A Walk Through HashiCorp’s History from 2012 until Today
Over the past decade, Terraform has become the standard platform for defining infrastructure-as-code (IaC) and making it part of the GitOps lifecycle. Hardware and software vendors make their products available through Terraform providers, as this will make them easier to consume for developers and platform engineers. Even AWS, Azure, and GCP maintain their own Terraform providers for easy consumption of most of their products. The Terraform Registry currently offers 4,137 providers for download, covering basically any relevant infrastructure technology. Based on this massive popularity you might think that HashiCorp is a highly successful enterprise, but you would be wrong, as the chart showing the company’s stock performance since the December 2021 IPO illustrates. You can find my analysis of the HashiCorp IPO right here.
Let us travel back in time and explain how HashiCorp, a company with significant achievements and a vast user base in cloud native infrastructure ended up with the above stock chart.
The Root of HashiCorp’s Financial Challenges
In 2012, Mitchel Hashimoto started HashiCorp with the vision of transforming infrastructure management and application delivery through innovative open source tools and without any concrete plans of monetizing his project. The screenshot from HashiCorp’s original website from December 2012 shows the company’s first product Vagrant, a tool to automate the creation of consistent and production-like development environments.
First Monetization Attempt in 2014: Atlas
Up until 2016, the company focused on serving SMBs with its Atlas platform, a paid hosted service that was licensed per node and included Terraform for resource provisioning, Consul for runtime management, Packer for image packaging, and Vagrant for creating development environments.
2017, the Year of HashiCorp Enterprise Products
In 2017, HashiCorp retired Atlas in favor of focusing on separate enterprise-grade product offerings, leaving its SMB roots behind. Terraform Enterprise, Consul Enterprise, Vault Enterprise, and Nomad Enterprise were the individual products. The key initial differentiators for Terraform Enterprise focused on private registries, policy enforcement, secure authentication, self-hosted operations, and premium support — features aimed at better enabling Terraform collaboration, compliance and integration within large enterprise environments. However, organizations often rolled their own enterprise solutions based on open source code and were not interested in the purchase of commercial licenses, while at the same time many major enterprise software vendors leveraged HashiCorp’s open source code to build their own commercial solutions on top of the open source Terraform ecosystem. SUSE, Dell, AWS, Google Cloud, Azure, IBM, and many others built custom integrations with open source Terraform to make it easier for customers to consume their respective solutions, but without HashiCorp getting paid. This story may (rightfully) remind you of why the original Docker company failed and was sold off for parts back in 2019 (my take from back in the day).
The HashiCorp IPO in December of 2021
In December 2021 HashiCorp had its IPO, but lost over 70% of its initial value by June of 2022 and was down by nearly 80% in late 2023. This was due to the very aggressive opening price that raised the company another $1,25 Billion in cash and valued HashiCorp at $15 Billion. HashiCorp was able to achieve these massive figures based on successfully moving most of its revenue to an annual subscription model and due to its 75% YoY revenue growth in 2021. While the company was able to increase its paying customer base by 80% from 2,500 in December of 2021 to 4,500 today, in April 2024, and increased its annual revenue from $212 million to $583 (+175%) million over the same time period, NetIncome in January, 2024 was still at -$191 Million. While these are strong numbers, the market expected more in terms of revenue growth and profitability, resulting in a significant loss in terms of share price.
A Black Day in August of 2023
August 10, 2023 was not a good day for HashiCorp and also not a good day for open source. This was the day when HashiCorp made the decision to unilaterally change the license of Terraform, their flagship open source product, from the Mozilla Public License (MPL) to the much more restrictive Business Source License (BSL). The open source community responded instantly by creating a fork from the still open source Terraform code. OpenTofu was born. The fact that the OpenTofu manifesto gained almost 35,000 stars on GitHub in only a couple of days, and the actual OpenTofu code repository has already gained half as many stars (20,142) only 8 months later is testament to the fact that the community is serious about not accepting an individual corporation to grab many years worth of community contributions and take them private.
The Manifesto: Liberating Terraform as a Building Block of the Modern Internet
“We believe that the essential building blocks of the modern Internet, such as Linux, Kubernetes, and Terraform need to be truly open source: that is the only way to ensure that we are building our industry on top of solid and predictable underpinnings.” — OpenTofu Manifesto
The statement in combination with the fact that the OpenTofu project is maintained directly by the Linux Foundation underlines the importance attributed by the community to Terraform. This explains the instant community outrage and the strong backing for OpenTofu from the start. Currently there are 161 software vendors listed on the OpenTofu website today (April 26, 2024) and GitHub shows 2,106 contributors to the OpenTofu repository so far.
Digging a Deeper Hole: The Cease and Desist Letter
HashiCorp’s cease-and-desist notice accusing OpenTofu of having used code from after the licensing change was in effect further strengthened the OpenTofu community, while isolating HashiCorp.
April 2024: IBM Received Great Strategic Value at a Reasonable Price
Picking up the company behind Terraform, the de-facto standard for defining cloud resources as code, for $6,4 billion (a revenue multiple of 11.5) was a good deal for IBM, for numerous reasons:
- Terraform and also Vault already are deeply engrained in IBM’s extensive product portfolio. Now Big Blue can double down on this integration by pairing Terraform and Ansible to consistently automate the resource infrastructure and application layer across its portfolio.
- If IBM rolls back the licensing change, restoring MPL for the Terraform code base, the company will become the ‘hero’ of the open source community. This should boost sales for the new HashiCorp Infrastructure Cloud and channel efforts and innovation from OpenTofu back to Terraform.
- Cloud native applications need cloud-independent infrastructure as code, as Kubernetes requires infrastructure automation, persistent storage, zero trust authentication, multi-cluster policies, app-centric operations management, app-centric networking, service mesh, and many other components that Terraform can deliver.
- Terraform brings the ability to define application infrastructure independently from the underlying cloud. The increasing number of organizations aiming to leaverage the advantages of different clouds needs a platform for cloud-independent resource definition and automation.
Now let’s grab some popcorn, lean back, and see the next chapter of the HashiCorp story unfold.
