Loading page…
The first cloud company won by being cheaper, not better: tens of thousands a year against Siebel's one to two million, sold by calling Siebel's own customer list and asking again every month until they said yes.
roughly $26B guided for the year, with a forecast of $50B by fiscal 2026 · valued north of $270B · about 20% of a $100B CRM market · 150,000 customers, including 90 of the Fortune 500 · Sales Cloud ~$5B and Service Cloud ~$5B as standalone businesses · ~90% logo retention, over 100% net retention · 73% gross margin, ~18% operating margin, ~16% cash-flow margin · IDC: every $1 of Salesforce revenue creates $6 of revenue across its partner ecosystem
The early reps had no lead scoring and no funnel. They pulled the list of Siebel customers, called every one to ask if they wanted to buy Salesforce, and if the answer was no, called back in a month or two to ask again.
Fewer bars = easier, cheaper, or faster for an AI-assisted solo builder. Editorial judgments based on the case details.
Marc Benioff left Oracle in 1999 with an insight and a co-founder. The insight was that enterprise software did not have to arrive on discs and live on your servers; the co-founder was Parker Harris, plus a couple of others. The obstacle was that he had to convince companies to hand their contact data — the most valuable list they owned — to a third party, at a moment when AWS did not exist. He did it with two words. "No Software" was the message, because customers could not picture the cloud but they could vividly picture the pain of discs, reinstalls and service contracts. The commercial argument underneath was even simpler: Siebel implementations ran one to two million dollars, and Salesforce sold for tens of thousands a year. So they went to CFOs and to the small and mid-sized companies nobody was serving. Twenty-plus years later the same business did about $6.3 billion in the quarter discussed, up 23%, with roughly $26 billion guided for the year, a $50 billion forecast for fiscal 2026, and a valuation north of $270 billion. It holds about 20% of a $100 billion CRM market across 150,000 customers. The part the guest thinks is most underrated is not the cloud delivery — it is the platform. force.com let customers build what Salesforce did not; AppExchange let third parties plug in without integration work; the APIs shipped early and stayed good. That combination produced an economy: IDC found every dollar of Salesforce revenue generates six dollars of revenue for partners building on it. Matt Garrett, who ran Salesforce Ventures for part of eight years there, sums up the pattern as focus. Front office, customer data, nothing else — and every acquisition compounding onto the last.
The first customers came from the incumbent's own customer list, worked by brute force. Early reps qualified leads by pulling whoever was a Siebel customer, calling to ask whether they would like to buy Salesforce, and calling back in a month or two to ask again if the answer was no.
The quick-reference card — acquisition channels, replication playbook, risks & traps, and our verdict — is members-only.
Join the Unicorn Club — $5/moCancel anytime, in one clickPaste into Claude Code / Codex and get a working version of this product end-to-end.
Two production-grade prompts per idea — a full build spec and an SEO growth plan — plus 10 tailored build specs a week. Members only.
Join the Unicorn Club — $5/moCancel anytime, in one clickEvery claim traces back to a listed source — open any of them and check the number yourself. The exact words are quoted below, with a timestamp that jumps to the second each figure is said.
“…Salesforce has grown rapidly to become the global leader in the 100 billion dollar CRM Market the business has 150 000 customers including 90 of the Fortune 500 and is currently valued…”
2:27 · hear it in the video“…so if you look at their Q2 number Salesforce did about 6.3 billion in Revenue that's up 23 in year-over-year growth they've given guidance that they'll do about 26 billion of…”
5:09 · hear it in the videoData credibility: Podcast breakdown, Business Breakdowns episode 29, published May 2023. The host is not named in the transcript; the guest is Matt Garrett — general partner at CRV and formerly head of Salesforce Ventures, with eight years inside the company. Graded verified hard data because the figures are public-company numbers: a quarter's revenue of about $6.3B at 23% growth, roughly $26B of full-year guidance, a $50B forecast for fiscal 2026, a valuation north of $270B, 150,000 customers, ~73% gross margin, ~18% operating and ~16% cash-flow margins, plus a named third-party study from IDC putting $6 of partner revenue against every $1 of Salesforce revenue. The important caveat is that these are quoted from memory in conversation rather than read off a filing, and the episode is from 2023 — treat every number as of that recording, not of today. Two specific caption problems: the auto-transcript strips percent signs, so "90 of the Fortune 500" may be 90 companies or 90%, and we have left it as the transcript has it rather than guessing; and the line about roughly 17% of employees being quota-carrying sales staff is garbled enough that we flag it rather than rely on it. The auto-caption also mangles proper nouns — Siebel appears as "sequel" and "sibo", nCino as "Casino Encino", Veeva as "Vivo", Dreamforce as "Stream force" — and those spellings are corrected here from context; ExactTarget's $2.5B purchase price and the 2013 date are the guest's recollection. Salesforce Ventures figures — from $10-20M a year to about $1B a year, $4B on the balance sheet, 300-400 investments — come from Garrett's own tenure running it, which makes them first-hand but still unaudited here.
The first dollars arrived as subscriptions priced per seat, at tens of thousands of dollars a year against a competitor charging one to two million for a licence plus the service contract around it.
Retention turned it. About 90% of customers stay, and because seats and modules expand inside each account, net retention runs above 100% — the installed base grows without a single new logo.
