Upstart graph analytics company TigerGraph today announced the completion of a Series C round of funding worth $105 million. The funding validates TigerGraph’s approach, according to the company, which says it will use the money to bolster the creation of a graph analytics ecosystem, including greater cloud adoption and expansion of AI and machine learning applications.
“This is a big milestone for the company,” TigerGraph founder and CEO Yu Xu tells Datanami. “I think people recognize the graph space has a huge potential. The market is huge. And particularly in this space, they choose TigerGraph as the best company in terms of technology, performance, and functionality.”
Xu says TigerGraph’s early focus on performance and scalable has positioned the company as a leader in the graph database and graph analytics space. Now it’s looking to build on that reputation by focusing on other areas, including AI and machine learning, visualization of graph data, and bolstering the graph ecosystem
The cloud will play a major role in everything TigerGraph does, says Xu, who led Teradata’s Hadoop and MapReduce efforts before founding TigerGraph in 2012.
“We’re going to double and triple down on graph in the cloud,” Xu says. “We’re going to support Google GCP soon. We’re already on Amazon and the Microsoft cloud. We’re going to add more cloud-first, cloud-native features.”
Link analytics, running atop graph database, can speed the discovery of patterns buried in data (image courtesy Tigergraph)
The combination of graph databases with machine learning and AI is a fruitful area that many TigerGraph customers are looking to exploit. The company says its approach to “deep link analytics” in its graph database can simplify the development of the data features that are so critical to machine learning systems.
For example, one of TigerGraph’s customers, China Mobile, has developed nearly 120 graph features for each of its subscribers’ phones, which number in the hundreds of millions. These graph features are fed into China Mobile’s machine learning model to help detect fraud and spammers. The more a given phone stays within a known “good network” of callers and avoids short calls, the less likely it is to be marked as a phone used for fraud or spam calls.
Up to this point, graph analytics has been primarily been the domain of large banks and other big institutions, like China Mobile. But the benefits of graph analytics are so compelling that mid-size companies with fewer resources are looking to exploit the technology. With the cloud providing compute resources, this is now possible, Xu says.
TigerGraph is looking to scale up its work with customers and other software vendors in the graph ecosystem to build next-gen graph analytics applications. “We’re going to invest a lot more in our partnerships,” Xu says. “In the last three years, we’ve proved we can solve big problems, high value problems, which were previously unsolvable….Bigger partners have noticed this.”
Todd Blaschka, TigerGraph’s COO, is looking forward to using the $105 million to expand not only the company’s reach and capability, but to show the world what graph databases are capable of. “We’re seeing graph really coming of age,” Blaschka says. “Where there’s data, there will be graph.”
The key is achieving this is to lower the barrier of entry to utilizing graph techniques, which have generally been the provision of large, well-funded companies. TigerGraph has the opportunity to simply the domain in a way that hasn’t occurred before.
TigerGraph made the leader’s segment of Forrester’s Wave for Graph Data Platform in November 2020
“It needs to be something where the C-level is not being caught up on graph, relational, NoSQL,” he says. “They just want to know, how can I get better insights and make better decisions faster with the data? How do I get there? So that’s where our vision and the acceleration is, to make that more mainstream.”
In 2020, TigerGraph launched its GraphStudio, which includes a drag-and-drop query builder that simplifies the process of developing graph queries. That was a key development for TigerGraph, and the company plans to build on that approach in 2021.
“We are bringing innovation to the BI space for graph databases,” Xu says. “This is really important, especially for the cloud. This will attract more first-time business users who are new to graph. They don’t need to learn any graph language. They just need to use the browser, drag and graph, to create a tree or small graph to ask questions. And everything is done visually and they get back results in real time.”
The company is also working with BI tool vendors to make it easier for analysts to use their existing visualization and analytic tools to tap into graph databases alongside their other data sources, including data warehouses built on relational database technology.
For example, TigerGraph developed a JDBC/ODBC connector that enables Tableau and Microsoft PowerBI customers can query the vertexes and edges in the TigerGraph database as regular relational database tables, Xu says.
“This makes sure that more customers can use the data stored in the graph database,” he says. “They don’t need to know any graph language or graph theory. They just need a familiar interface from Tableau or PowerBI or other BI companies.”
TigerGraph has something north of 100 paying customers. However, Blaschka notes that that number does not include the TigerGraph cloud, which has a lot of customers coming and going frequently. According to Blaschka, TigerGraph’s cloud business model looks a lot like MongoDB’s plan with Atlas, which has been a rousing success.
“We expect hyper growth in the number of users on TigerGraph Cloud,” he says. The comapny expects that number to “grow exponentially as cloud becomes even more and more prevalent.”
The Series C round of funding was led by Tiger Global and brings TigerGraph’s total funding raised to over $170 million.
TigerGraph Offers Free Graph Database for On-Prem Analysis
TigerGraph Cloud Goes Live as Firm Raises $32M
TigerGraph Emerges with Native Parallel Graph Database