets better as it gets bigger: → One umbrella brand amortized over 30 acres instead of 3 → Mortality risk spread across bays — one bad closure no longer sinks the year → Seed bought at scale through the hatchery bottleneck → Year-round supply consistency, which is exactly what distributors pay up for Small farms compete. Scale compounds. The consolidation has already started — with nobody driving. Alabama, 2018–2022 (per a study in Aquaculture Economics & Management): production doubled, 22 farms became 10, same acreage. A rollup with no roller. So there you have it. A way to capitalize on every jock who now tracks his sleep score and measures nighttime erections like Bryan Johnson. A high-T business model. Through and through. PS: Oyster map doesn't exist yet. I'll build it if enough people ask. The Ohio trades map does exist though: screen.searchloop.ai

Blackstone just paid 18.5x for a bunch of plumbing and HVAC shops. If you're hunting trades deals, that number should worry you. Not because you're competing with Blackstone. You're not. It's worse than that. $2.5 billion for Champions Group in February. Goldman, $1.7B for Sila Services a few months before. The mega-funds aren't buying trades businesses anymore. They're buying the roll-ups. The platforms bought independent shops at 4-8x over the last five years. Blackstone just validated the exit at 18.5x. No wonder add-on activity in HVAC is up 88% year over year. PE and strategics now do roughly 80% of all HVAC deals. So I went looking for a real answer on how closed the window actually is. I screened 1,013 home services companies across Columbus and Cincinnati. Ownership, reviews, traffic, owner identification. Audit trail on every call. Two things the databases won't tell you: → The platforms already own the biggest footprints. The four PE-platform operators in my demo sample include the largest players in these metros. Columbus Worthington Air (6,824 reviews) is ARS, which is GI Partners. Service Experts is Brookfield. Erie Home is Gridiron. Bassett is Alpine. That's who you're actually bidding against. Not Blackstone. The platforms Blackstone just handed an exit valuation to. And they're working from better maps than you are. → But 19 of the top 25 operators are still independent. Family shops with 200 to 4,431 reviews, 30+ year histories, named owners. One is on its 4th generation. Several have succession signals hiding in plain sight. The window isn't closed. It's closing in a specific, mappable order. One more thing the screen caught: a PitchBook "PE-backed" tag that fell apart when we checked the actual evidence. Independent, owner named. That's the difference between a database and a screen. The demo is live and clickable. Link in comments. Want the full Ohio inventory? Comment "Trades" or DM me. Free. Whose maps are you working from? P.S. Different vertical? Same machinery points anywhere. DM me your market and I'll show you what it looks like aimed at your targets.

I used to feel like Adam Sandler in 50 First Dates. Every morning, re-introducing myself to my AI, along with a bunch of frustrated swear words, like the last 24 hours never happened. Now I don't. Because I, at long last, figured out how to use Obsidian. Obsidian is a free app that turns a folder of plain text files into a connected knowledge graph. Point Claude at the folder and it inherits every connection you've built. The problem is it's not intuitive. Most people never see it working. So I built a starter vault you can download and open in a minute — a fictional lower-middle-market deal with a thesis, a CIM, a founder, a banker, and a four-gate scoring model, all wired together. (You'll need Obsidian too — it's free.) Open the graph view and watch everything connect. Point Claude at the folder and it reads the whole thing. No setup. No "let me tell you about my work." Just context, instantly. But here's the part that actually matters — it's recursive: Every session starts by pointing Claude at the vault. Every session ends by asking it to write up what you did and link it back in. It already knows how — the rules live in the vault. So every session makes the next one smarter. The vault never forgets, and it keeps getting richer. You stop repeating yourself — for good. I built it for private equity pros. But it works for anyone who keeps starting from scratch — anyone whose AI should remember how you think and build on what you've already done. Your rules. Your systems. Your frameworks. Link's in the comments. No email gate, no sign-up. What do you find yourself re-explaining to your AI every single time?

