Anand Sanwal

The 3-tier price anchor: PDF, PDF+call, PDF+call+hotline

CB Insights' early advisor Nick priced the same credit-card survey PDF at three tiers ($12K PDF, $50K PDF+call, $100K PDF+call+on-demand hotline limited to 10 buyers). The top 'feet' tier almost never sold but anchored buyers toward the middle $50K package.

“no, no, no, the way we're gonna do this is $12,000 just gets you the PDF. $50,000 gets you the PDF and a call. And $100,000 gets you the PDF, a call, and anytime we hear something juicy, we pick up the phone for you. And only 10 of, only 10 people can get that one, right? So like we get off the, it's like OnlyFans for hedge funds.”

Steal thisAdd a deliberately expensive top tier nobody buys to anchor buyers into your real target package.

CB Insights made ~$700K selling credit-card sentiment PDFs

Before CB Insights existed, Anand Sanwal's team made roughly $700,000 in about a year selling a credit-card sentiment survey PDF built from interviews with ~25 industry insiders. That cash funded CB Insights.

“Uh, we probably made like $700,000.”

ECO: Edge, Collect, Opportunity for any data business

Anand's test for whether a data business will work: data has no intrinsic value, only what edge it gives the customer. Run every dataset through Edge (what advantage does it create), Collect (can you feasibly gather it), and Opportunity (how big and data-hungry is the vertical).

“So I think of it as like 3 things, like I call it like ECO. So you got your— the edge, you have to define what that's going to be. You have to, the C is sort of, can you collect it? And then the O is just the opportunity, right?”

Steal thisBefore building any data product, define the customer's edge first, then prove you can collect the data, then size the vertical.

CB Insights' data started as manual 'ground and pound'

CB Insights' first data collection was brute force: Anand and his team manually read ~50,000 funding/M&A articles and typed investors, amounts, and valuations into spreadsheet columns before engineers automated it. Many great data businesses start with 'data janitor' work.

“And then collection was, in the beginning it was just ground and pound. Like it was, we've had 50,000 articles about funding and M&A events. And me and the guys just went through manually and put in columns in a spreadsheet. Here's the investors, here's the amount, here's the valuation, whatever we could find.”

Data to data-co-op to workflow tool

Anand's admired formula (seen at Razor's Edge): start by ground-and-pounding a proprietary dataset, then let customers upload and clean their own data so everyone benefits from a pooled 'data cooperative', then add a workflow/CRM tool on top to embed in the customer's daily work.

“And so like, it's become a, like a, a pooled data, data cooperative, and then they've added workflow on top of it. So I think that that's an, an, really impressive formula. Going from data to data co-op to workflow tool, right? I think like that's kind of the future, right? Because you want to get into people's workflow.”

Steal thisLayer a workflow tool on top of pooled customer data so you become embedded software, not just a dataset.

Data to data-co-op to workflow tool

Anand's admired formula (seen at Razor's Edge): start by ground-and-pounding a proprietary dataset, then let customers upload and clean their own data so everyone benefits from a pooled 'data cooperative', then add a workflow/CRM tool on top to embed in the customer's daily work.

“And so like, it's become a, like a, a pooled data, data cooperative, and then they've added workflow on top of it. So I think that that's an, an, really impressive formula. Going from data to data co-op to workflow tool, right? I think like that's kind of the future, right? Because you want to get into people's workflow.”

Steal thisLayer a workflow tool on top of pooled customer data so you become embedded software, not just a dataset.

IMG Academy but for entrepreneurship: go pro in business out of high school

Anand is building an in-person 6th-12th grade boarding school for entrepreneurship, modeled on IMG Academy for sports. The thesis: if kids can go pro in sports out of high school, they should be able to go pro in business.

“So what I'm trying, what we're working on is building a school of entrepreneurship. So it's an in-person 6th to 12th grade school. You know, if you know the IMG Academy, right? Like for sports. Right? Like the basic idea is like people go pro in sports outta high school. Like you should be able to go pro in business.”

