← all stories

After the Web · Era 4 · Dot-coms and the crash

1998–2016acquired by Oracle for $109 a share (≈$9.3B) · 2016

NetSuite

NetSuite was a California company that from 1998 sold small businesses accounting, record-keeping, and CRM not in a box but by subscription through the browser. It was founded with partners by former Oracle developer Evan Goldberg with Larry Ellison's money; it was listed on the stock exchange from 2007, and every year from 2002 to 2015 it operated at a loss. In 2016 Oracle, where that same Ellison was chairman of the board of directors, bought it for $109 a share, about $9.3 billion.

Founders Evan Goldberg
Domains netsuite.com
saassubscriptionerpaccountingcontrolled-company

In brief

NetSuite was a San Mateo company, incorporated in September 1998 and headed by co-founder Evan Goldberg, a former Oracle developer, with Larry Ellison's money. It sold small businesses accounting, then an online store and CRM, as a single subscription through the browser. By its 2007 IPO it had 5,400 active customers and an accumulated deficit of $241.6 million; it closed every year from 2002 to 2015 with an accounting loss, and in 2016 it went to Oracle at $109 a share, about $9.3 billion, while its largest owner was that same Ellison, the buyer's chairman of the board and chief technology officer. It is a story about how long a company can live at a loss when its owner has an almost bottomless wallet, and about how such a deal had to be answered for in court for almost eight years.

How it started (the founders)

From August 1987 to November 1995 Goldberg worked in development at Oracle and rose to vice president of development in the New Media division; he then founded mBed Software, which made multimedia tools for websites, and was its CEO until September 1998. So by the time of the founding he knew Oracle from the inside, knew the web, and knew Ellison personally estimate: there are no direct documents about the acquaintance, but the link between a former employee and a former employer is visible from his biography.

The first version of the founding myth goes like this: a short phone call between Goldberg and Ellison, in which Goldberg proposes a tool for working with sales and Ellison insists that the back office, accounting and ERP, has to come first rumor: there is one source, and it is secondary. What the documents show is something else: the company was incorporated in California in September 1998, the first application, NetLedger, with a focus on accounting, came out in 1999, the online store in 2000, CRM and sales force automation in 2001, and the suite was released under the NetSuite name in 2002. The order of accounting first and sales second matches what gets retold, and differs from Salesforce, which started with salespeople.

The money came mainly from Ellison through Tako Ventures. The figure of $124–125 million is cited only in retellings rumor; the S-1 gives no amount, only that by 2005 Tako was the "majority stockholder" and had given the company a line of credit. Ellison was chairman of the board until 2003; in January 2003 the management was restructured: Nelson became CEO, and Goldberg CTO and chairman.

Year-by-year timeline

Lesser-known but significant facts

  1. Ellison was not only the owner but also a lender secured by all the assets. From 2005 there was a $10 million Tako Ventures line of credit, from 2006 a $20 million one, at prime + 1% and secured by "substantially all of the Company's assets"; the $8.0 million balance as of 2007-09-30 was to be repaid with IPO proceeds.
  2. In 2002 the company sold for less than it cost to serve. Revenue of $3.1 million, cost of revenue of $5.3 million, a gross loss of $2.1 million, and $9.9 million went to sales and marketing. This was not a slow start but a deliberate payment for growth.
  3. Ellison gave up the vote of his shares before the IPO. He transferred 31,964,898 shares to an LLC whose agreement neutralizes their vote while he works at Oracle, with the stated aim of avoiding and mitigating possible conflicts of interest. Nine years later it was this very structure that became one of the arguments of the defense (the court did not find him to be a controlling stockholder) estimate. A caveat: the LLC agreement neutralizes the vote only in certain situations, and Ellison kept the right to vote on a merger for cash or securities.
  4. NetSuite bought Oracle licenses from Oracle. From 2005 there were three agreements for the database and the application server: $2.5 million, $0.9 million, and $4.2 million plus $1.4 million for support; the prospectus itself calls Oracle a potential competitor. From 2005 NetSuite's cloud ran on the licensed database of its future buyer.
  5. The IPO went by auction and priced above the assumed price. The assumed price in the S-1/A was $20.50, the midpoint of the $19–22 range; the result was $26; the range was raised twice during marketing estimate.

Legend vs. the record

The first growth lever

Not a channel but a source of money. Revenue grew 1.8–2.7 times a year: $3.1 million (2002), $8.3 million (2003), $17.7 million (2004), $36.4 million (2005), $67.2 million (2006). Sales and marketing meanwhile cost $9.9 million, $15.4 million, $27.0 million, $39.2 million, and $43.9 million, more than revenue in 2002–2005 (about 152% in 2004 and about 108% in 2005). The owner covered the difference, with equity and with a $10–20 million loan secured by all the assets. The second part of the lever was the suite: the customer bought accounting, the store, and CRM together; according to the company itself, by 2003 it had more than 6,000 customers estimate. The subscription pricing scheme does not appear in the primary sources unverified.

The paired story

Parallels today (projects from the catalog)

What a builder can take from this in 2026

  1. A subscription needs a wallet, not just an idea. With revenue of $3.1 million in 2002, spending on sales and marketing was three times larger; the company survived because its owner was willing to lend. Today, first count the months of runway at a negative margin, not just the growth.
  2. A loss during growth is not a death sentence if cash flow is growing too. In 2015 the loss was $124.7 million with operating cash flow of $100.4 million, according to the company's press release estimate. Track GAAP and cash separately, as NetSuite did.
  3. Put the conflict of interest on paper before the deal. The transfer of the shares to the LLC and the renunciation of the corporate opportunity doctrine in 2007, the independent committee and the majority of unaffiliated shareholders in 2016: these papers became part of the defense, and the case was decided in Ellison's favor in 2023 and 2025 estimate. If you have a large strategic shareholder, formalize this on the day they come in, not on the day of the sale.
  4. Dependence on the buyer's database is both a risk and an asset. Three Oracle licenses made NetSuite cheaper and at the same time more dependent; the platform you stand on is better formalized in a contract with a reciprocal clause.
  5. Do not call yourself the first without a document. The slogan about the very first cloud company is easy to check: incorporation in 1998, product in 1999. Today the same applies to any claim to be the first AI something: a verifiable date matters more.

Discrepancies and what we could not verify

Sources (primary first)

Internet History Keeper

Every historical dossier here is free and open, and it will stay that way. If you want the series to keep going (new dossiers, checks against primary sources, the English edition), support it with a subscription for $9.90 a month. It is support, not access, and you can cancel at any time.

Become a keeper — $9.90 a month →

The same thing, about today

The same breakdowns, but of projects launching right now: what the product is, where the first users came from, how they charge. The card is free, the full dossier is $5 (the dossier itself is written in Russian).

Browse the dossier catalog →