Most acquirers pay $25k+ a year for a deal database. And it's worst at the exact companies they most want to buy — the small, founder-owned, off-grid ones that never make it into a firmographic feed. You don't need it. You can build a better, proprietary screen yourself — all you need is Claude Code. I've spent the last 9 months building exactly this for 15 private equity clients. So I wrote the whole playbook down and put it online, free: → Where the real data actually lives — Google Maps for trades, the NPI registry for healthcare, Companies House for the UK. All free. → The cheap tools that do the work — Apify, Firecrawl, Claude Code. → How to scrape, enrich and score targets without it hallucinating (the accuracy discipline is the whole game). → A downloadable starter kit — the rules, prompts and playbooks. Just point Claude Code at it. Two live examples on the page: a home-services screen for Ohio, a dental screen for Texas. Every company scored, private-equity ownership flagged, the owner identified — with a full audit trail. No black box. Cost? ~$50–300 in API calls. Versus $25,000 a year. And it surfaces the owner-run targets the databases miss completely. 2-minute walkthrough + the full guide 👇 (free - no sign-up) screen.searchloop.ai — P.S. Prefer it done for you — built, with the outreach run (usually 20%+ reply rates)? That's what I do for funds. DM me. #privateequity #growthequity #searchfunds #dealorigination #manda

35,000 companies scraped. Under 100 worth contacting. 20% reply rate. One touch. The full sequence hasn't even started. A UK fund running a platform bolt-on strategy came to me with a problem they knew was important but not yet urgent. Both partners were buried in portco integration — hiring, operations, the real work of running businesses. Origination sat on the back burner. Nobody was doing it. Everyone knew it mattered. There wasn’t enough bandwidth. So they asked me to help. We pulled 25,000+ companies from Google Maps across their target sectors. Then filtered. Hard. Not just size and location. The client needed us to understand the business model — several specific things that mattered for whether a bolt-on would actually integrate. Including business model and whether the revenue mix matched the platform. If the fit wasn't right, it was out. After that analysis: fewer than 100 companies cleared the bar. 25,000 in. Under 100 out. Then the real work started. We set up dedicated sending infrastructure: private IPs via emailBison, proper warm-up, domain reputation dialed in so nothing bounces. The copy was specific to each business. Non-needy. No fund size in the first paragraph. Just the reason this owner, this company, matters. We launched last week with the first touch. Reply rate: above 20% including 3 owners you just called back directly. Not open rate. Reply rate. Most PE cold outreach lands between 2% and 5%. We're at 4-10x that. And the full sequence is still coming — LinkedIn touches, WhatsApp, letters, all of it. This is just the opening move. Most funds think the problem is one thing. It isn't. It's the whole system. The filter finds the targets. The infrastructure ensures delivery. The copy gets the reply. You need all three. Most funds have none of them dialed in. The partner who was too busy to source? Calls on his calendar now. Get in touch if you want the full breakdown.

I've been on both sides of deal outreach. Sending it as a buyer. Receiving it from bankers trying to get my attention. The emails that get replies all share one thing. And it's not what most teams optimize for. The standout emails are short. Four or five sentences. No fund size. No last deal closed. No "checking in." They mention something specific about the business. A detail that tells the owner: I actually read your materials and I understand what you built. Then they say, roughly: "We're looking at a few opportunities in this space. If a conversation is useful, I'm around. If not, no sweat." That's it. No pitch. No urgency. No credential flex. Here's why this works. Most outreach telegraphs need. The subtext is: please take my call. The recipient feels pursued, not respected. The short email does the opposite. Its subtext is: I have a thesis and capital. If our paths overlap, great. If they don't, I won't waste your time. That posture is non-needy but also nice. It signals competence without demanding attention. Sellers have ego. They built something from nothing. They can tell the difference between someone who understands the business and someone who bought a mail merge tool. The first message sets the entire dynamic. Now — if you already own a business in the sector, this gets even easier. Your portco does the credentialing. The seller doesn't wonder if you're real. You've already put capital to work in their world. You understand the margins, the customers, the headaches. That shortens the distance between "who is this" and "let's talk" by half. But even without one, the principle holds. Be specific. Be short. Don't chase. Credibility isn't something you announce. The seller decides it by the end of the first paragraph. This is where most lean PE teams get the AI question backwards. The instinct is to automate the message. Feed the CRM into an LLM. Let it write. The output passes a grammar check but fails the credibility check. It sounds identical to every other buyer using the same tool. The better answer: AI runs the system. Human judgment reviews the message. AI can draft a smart template, enrich it with details about the owner and the business, manage the cadence. But someone who knows the sector reviews every message before it ships. If it sounds like a bot wrote it, you never get the reply. Automate qualification. Automate deliverability. Automate follow-up so nothing falls through the cracks. That's the part that burns time for teams with capital and a mandate but no analyst army. But the first touch needs human review. Someone who knows why this business, why this owner, and what a real conversation sounds like. Differentiation has to show up in the first message. Not the pitch deck. Not the second call. The email. Being easy to work with is the most underrated differentiator in M&A outreach. The seller decides whether you're easy to work with by the end of the first paragraph.