Sloomoo slime museums hit $30M revenue with tiny headcount

Anand's napkin math on one NYC slime museum (Sloomoo): ~500-600 visitors/day, $40 tickets plus $30 slime-dump upsells, only ~10 staff. He estimates the single location does $6-8M and read the chain hit $30M across NYC, LA, and Atlanta.

“I think I just read that they did $30 million. Now they have LA, Atlanta, a couple other places. So I think last year they did $30 million at these slime museums.”

Dilo: unstaffed Amazon-Go convenience stores inside apartment complexes

Anand describes Dilo, which installs compact, 24/7 unstaffed Amazon-Go-style convenience stores in multifamily complexes of 200+ units. Dilo funds the ~$100K buildout, restocks, and cuts the property manager in on revenue, turning the store into a tenant amenity and revenue stream.

“And so, you know, they go to the property developer, the property manager, say, listen, I'm going to put up, we'll put up the $100,000 to build this out and we'll restock it and we'll cut you in as a property manager on some of the revenue. You know, once we sort of recoup our investment. And so I thought like, this is an amenity now that the property manager can offer to their tenants.”

Steal thisSell B2B2C: let the landlord be your distribution channel by giving them an amenity plus a revenue cut.

Online-addiction rehab centers in cheap commercial real estate

Anticipating fallout from mass legalized online gambling, Anand pitches offline 'online addiction centers' for phones, porn, DraftKings, and Robinhood. The wedge: nerdy phone-addicts are lower-risk tenants than hard-drug patients, so you can get distressed commercial real estate cheap.

“I think there's a, again, another offline play in online addiction centers, right? So parents want their kids off the phone. I'm addicted to porn. I'm addicted to DraftKings. I'm addicted to, you know, trading naked calls on Robinhood, like whatever it might be. And so you go there and I think the interesting thing here is there's all this commercial real estate that's available.”

Chasing Perfection: a high school football book that's really a leadership book

Anand recommends 'Chasing Perfection,' on the De La Salle high school football dynasty (the winningest in history). He says the upfront chapters on leadership, motivation, and accountability, where players feel a deep sense of responsibility to each other, should be read by anyone in a company.

“Like it is unreal because these kids aren't responsible to just the coaches or their school. Like they feel a really deep sense of responsibility to each other. And I actually think like that's something we need more of. And I really like, there's, I think there's a lot of lessons that can be applied from youth development and sports.”

Sawdust Data: sell the exhaust data from other companies' products

Anand pitches building a platform that takes the data 'residue' from existing products (OpenTable reservations, dog-collar trackers, procurement systems) and sells it to outside buyers like real estate firms, pet insurers, and hedge funds.

“sawdust data is people create products and sometimes the residue of that product is actually really interesting data, right? And so that's the sawdust from the data. And so it's basically taking that exhaust. And so I'll give you a couple of examples, right? You have, I'll just, and I don't know if these are businesses already, but you know, OpenTable and Resy have all this reservation data. I wonder if you could take that data exhaust and sell it to real estate firms”

Steal thisFind a company sitting on data exhaust it doesn't monetize, then build the data product that sells it to an outside industry that craves the signal.

Variable-SKU, opaque, high-consideration data businesses

Anand's filter for a defensible data business: it must be high-consideration (real money/risk on the line), opaque (hard to find, people ask in Slack groups), and variable-SKU (products too different to drop into a comparison grid). Hitting all three signals a painful but moated opportunity.

“my framework is variable skew, highly opaque, high consideration. And so I'll break that down. So high consideration means like somebody's taking risk or putting a lot of money out there, right? It's just like, it's got— there's some skin in the game. Opaque, meaning it's just hard to find, you know, and I think if you go into Slack groups and people are like, hey, does anybody know X? That's usually an asset class that's like opaque”

Steal thisScore any data-business idea on three axes: high consideration, opaque, variable SKU. Pursue only the ones that hit all three.