The hard stretch here was not silence, it was pressure to take the wrong money. Large financial customers offered substantial sums for an on-premise version of Salesforce, which the company would have profited from immediately.
At the bottom it is a customer database with workflow on top. Everything else is a layer over that. Sales Cloud began as sales force automation — pipeline, ownership, deal size, close dates, so a large team stops standing on each other's toes. Service Cloud logs the support side of the same relationship and is now the bigger and faster-growing of the two, each around $5 billion on its own. Marketing Cloud, acquired as ExactTarget, is the email and customer database behind campaigns. Commerce covers the transaction. MuleSoft moves the data between them, and Tableau visualises it. The platform is the part the guest calls most underrated: force.com exposes the data objects as a flexible database with Apex, a language of its own, on top — genuinely low-code for simple workflows and fully programmable when it is not. Our read: the product a buyer actually purchases is the single source of truth about a customer, and every module is a reason not to move it. That is why the guest's framing of one product getting you to a $10 billion market cap and a platform being required for $100 billion is the sentence worth keeping. A product is bought; a platform is built on, and the things built on top are switching costs you did not pay for.
Per-seat subscriptions, expanded two ways: more seats inside the same enterprise, and more products sold to the same account. About 90% of customers renew and the account base expands, producing net retention over 100%. Gross margin is 73% because there is no incremental cost to serving another seat. Sales and marketing is the largest expense line at more than 30% of expenses, and the guest is unsentimental about why: it is the engine, not an overhead. High-growth SaaS companies routinely spend 50% to 125% of revenue on it, because bookings recur and the spend does not. Acquisition became the second engine. Small talent deals built the muscle first; the 2013 purchase of ExactTarget for $2.5 billion established the pattern of buying category leaders when building would take too long — Tableau in analytics, MuleSoft in integration, Quip and Slack in collaboration. Salesforce Ventures ran alongside it, scaling from $10-20 million a year invested to about $1 billion a year with $4 billion on the balance sheet, and three to four hundred investments produced only a handful of acquisitions. Our read: the discipline is the interesting part. Everything bought had to be front office and customer data, which is why the acquisitions compound instead of sprawling.
Gross margin 73%, operating margin around 18%, cash-flow margin around 16%. The gap between the first and second numbers is the sales and marketing spend that lands before the revenue it buys, and the near-parity of the second and third is the subscription model's quiet advantage: cash is collected up front even as revenue is recognised across the year. On headcount, the guest gives a rule of thumb that SaaS businesses land near 17% quota-carrying sales staff, though the caption is garbled at that point and we would not build a plan on it. The operating system of the company is the V2MOM — vision, values, methods, obstacles, measures — filled in by every employee, cascading from Benioff downward, revisited two or three times a year, and forcibly prioritised rather than left as an infinite list. It was paired with "hack the V2MOM", an executive off-site broadcast to the whole company where several hundred people out of fifty thousand could stand up and say a part of the business was wrong, with no sacred cows. Our read: this is the most portable asset in the whole breakdown, and it costs nothing. A one-page prioritised plan per person, reviewed three times a year, with a public forum for saying what should stop — that works at five people as well as fifty thousand.
Copy the message discipline. One claim, the one the buyer already feels, beats five advantages — "No Software" and "it costs a lot less" did more work than any feature list, and the guest names diffuse positioning as the mistake he sees most in founders. Copy the asymmetric attack. Salesforce beat Siebel not by being better but by choosing the axis Siebel could not follow on: matching it would have meant destroying their own licence and services revenue. Before building anything, find the thing the incumbent is structurally unable to do, not the thing they have not done yet. Copy the ecosystem instinct early. Good APIs and a place for third parties to plug in look like premature infrastructure at small scale and turn into the moat later. The six-to-one partner revenue ratio did not appear by accident. Change the capital assumption. This case was built with venture money, an enormous sales organisation and a decade of acquisitions. The solo version is the same wedge at a smaller scale: one vertical, one incumbent charging six figures for on-premise software, a hosted product at a tenth of the price, and a customisation layer so your customers extend it themselves instead of asking you to.
The guest's own answer is that nothing kills it soon — do nothing for five years and the model, the sales team and the product quality still hit the numbers. The threats he names are slower. The first is the SMB end, where a newer generation of tools is easier to buy, implement and run. He expects these to nip at margins and take a little share rather than to displace anything. The second is data gravity moving. As companies pull everything into Snowflake or Databricks and then push results back into their applications, the value of holding the data erodes — and applications that run directly against the customer's own warehouse are, in his words, potentially very disruptive. The third is bottom-up adoption through engineers. Open source and developer-first tools get adopted without a sale, then get handed to the business side. His example is a renewal where the customer realises they are only using 30% of a vendor's product and reopens the contract. Our read: for a reader, the risk that matters is misreading the era. Nothing here is a template for 2026 as-is, and the figures describe a specific quarter discussed in an episode published in 2023, not today's Salesforce.
The four founder playbooks — first customers, first dollar, the flywheel, the silent stretch — are in the Unicorn Club, on this idea and on all 1,403 others. The story, the numbers, the sources and every later section stay open to everyone.
Join the Unicorn Club — $5/moCancel anytime, in one click