Marc Benioff had a funny line on All-In: "Sex bots off. Cursor on." lol, but the point: AI is moving toward actual work infrastructure. And a lot of PE firms I talk to are still stuck on step one. Using AI like a smarter intern. Draft the memo. Summarize the CIM. Clean up the deck. Write diligence questions. Polish the email. Useful. But still just a human sitting inside a chat window. The real leverage is much less glamorous: turning messy workflows into systems. At the fund level: A banker sends over a new opportunity. Who reads it? Who checks the CRM? Who looks for conflicts? Who decides whether it's worth a first call? Where does that decision get recorded? Is it a different process next time? At the portfolio company level: A lead comes in. An RFP lands. A quote goes out or invoice gets generated. Someone checks it. Someone follows up. Someone records the review. Someone updates the customer record. Someone knows the weird exception that never made it into the SOP. That's the operating layer. And at most companies, it lives in people's heads, inboxes, spreadsheets, half-used systems, and "Sarah usually handles that." AI doesn't just automate those workflows. It exposes whether the workflow actually exists. The next wave of AI value won't come from better prompts. It'll come from turning messy human operating knowledge into repeatable systems. That's the layer I care about at SearchLoop: turning messy origination and operating workflows into systems that compound. Happy to compare notes if you're thinking through where AI actually changes your operating model.

7 years at Credit Suisse, Greenhill, and Treis taught me one thing about PE that doesn't show up in the pitch deck: The firms that win the next decade aren't the ones with the best capital. They'll be the ones with the best origination engines. A year ago I set out to build that engine. 9 months in: → 12 paying clients — 6 running platform theses, 6 running roll-ups — across home services, dental, healthcare, HVAC, and B2B services & software → 700+ owner conversations → 2 proprietary databases shipping soon — one this Thursday Going solo against an industry of 100-person firms taught me something: you don't need to be bigger. You just need to be faster - and keep the human in the loop. If you're in PE or are an advisor building a thesis in a fragmented vertical — the two DBs dropping over the next two weeks are for you. This Thursday: every specialty trades operator in Ohio, scored on 7 acquirability signals. Next Thursday: every acquirable dental practice in Texas, ranked by succession risk. Want the DB early? Comment Trades or Teeth below — or DM me directly.

What used to take a 20-person PE shop can now be done with 2-3 people and AI. Fragmented industries — pool services, dental, specialty trades — have been a real roll-up grind for years because the operational complexity didn't scale. I think that could be chaning fast. Here's why.

Two searchers bought a home care business for $88M in 2020. It's now tracking toward an exit close to $1B. The differentiated mechanism: The business had two halves — a software company and a service company. When they entered a new state, they deployed the software sales team FIRST. Sold into the market. Got operators implementing. Then acquired one of those software customers as the platform. By LOI, they already knew: → the target's exact financials → their operational patterns → whether the team could actually execute → how fast integration would go (under 30 days, because the data was already in their system) That's not proprietary deal flow. That's manufactured deal flow. Most searchers and ISPs treat "proprietary" as "cold outreach to an unbrokered deal." That's just earlier. The real edge is being inside the operator's business for 2 years before you ever send an LOI. Software is the most obvious Trojan horse, but it's not the only one. Any service you sell into the sector works — audits, recruiting, implementation consulting. Anything that gives you visibility into the actual operations before you offer to buy them. The next $1B search outcomes aren't coming from a broker's teaser. They're coming from operators who sold into their sector first and acquired from inside their customer list. (Story is from Jenna Wigum on Acquiring Minds — the Abound Health build-up is worth the full listen.)

