If we have learned anything in 2021, it’s to bet on open source technologies - see funding rounds (Snyk…again!) and public market performance (MongoDB & Confluent!). We’ve also learned not to bet on anything with the prefix “Ever”.
Just last week we saw the Evergrande real estate crisis in China cause people worldwide to wonder how the failure of a company they’ve never heard of could roil global markets which harkens back to earlier this year when the Ever Given container ship got stuck in the Suez Canal stalling global supply chains. Kudos to Ever.AI renaming themselves to Paravision to avoid any bad halo effects.
This week we interview Naveen from NocoDB - the open-source company that turns any database into a smart spreadsheet. They are an open-source Airtable alternative that has amassed 18,000 GitHub stars and strong growth across the board.
Snyk, building the most trusted developer security platform, announced their $300M Series F co-led by Sands Capital Ventures and Tiger Global.
Opentrons, providing open-source solutions to lab automation, announced their $200M Series C led by SoftBank.
ClickHouse, makers of a database management system for real-time analytics with SQL, announced their $50M Series A led by Index Ventures and Benchmark.
Jscrambler, protecting the source code of web and mobile apps from companies around the world, announced their $15M Series A led by Ace Capital Partners.
Exein, the first open source IOT Framework, announced their €6M Series A led by Future Industry Ventures and eCAPITAL Entrepreneurial Partners.
Loft, enabling platform teams in large enterprises to provide engineers with self-serivce Kubernetes access, announced their $4.6M Seed led by Fusion Fund.
Infracost, showing engineering teams how code changes will affect their cloud bills before production changes, announced their $2.2M Seed led by Sequoia, Y Combinator, SV Angel, and Yun-Fang Juan.
BattleSnake, building eSports for developers, announced their $1.5M Seed led by Madrona Venture Group.
To track the performance of COSS companies, we’ve created an equal-weighted index comprised of public names including: Kaltura, Couchbase, Confluent, MongoDB, Elastic, Talend (acq. by Thoma Bravo announced), Cloudera (acq. by KKR/CD&R announced), Rapid7, Fastly and Jfrog.
The COSS Index has been in positive territory over the last four weeks but continues to lag the major indices this year.
COSS Index +4%
S&P 500 +18%
The COSS Index flip-flopped with the NASDAQ over the last three years.
COSS Index +90%
S&P 500 +53%
COSS companies continue to ride large share gains from MongoDB and Confluent in jumping to nearly their highest level of the last five years. All three indices continue to trade significantly higher than their rolling five-year average.
COSS Index: Current Multiple 19.0x | Five-Year Mean: 8.6x
Emerging Cloud Index: Current Multiple 14.2x | Five-Year Mean: 9.7x
NASDAQ Composite: Current Multiple 4.4x | Five-Year Mean: 3.3x
Interview with Naveen, the Founder of NocoDB.
What’s your background?
My background is in computing primarily. I have done my Masters in mobile and computing at Lancaster University in UK. For the bulk of my career, I've been working close to hardware, close to operating systems, close to networking, and nothing was related to web per se. And I happened to come across this existing MySQL database on which I could not write REST APIs. Even the best API frameworks wouldn't fit in with an existing MySQL database. And I said, "Hey. This cannot be true. There's something wrong in this world if this is true, and the database should be accessible in an easy way through APIs."
So I think my first intuition was to set that right. I made a project back in 2017 and it caught a fair amount of traction. A lot of people started to use it in production. A lot of people started to blog about it. So the SEO kind of started to race for that data project. My first attempt to build an open-source product was last year where I open-sourced XgeneCloud, which was the evolution of Xmysql. And it created REST + GraphQL APIs for all the databases and it had a GUI element to it with a desktop application. So that was the precursor to NocoDB.
What is NocoDB?
NocoDB is a no-code database that makes the existing SQL databases available as spreadsheets. It also makes it a smart spreadsheet. So the idea is, Google spreadsheets or Airtable is very intuitive to use when compared to other applications. There is little to understand on how to use the spreadsheet products, because they are so ubiquitous. And so the idea we are bringing is why can't the smart spreadsheet exist in SQL databases? So that's the whole idea of NocoDB.
How is NocoDB being used today?
Currently, it's being used by developers, non-developers, business and operation people. But the use cases are so varied that I can't pinpoint and say, "Hey. This is exactly what they are working on," because NocoDB is not just giving spreadsheets on the database. It also gives you APIs, which have programmatic access to these databases. So it facilitates a wide variety of use cases. Most of the time, a developer sets up NocoDB, and brings in business and operations people on top of it. So that's how I see the use cases.
NocoDB has over 17,000 Stars on GitHub and 2.2 million downloads in 15 weeks. How did it become so popular?
To be honest, I could not have anticipated that this would happen. It was hard to predict, because we were just building the product. I think some of the things went our way, because there is very little to educate about the product to the people. Because when I say this is a smart spreadsheet on an existing MySQL or Postgres database, they intuitively understand what the product is and they want to try it. So there's no catch. The messaging is very simple. And whoever sees it easily understands it. And I think another thing that is going our way is we have made the setup very simple with very minimal steps to get started. Even if they don't have certain dependencies on their system, we default to certain backups, like a file-based database.
So we take care even if you don't have certain things to get started. And if you want to run it in production, we have a production setup. But to get started, we have made it seamless. So the bulk of the users get a taste of the product without having to go through multiple steps.
Where do you want to take NocoDB?
Honestly, we're keeping it simple. Why can't any end-user access databases just as Google spreadsheets? Why can't it be done? Our whole bet is to build a Google spreadsheet or a Airtable product on top of existing databases. And you should not feel, "Hey, I'm accessing a database," as a end user. Databases and spreadsheet have been existing for decades. And spreadsheets have been always accessible and plenty of people use them for all sorts of reasons. But databases are really powerful computing tools when it comes to building better applications. But they are very inaccessible with the developer tools that are out there. So our idea is to democratize this. You should not feel that friction when you go on your database. That's what we want to go into.
What advice would you give an open-source founder in terms of creating a successful project?
Most of the time, the first version of the product looks a little basic. Most open-source products in general look basic. So we tried to do a little more than a typical first version with our launch. We wanted to take it a couple of steps further and then launch it. I think that resonated well with our users. It was feature-rich and I think that helped. They said, "Hey. This can do this. Oh, this can do this." They weren't expecting many of the features to be there. But when I say this is an open-source Airtable, there are so many features missing as well. But the idea is we have pushed the product to be better than the first version that any user expects.
I would say the rest of them are open secrets on how to launch a good open-source product. Keep your UI really nice and beautiful, and keep the number of steps to get started minimal. Those are well-documented in open-source communities. We just followed them. But, we just pushed the boundaries a little bit on what users expect from a launch version of the product. And I think that helped.