Neo4j raises $325M in funding to advance graph database

Open source graph database vendor Neo4j said on Thursday it raised $325 million in a Series F round of funding earmarked toward building out the vendor's go-to-market efforts and data science and cloud technologies.

Neo4j, based in San Mateo, Calif., was founded in 2007 and helped pioneer the graph database market. A graph database enables users to connect data contextually, as opposed to the binary operations of a traditional relational database.

Neo4j's graph database platform has found wide adoption across different sectors including retail, where Walmart is a key customer. Neo4j has also had a strong focus on data science and its technology played a notable role in helping researchers during the COVID-19 pandemic.

The funding round was led by Eurazeo and GV (formerly Google Ventures) and gave Neo4j a valuation of more than $2 billion.

In this Q&A, Emil Eifrem, Neo4j's CEO and co-founder, outlines the opportunities and challenges of graph database technology.

Why are you now raising a Series F funding round to advance Neo4j's graph database efforts?

Emil Eifrem: The pandemic has been a real human tragedy. 

Emil Eifrem

For us over the past year, business has actually accelerated and we ended up coming out of 2020 doing very well and we had tons of cash in the bank.

I really wasn't planning on raising new funding, to be honest, but as an entrepreneur you have to be opportunistic. It just seemed like everyone with a pulse was raising money and so we figured at least it's worth kind of checking the market.

Since the beginning, Neo4j was never about fighting with a relational database. It was about fighting the concept that you put everything in the relational database, the one-size-fits-all database type of stuff. It's very clear at this point that graph databases are going to be one of the pillars of the future data landscape.

We have an opportunity to build something really big here and now it's time to accelerate.

Where do you see the intersection between graph databases and GraphQL?

Eifrem: I love GraphQL. I think it's an amazing technology stack and in my mind what it does is it allows the front-end developer to focus on building front-end code, like mobile and web applications, and then think of their application data as a graph.

It just seemed like everyone with a pulse was raising money and so we figured at least it's worth kind of checking the market.

For back-end developers, GraphQL allows them to expose what they've written at the back end, which typically includes business logic and managing data, exposing that very easily as a graph up to those front-end developers.

Now, how does that fit in with us? Because GraphQL looks at data as a graph, it fits hand in glove with a graph database, so it's really easy to take a graph database and expose that via GraphQL. That's how I see those two things, how they're separate and distinct, but fit really well together.

What do you see as the current challenges of graph database adoption?

Eifrem: I think there's still an awareness issue. People today know what a graph database is, not everyone, but most people do. But today, they underestimate just how broadly applicable it is.

At Neo4j, the breadth of use cases that we can handle today far outpaces where we were even a few years ago. So if you got educated about what graph databases were back in 2015 or 2017, you will have a much narrower conception of graph database use cases.

When you produce a horizontal product like a database, you don't foresee all the various use cases and where it's going to be deployed. You think of more fundamental things like the shape of data. Then all of a sudden, a pandemic hits, and people start doing COVID tracing with your application, as several governments around the world used Neo4j for COVID contact tracing.

Another use case that wasn't a use case for us until a few years ago is supply chain. Today any company that ships things is tapping into a global supply chain spanning continents. All of a sudden, you have a ship captain in the Suez Canal who falls asleep or whatever and blocks all traffic in the canal for a week. Then you need to figure out, well, 'How does that cascade across my entire supply chain, and how does that affect my customers?'

The only way you can do that is by looking 20 or 30 levels deep at the data relationships and that requires a graph database. So we've seen a surge in supply chain usage for Neo4j and it's a great example of how the world is becoming more connected.

If the world is becoming more connected, that means more connected data. This is why I just feel so confident that every year there's more and more need for our product in the world more and more use cases that ultimately have to use a graph database.

What is the role of open source for the Neo4j graph database?

Eifrem: I think we're going to be open source for as long as developers are an important constituency for us, which I think will be forever.

The goal here is to make sure that we're always choose-able for developers. We want to build up this massive community, of hundreds of thousands and ultimately millions of developers who are building applications with Neo4j.

The goal here is to enable low-friction adoption, so people can really easily get up and running with our products.

Editor's note: This interview was edited for clarity and conciseness.