I don't know how to code. But last week I vibe-coded a full control center for a client's deal origination system. Launch searches. Initiate outreach. Check the funnel. Run enrichment. Verify ownership. One interface. It looks better than half the SaaS tools I've paid for. So if anyone can build that now — do SaaS investments even work anymore? Here's the thing. That control center only works because we spent a lot of time building what sits behind it. The workflows. The qualification logic. The CRM integrations. The enrichment chains. The error handling. The front end just gives you buttons for infrastructure that already exists. I couldn't have vibe-coded any of that. And that's exactly what PE is paying for when they roll up vertical SaaS. Nearly half of all SaaS M&A last year was vertical SaaS. Roofing ops. Cemetery management. Public safety dispatch. Nobody's acquiring these companies for the UI. They're acquiring the system of record — embedded workflows, compliance logic, integrations that took years to build. Vibe coding made the front end a commodity. The SaaS companies I'd worry about are the ones where the product IS the pretty interface. That's a weekend project now. The ones I'd buy? The ugly vertical platform with mission-critical guts. So did Claude Code kill vertical SaaS? No. But it killed the wrong reasons to invest in it. — If you're hunting vertical SaaS compounders before they hit the broker desks, I put together a free tactical guide with the origination stack we run at SearchLoop: https://lnkd.in/e6GBgwxR #PrivateEquity #B2BSaaS #VerticalSaaS #DealOrigination #SearchFunds

Med spas. HVAC. Landscaping. Lots of funds is chasing fragmented plays. But the weirdest market I've looked at? Dead people. Making money from dead people seems kinda unethical, but it's not really. You're providing a service. To their loved ones, I guess. It just hits different. Because we're talking about a business model that shares commonalities with our deepest fear. Or uncertainty. People die every day. Recurring revenue box ticked. And more people are going to die as the boomers age out. PE doesn't care. And the deals are being done: → Birch Hill Equity Partners took Park Lawn private for $1.2B in 2024 — 170+ funeral homes and cemeteries, now expanding into new states → Axar Capital took StoneMor private, rebranded it Everstory Partners, now runs 460+ locations with drone-mapped digital operations → Serent Capital backed PlotBox in 2025 to scale cloud-based cemetery and crematory management software → Rosewood Private Investments quietly rolled up 77 funeral homes across the northeast through Milestone → SCI spent $71M on acquisitions in just the first nine months of 2025 — sitting on a $16B preneed backlog Also, the smartest players aren't just consolidating — they're layering in software. Tech-enabled death care. That's a phrase I never expected to type. Often in PE a business is a business. Landscaping, vertical SaaS, healthcare — they all get a nod when mentioned in a convo. But could you get up every day and get excited about the fact that you're rolling up funeral homes? — If you're building out your own origination engine to find deals in sectors most funds overlook, I put together a free tactical guide with the systems we use at SearchLoop: https://lnkd.in/e6GBgwxR #PrivateEquity #GrowthEquity #SearchFunds #DealOrigination #PrivateMarkets

There are 19,000 private equity funds in the US. There are 14,000 McDonald's. KKR's Alisa Wood dropped that stat last year and it still lands the same punch in 2026. And it's not just a US story. Europe has roughly 9,000 PE funds and around 9,000 McDonald's — nearly one-for-one. The saturation is everywhere. More PE vehicles than fast-food outlets. That's not a fun fact — it's the competitive reality every fund is living inside right now. When there are this many funds chasing the same deals, the old channels get noisy fast. Broker networks, conference circuits, the same Grata and SourceScrub lists — everyone's fishing the same pond. The funds pulling ahead aren't working harder on the same playbook. They're building proprietary origination infrastructure that compounds — systematic market mapping, thesis-driven qualification, persistent outreach that actually lands. It's not about "using AI." It's about having a repeatable system of record you can point to. Disagree? PS: If you're thinking about building your own origination setup, I put together a free 10-chapter playbook with the exact stack and sequences we've used across funds: https://lnkd.in/e6GBgwxR

