Skip to content

Databricks Just Signed a Term Sheet at a $188 Billion Valuation, Up 40% From February

DS
LDS Team
Let's Data Science
9 min
Databricks signed a term sheet on July 16 for a strategic funding round led by Coatue Management, putting the company's valuation at $188 billion, a 40 percent jump in five months. The Wall Street Journal reports the new round totals roughly three billion dollars, though Databricks has not confirmed the exact figure and the cash has not arrived yet. CEO Ali Ghodsi says the money will fund the tools enterprises use to control what they spend feeding data into AI models.

On June 9, The Information reported that Databricks was in talks for a new funding round, one that might value the company somewhere in the $165 to 175 billion range. Nothing was signed. Nothing was final.

On July 16, the talk became paper. Databricks signed a term sheet for a new funding round at a $188 billion valuation, a number higher than anything floated during the June negotiations.

The round is being led by Coatue Management, an existing investor, and Databricks expects it to close later this summer. The company has not disclosed how much money is actually changing hands. The Wall Street Journal first reported the figure, roughly three billion dollars, and Reuters has since repeated it, though Databricks itself has not confirmed the amount. The company says the capital is aimed at three specific products: Unity AI Gateway, Genie, and Lakebase, the tools Databricks is betting will carry it past its old reputation as a data warehouse company.

A Signed Term Sheet, Not a Wire Transfer

A term sheet is a promise, not a payment. It locks in price and terms so lawyers can finish the paperwork and investors can send money, but it is not the money itself. Databricks was explicit about this in its own announcement: the $188 billion figure comes from a signed term sheet, and the company expects the round to close "later this summer."

TechCrunch called the timing unusual. Companies typically wait until cash is in the bank before announcing a new valuation. One venture capitalist told the outlet the deal is solid regardless, with so many firms competing for a piece of Databricks that there was no reason to keep a higher valuation quiet.

The company has reason to want its number in headlines now. Databricks is privately held, but it operates at a scale most public companies would envy: more than 20,000 organizations, including adidas, AT&T, Bayer, Block, Mastercard, Rivian, and Unilever, run on its platform, along with 70 percent of the Fortune 500.

The Valuation Kept Climbing

This is not Databricks' first trip through the venture-funding wringer this year. The company closed a private round in February at a $134 billion valuation, then spent the following months watching investors speculate about what would come next.

For context: LDS covered the talks stage on June 9, when The Information first reported Databricks was negotiating a round in the $165 to 175 billion range. Neither figure survived contact with the final term sheet.

Run the two confirmed numbers, and the jump holds up: a 40.3 percent increase, which lines up with what Databricks, Benzinga, and MarketScale all independently describe as roughly 40 percent.

DateValuationRound TypeLead Investor(s)
December 2024$62 billionGrowth round, $10 billion raisedMultiple investors
September 2025$100 billionValuation confirmed, tied to $4 billion revenue run rateNot disclosed
February 2026$134 billionSeries L, roughly $5 billion raisedInsight Partners, Fidelity, J.P. Morgan Asset Management
June 2026$165 to 175 billionReported in talks; round never confirmed at this rangeNot reported
July 16, 2026$188 billionStrategic round, term sheet signed, roughly $3 billionCoatue Management

Three of those five milestones landed within the past thirteen months. Few private software companies climb a valuation ladder that fast.

How Databricks Got Here

The climb did not start this year. Here is the full sequence, going back to the round that first made Databricks a household name in venture capital.

DECEMBER 17, 2024
A record-breaking round
Databricks raises $10 billion at a $62 billion valuation, a record for a private software company at the time.
SEPTEMBER 8, 2025
Valuation confirmed at $100 billion
Databricks confirms a $100 billion valuation, tied to surpassing $4 billion in annualized revenue.
DECEMBER 16, 2025
A new Series L is announced
Databricks says it has raised more than $4 billion at a $134 billion valuation, led by Insight Partners, Fidelity, and J.P. Morgan Asset Management.
JANUARY 23, 2026
A separate $1.8 billion in debt
Databricks secures $1.8 billion in debt financing from institutional investors, unrelated to its equity valuation.
FEBRUARY 2026
The Series L closes at $134 billion
Total commitments reach roughly $5 billion. Databricks discloses a $5.4 billion revenue run rate, up 65 percent year over year.
JUNE 9, 2026
Talks of $165 to 175 billion surface
The Information reports Databricks is negotiating a new round in that range. Nothing is signed.
JULY 16, 2026
The term sheet is signed at $188 billion
Coatue leads a strategic round, reportedly worth $3 billion. The deal is expected to close later this summer.