Representation Media: clone Rebel Girls for every community

Anand notes the Rebel Girls book sold ~5.5M copies at $20 (~$110M) and pitches extending the format to other underserved groups (Hispanic entrepreneurs, autistic scientists), since micro-communities want inspiring stories featuring people like themselves.

“That book is sold, I think 5.5 million copies at like $20. So it's like $110 million and that's probably dated, right? But I think there's like this idea of like what I, what I call representation media, right? I bet you could extend that to I don't know, like Hispanic entrepreneurs or like autistic scientists”

Steal thisTake a proven inspirational-story format and re-cast it for a specific identity community that has no equivalent of its own.

Unsexy data businesses keep the smart competitors out

Anand argues build-once-sell-many data and information businesses (Informa, Euromoney, Cision) are great precisely because they're unsexy: nobody glamorous shows up to take your lunch money.

“it's a build once, sell multiple times kind of business, right? So like once you build that data asset or that info asset, like you can keep selling it. So yeah, they're great businesses. They're unsexy, which is always good because like then you don't get all the smart people kind of trying to come in and, you know, eat your, take your lunch money.”

Scrape hidden B2B SaaS pricing and sell it back

Anand pitches crawling public SaaS pricing pages and, for enterprise tools that hide pricing, hiring people to pose as buyers to extract quotes, then selling the assembled pricing dataset because B2B pricing is a 'dark art' where everyone has opinions and no one has data.

“you actually, cause a lot of enterprise companies don't put their pricing on their page. So you hire somebody who goes and pretends that they're a customer and gets the pricing. And then that's really, really valuable, right? Because like now if I'm— because pricing is like a dark art in B2B, you're like, yeah, sounds good, right? And like everybody's got opinions and nobody has data.”

Steal thisCollect the pricing competitors hide behind 'request a demo', package it as a benchmark dataset, and sell it to CROs and product teams.

Why $395/month attracted CB Insights' worst customers

Anand recalls launching CB Insights at $395/month, which drew low-paying, high-support churners. A consulting firm told him to add a couple of zeros so the price wouldn't look like a joke internally, teaching him that price itself signals quality.

“Our first CB Insights when it started was $395 a month. And what you actually got was a bunch of people who didn't know how to use data, who were a pain in the ass, right? And so it was like your worst nightmare. It was people who churned quickly, who paid you little, and who were blowing you up on chat all the time for support. And then I remember meeting a management consulting firm and they were like, hey, can you add a couple zeros to the price? Because like, one, your price will look like a joke internally”

Steal thisRaise your B2B price until it signals quality; cheap pricing recruits your worst, highest-support, fastest-churning customers.

College rankings based on graduates' actual W-2 income

Anand pitches disrupting US News rankings (which schools game on subjective inputs) by collecting graduates' W-2s and ranking colleges purely on tuition-paid vs. income-earned, starting niche with one state and expanding in concentric circles.

“I think the thing you do here is you actually go to graduates of the school and just ask them for their W-2s, and you're like, how much are you making, right? You just took on a bunch of debt. Like, does this place actually get you a job that gets you paid? Or are you like, you know, a server in a restaurant and you got like your $200,000 in debt? And so I think you reach out to alumni, go get their W-2s, and basically put a ranking together of here's how much I paid and here's how much I make when I get out.”

Steal thisRank colleges by ROI using alumni W-2s instead of gameable inputs, starting with one region before expanding.

Door-to-door pool pricing database

Anand's favorite side hustle: pool pricing is opaque, variable, and high-consideration, so send someone door to door asking homeowners what they paid for their pool and monthly cleaning, then build a local pricing database buyers can't find anywhere else.

“Some kid should just go door to door at every person's house that has a pool and just be like, "How much did you pay for your pool?" And get the specs and then go find out and find out what do they pay per month for cleaning and whatever. I don't know what a pool requires. And like literally build the database of like—”

Steal thisManually collect pricing that nobody publishes (door-to-door, phone calls), then sell the resulting database to both buyers and vendors.

ChatGPT will make private data far more valuable

Anand predicts that as AI replaces the old 'give Google your data for traffic' bargain, owning private, hard-to-get data becomes much more valuable, making him extra bullish on proprietary data businesses.