Search funds are quietly moving upmarket. For years the classic target was $1–3M EBITDA businesses — the smaller "searcher special" deals that bigger PE funds usually ignored. Now the new wave is chasing $5M+ platforms that used to be pure middle-market PE territory. Mineola Search Partners just published a good piece on the trend last week (https://lnkd.in/ebvDekUA), and EPA Investissements announced they're scaling up to back 40 new searchers this year (https://lnkd.in/efhduWEc). Same solo operator. Same 18–24 month clock. But now you're competing head-on with real PE money for the same targets. Origination just got harder — and a lot more valuable. You can't just pull the same Grata, Inven, or SourceScrub lists everyone else is using when bigger players are circling the exact same companies. The edge now belongs to the people building proprietary lists from raw sources — Google Maps, regulatory databases, piecing the puzzle together — then layering on outreach that actually lands in the inbox. I've laid out exactly how to build this infrastructure this 2026 PE Origination Playbook — going from raw data to a qualified calls. Grab it here: https://lnkd.in/e6GBgwxR Anyone else feeling this size creep?

6 million businesses are going to change hands in the next decade. Most won't find a buyer. McKinsey estimates that in 2022 alone, 92% of small business exits ended in closure — not sale. Not because the businesses lacked value. Because no buyer showed up in time. Half of all small business owners in the U.S. are already over 55. One in four are 65+. Regional manufacturers, specialty contractors, local distributors built over 30 years — with real cash flow and no obvious successor. This is the largest wave of quality deal flow any of us will see in our careers. And most PE and search fund teams know it. They're running Instantly sequences, scraping Clay lists, working their broker networks. But it's all disconnected. A campaign here, a list there. No systematic way to map a market, qualify at scale, and build persistent outreach that compounds over time. So they end up reactive — catching owners after they've already decided to sell, in a process, talking to three other funds. The edge goes to whoever reaches these owners while they're still running the place — before they've even decided to sell. That's not a relationship strategy. That's an infrastructure problem. The $5 trillion transfer is coming regardless. What's your current system for finding owners before they raise their hand?

The 2026 PE reports from Bain and McKinsey are out. Headlines look great. Details don’t. McKinsey says global PE deal value +19% to $2.6T. Bain shows buyouts +44%, exits +47%. Peel it back and the recovery is narrow, concentrated, and brutal. Deal count actually fell. A handful of megadeals did the heavy lifting. There’s still a $3.8T backlog of unsold companies. Holding periods pushing seven years. Distributions stuck below 15% NAV for four straight years. LPs aren’t being picky — the math is forcing them. Bain puts it plainly: “Multiple expansion and ultra-cheap debt have waned… The firms that stand out in this difficult new environment will be those that can find or sharpen a repeatable model for sourcing deals, determining early how to create value and executing at speed. Relying on the same old, same old has never been riskier.” Most funds are still running on networks, broker deals and micky-mouse proprietary outbound. Few have a system they can actually point LPs to. The real edge belongs to the teams running a full repeatable system: comprehensive market mapping, thesis-driven qualification, persistent multi-channel outreach that gets A/B tested — all CRM-connected and compounding every week. That’s the infrastructure we build at SearchLoop. This isn’t “we use AI.” This is the repeatable model Bain is talking about. The rebound is here. It’s going to reward the builders. https://lnkd.in/e_w6BCfX

The most overhyped AI project right now is OpenClaw. Not because it's bad. It's actually cool conceptually — a bunch of AI agents in a trench coat that you talk to through Telegram. The problem? Nothing it does is novel. It summarizes emails. Monitors competitors. Posts to Twitter. These are API calls with extra steps. And it burns through a billion tokens doing it. Every single use case can be done faster, cheaper, and more reliably with purpose-built tools wired together: n8n, Make, Claude … Code lol. Oh, and it's a pain to set up — ironic for something that's supposed to simplify your life. When users are pressed on what value they're actually getting, the answer is usually "it remembers my conversations". We're talking about a database. The hype is wildly out of proportion to what it buys you. If you're thinking about automation, don't start with the shiniest all-in-one tool. Start with the specific problem. Build around that. Purpose-built beats bundled. Every time.