What the Money Is Actually For

Databricks framed the raise around a shift it says its own customers are going through. In a statement, co-founder and CEO Ali Ghodsi explained the thinking:

"Enterprises are moving from tokenmaxxing to valuemaxxing. They don't want to burn expensive tokens on the smartest model for every task — they want the best outcome per dollar. That means having the freedom to choose the right AI for the job. This new capital lets us keep pushing our multi-AI strategy forward to meet massive customer demand, so we can keep strengthening Unity AI Gateway, expanding Genie, and advancing Lakebase." — Ali Ghodsi, Co-founder and CEO, Databricks (Databricks Newsroom, July 16, 2026)

Translated out of Silicon Valley shorthand: companies got burned running every task through the most expensive frontier model available, and now want the cheapest model that still gets the job done. Three products carry that bet.

  • Unity AI Gateway: a governance layer that routes requests across different AI models and tracks what each one costs, so a company is not paying frontier-model prices for a task a cheaper model handles just as well.
  • Genie: Databricks' term for an AI "coworker" that turns a company's internal data into answers and, increasingly, actions, without an engineer writing a query first.
  • Lakebase: a serverless Postgres database, built on the company's 2025 acquisition of Neon, designed for AI agents that read and write data as part of finishing a task rather than waiting on a human.

Databricks isn't the only company betting that AI agents, not humans, will be the ones provisioning databases. Supabase hit a $10.5 billion valuation in June on the same wager, after its CEO said more than 60 percent of new databases on the platform are now started by an AI coding tool.

The company also said the capital will fund acquisitions. It has a recent one to point to: in June, Databricks bought Panther Labs, a cybersecurity startup, for an undisclosed price, extending the platform into security monitoring for AI systems. Benzinga also reported that Databricks SQL, the company's data-warehousing product, has reached $1.5 billion in annual recurring revenue, growing more than 100 percent year over year.

Not the Same Story as the Debt

Databricks has raised money in more than one currency this year, and it is worth separating the two. In January, the company secured $1.8 billion in debt financing, arranged by a group of institutional investors and reported at the time by Bloomberg and CNBC. That is not equity. It does not touch the company's valuation.

The debt arrived in two pieces, according to SiliconANGLE's reporting on the Bloomberg and CNBC coverage: most of it through an extension of an existing revolving credit facility, the rest through a delayed-draw term loan.

Combined with prior facilities, it gave Databricks access to as much as $7.05 billion in loans to draw down as needed, mainly to fund acquisitions and cover rising inference costs rather than to fund new products outright.

The debt financing followed the same pattern as the equity: it landed a few weeks after Databricks first announced its Series L round in December, while the company was still finalizing that raise. Loading up on debt and equity in the same season is not unusual for a company spending heavily on AI acquisitions and compute. Mistral borrowed $830 million earlier this year to buy Nvidia chips for its Paris data center, using the same logic on a smaller scale.

The Skeptics' Case

Not everyone treats a private valuation as gospel, and Databricks' own peer group makes the skeptics' case for them. Cursor doubled its valuation in five months earlier this year, landing at $50 billion in a round investors reportedly oversubscribed within days.

Anthropic's math is stranger. Its Series G priced the company at $380 billion in February, and within weeks, investors on private secondary markets were offering close to 900 billion dollars for the same shares, a gap LDS covered in detail.

Set against that backdrop, a 40 percent bump for Databricks looks almost conservative. But the harder question is why a 13-year-old company with a reported $5.4 billion revenue run rate keeps raising private money instead of filing for an IPO. Ghodsi's own signals point in different directions. He told Bloomberg Television on June 4 that 2026 is "a terrible year" to go public. He has also told investors an IPO remains possible as early as 2027, according to The Information's reporting cited by LDS in June.

TechCrunch found a lighter way to make a similar point: the pace of Databricks' fundraising has become an inside joke. One investor posted online, "Turning on alerts for when we get a Series AA," a nod to the company running through the alphabet of funding rounds faster than it seems willing to ring the opening bell on a stock exchange.

The Bottom Line

Strip away the multi-AI strategy language and the Postgres branding, and the raw situation is simple. Databricks signed a piece of paper on July 16 that says it is worth $188 billion. Nobody has wired the money yet.

That number is 40 percent higher than what Databricks was worth five months ago, and it arrived faster and larger than the range investors were floating just weeks earlier. Coatue, already an investor in the February round, is now leading this one. Databricks says the cash goes toward tools that help enterprises stop overspending on AI, a pitch that only works if enough enterprises are, in fact, overspending.

The open question is not whether $188 billion is defensible. Private valuations are opinions until a public market prices the stock, and Databricks has spent thirteen years avoiding that test. The open question is what happens if the answer, whenever it finally comes, is smaller than the number on this term sheet.

Ghodsi says Databricks is playing the long game on value, not tokens. The market has not yet asked him to prove it.

Sources

Practice interview problems based on real data

1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.

Try 250 free problems