“I think this whole ChatGPT thing is actually going to make private data like that much more valuable because like, you know, before it was like, oh, I'll give you my data and you'll just link to me and I'll get Google. But like, like that's all going away, I think. And so like having access to private data is going to be really valuable. So yeah, I'm, I'm extra bullish.”

Prefer data that requires constant refresh

Anand's counterintuitive rule: build data products around data that goes stale fast, because constant refresh is what keeps subscriptions alive. One-time data (like pool pricing) can still work via lead gen and advertising instead of subscriptions.

“I think you probably want data, maybe counterintuitively, that requires a lot of refresh, right? Because like that's what keeps people subscribing, right? If I can get the data once and it's good for 3 years, it's like, all right, I'll get it, I'll churn, I'll come back to you in 3 years when it changes. When it's actually a little bit more— when there's a little more volatility to it, it's like, hey, I need to know what's going on because like my business might suffer”

Steal thisBuild subscription data products on volatile data that decays fast; monetize one-time data via leads and ads instead.

Anand ran his late father's India chemical factory remotely for 18 months

After his father died in 2017, Anand ran the family's fine-chemicals factory in India alongside CB Insights, optimizing the eventual sale not for price but for a buyer who would keep his dad's team employed rather than tear it down.

“People kept reaching out to buy it, you know. A lot of them, like, you know, when somebody— once a proprietor dies, like, a lot of bottom feeders show up. So I was just like trying to make sure that these weren't those people because I want to make sure the team landed on their feet.”

Treat 'industry standard' as a red flag in negotiations

From his M&A experience, Anand warns that 'industry standard' is the phrase people use when they can't justify a term on its merits; it's always standard in favor of the other side, so it's a signal to push back rather than accept.

“when I hear it's industry standard, like my ears always perk up because it's like somebody's like trying to get one over on me on this, right? Or they're just trying to like make it— sweep this under the rug.”

Steal thisWhen a counterparty justifies a term as 'industry standard', stop and negotiate it; that phrase signals it favors them, not you.

When the innovation is the cap structure, run

Anand observes that convoluted org structures (FTX, Adani, Enron) get praised as genius in the moment but are usually shell games; if a company's real innovation is its capital structure or HR gimmicks instead of its product, that's a warning sign.

“when you look back, it's always like, no, it was just like a giant shell game. You know, Sean said like that, that's an interesting lens, I think, to look at companies through. And it's like the innovation isn't the product. The innovation is like cap structure or some weird thing. It's like, okay, that's a weird, that's a kind of a weird business to build, right? Where like, that's what you're innovating on.”

Spot weak startups via 3x liquidation prefs and exec churn

Anand says fraud is hard to catch, but underperformance isn't: signals include billion-dollar raises carrying unreported 3x liquidation preferences and a CRO or CMO leaving every six months, both of which indicate a company isn't doing as well as outsiders think.

“you'll see companies raise at a billion-dollar valuation, and when we dig in, we see like the liquidation preference is 3x, but that stuff doesn't get reported, right? Or you see the CRO or the CMO leaves every 6 months, kind of to Sean's point on the CFO. Like, that indicates there's problems.”

Steal thisRead past the headline valuation: high liquidation preferences and serial C-suite departures reveal a struggling company.

The Schultz strategy: interview adjacent, not identical, guests

Anand credits Andrew Schultz with growing his audience by interviewing people adjacent to (not the same as) his comedian peers, e.g. Theo Von booking Neil deGrasse Tyson, and applies the same low-overlap logic to B2B partnerships.

“this sort of like Schultz strategy of like adjacent folks, I think, is really smart to bring in. Like, the Venn diagram is not as heavily overlapped, and I think like he keeps building that way, which I think is really interesting. And so for us, like, it's always like, who has an audience that's not exactly our audience, but that has maybe a slight overlap?”

Steal thisGrow your audience by partnering with adjacent, low-overlap audiences rather than direct peers who share your existing followers.