The most expensive line item in your fund is invisible. You didn't lose that deal because you were outbid. You lost it because you were too slow. The founder had three conversations with another fund, built rapport, and went into exclusivity. By the time you knocked on the door, it was over. And the part that stings — that founder sold to the person they trusted. Not the highest number. People obsess over purchase price, IRR, and risk-adjusted returns when they find a deal. But no one measures: 1. Origination funnel speed from research decision to outreach action 2. Building enough trust that a founder who wasn't thinking about selling starts to consider it 3. Deals we never saw 4. Being a month late There's no line item for any of it. No IC memo. No post-mortem. And you know what actually affects IRR? Speed. That invisible cost — compounded across a fund life — is tens if not hundreds of millions of dollars. Every fund measures what happens after they find a deal. Almost none measure how fast they find it.

Living in Ohio connecting virtual lego blocks for private equity investors was not on my 2026 bingo card. Farm in Zimbabwe → liberal arts college in Massachusetts → investment banking in New York → PE in London → MBA in Texas → solopreneur in Ohio. That wasn't in my childhood journal. Back when I was recruiting for PE in London, I sent hundreds of cold DMs wondering: How do these guys actually find deals? Turns out I was sorta right. Deals don't just land in your lap. You have to go find them. I did a lot of that at my fund. Manually. Unsystematically. Events, LinkedIn searches, email blasts, grabbing coffees with other investors. After business school I decided to give this solopreneur thing a whirl instead of going back to PE. 80–90% of funds — including big ones — have no real system for proprietary sourcing. It's networks, broker deals, a rushed bit of market research, and a volley of emails. The process? Usually held together with duct tape and good intentions. Even the funds with "we use AI to intelligently source deals" on their website. Side note: a company ChatGPT subscription doesn't count as an AI system. You'd be amazed at what people say about themselves. A lot of it is smoke and mirrors. So now I build the actual infrastructure. White-label deal sourcing systems for PE and growth equity firms. AI that finds companies systematically, plugs into your CRM, and gives your team a pipeline that actually compounds. It's a lot of work doing everything yourself. But the gap between a ChatGPT subscription and an integrated system that actually runs is much wider than anyone thinks. And until that gap closes, there's plenty to build. Zimbabwe to Ohio. Stranger things have happened. Probably.

If I had to build a proprietary deal pipeline for a marina roll-up in South Florida, I wouldn't start with a database. I'd start with a satellite image. Marinas can only exist on waterfront. You can't build new ones — coastal permits, environmental regs, and finite shoreline see to that. Supply is locked. Every target already exists. You just have to find the right ones. In South Florida, any marina with 10+ slips needs an annual operating permit from the county. Public record. Slip count, operator, location. Filter to 50+ slips and you've got your target universe in an afternoon. Then layer the signals: → Satellite imagery shows dock condition, occupancy, and adjacent waterfront for expansion → Google reviews reveal management problems no spreadsheet will tell you → Florida publishes boat registrations by county — that's demand density → Single location + incorporated 20 years ago = succession opportunity → Median income within 10 miles = pricing power A marina with aging docks, a 4.2-star rating mentioning "new management needed," 60 slips on the Intracoastal, and an owner who incorporated in 1998? That's not in anyone's CRM. But it's visible from space.

Pessimists get to be right. Optimists get to be rich. I think Shaan Puri from My First Million said that. Doomsdaying about ai is a hot topic right now. Fear gets clicks, I guess. And humans have been doing it since that apple was eaten. But you know what's hard. Optimism. Doomsdaying requires no effort, no experimentation, no risk. Just loud opinions. My guess: those are the people who ai will replace. But the ones in the arena. The ones figuring out what can be done that couldn't be done before. The people who think of ai as outcomes & leverage. Those are the people that will have a place in the post ai world. Ai doesn't have judgment or curiosity. It can't go for a walk or let thoughts percolate in the shower. But optimists can. And with ai, they can give those thoughts some leverage. How exciting.