From Open Source To Proprietary License? RedMonk Data Shows Minor Impact
Does abandoning open source licenses increase revenue? The developer-focused analyst firm RedMonk wanted to find out…
We get asked a lot if we have any data on the financial impact of license changes.
The public data available is limited, but here’s a look at how at least some companies have fared pre- and post-license change. https://t.co/GBzj2oERAj https://t.co/LWg8MTqFFJ pic.twitter.com/0nE4kIHuQj
— Rachel Stephens (@rstephensme) August 26, 2024
Senior analyst Rachel Stephens explained her methodology in an interview with The New Stack: focusing the analysis “on public companies that changed software licenses to see if there were notable data patterns, as that was the problem space I was specifically curious about.” Using public companies ensures clearly-defined “financial metrics”, so Stephens checked the revenue, net income, and market capitalization at:
- MongoDB
- Elastic
- HashiCorp
- Confluent
The New Stack reached out to all four companies, but only two of them responded with a comment. But beyond that, Stephens’ analysis served as a touchstone for interesting conversations on social media involving other analysts and various tech-industry figures about open source licensing.
Conventional wisdom has it that companies sacrifice goodwill (and hence customers and profits) by abandoning their open source licenses.
But would Stephens’ analysis validate that perception?
The Results
Stephens first admits the possibility of other “confounding variables” — like everything else that impacts the companies’ financial results besides licensing. (For example, the state of the broader market…) And of course: “We don’t have data revealing what might have happened if the companies hadn’t made the change.”
“The limitation of this analysis is that it had to be confined to companies with public financial statements,” Stephens emphasized in our interview. “But that’s also a beauty of this analysis: all of this was constructed with publicly available data, so anyone who wants to broaden the scope to other companies can easily use this as a starting point to add in their own points of comparison.”
So, with all necessary caveats in place, Stephens first charted their revenue growth from 2017 through the first quarter of 2024 — and then drew a line marking the moment they’d changed their license. The results were remarkably consistent — and remarkably unspectacular. Stephens’ analysis acknowledged that “revenue increases after the license change, but the rate of increase is not materially different than the rate of increase prior to the license change…. we don’t see a notable change in the rate of growth.”
There does also appears to be a slight uptick in the rate of revenue change for MongoDB — albeit three years after the licensing change. (And MongoDB remains unprofitable.) Reached for comment, a MongoDB spokesperson started by saying “It’s important to acknowledge the ease with which large cloud vendors could historically capture the value of open source products without contributing to the open source community.
“We’ve seen continued success under our current license because it effectively addresses this balance.”
Indeed, when it comes to market capitalization, “MongoDB experienced tremendous growth in company value post-license change,” Stephens’ analysis conceded. But then looking at the other three companies, “We also see very mixed results in company valuation, and there does not seem to be a clear link between moving from an open source to proprietary license and increasing the company’s value.”
Graphing the change over time, Stephens found that HashiCorp and Confluent have both been below its IPO valuation since sometime in 2022, while Elastic has been above — although only once peaking near 250% of its initial valuation (before a sharp drop which then lasted nearly two years…)
As Stephens sees it, “Elastic grew, but at a much more moderate rate” than MongoDB — and in those same years, “HashiCorp has struggled and seen [its] valuations drop post-license change.”
And net income? Negative — for all four of them. (“This means that valuation is primarily driven by expected growth in future cash flow,” Stephens pointed out.) But again graphing each company’s numbers (going back as far as 2017, or as far back as they’re available), Stephens finds no spike — or even much of an increase — in their (negative) net income figures.
Reactions
I asked Stephens if she thought the analysis would have an impact in the future on companies considering moves to proprietary licensing. “I doubt it,” Stephens replied. “License changes are strategically important decisions that no company undertakes lightly.
“Decisions like that should be driven by much richer data sets than what you’ll find in a small sample size of publicly available financial data.”
But it prompted some interesting reactions online…
Chef’s co-founder, Adam Jacobs, had a funny response. “What I think changes most for these companies is that they stop saying ‘lost to ourselves’ in sales review meetings.”
Rachel might not agree with me, but what I think changes most for these companies is that they stop saying “lost to ourselves” in sales review meetings. https://t.co/rucWdmMhEv
— Adam Jacob (@adamhjk) August 26, 2024
And Phillip Carter, the lead for AI-related observability efforts at Honeycomb.io, shared a similar thought about companies moving to proprietary licenses.
“In my opinion, a lot of this is also just to placate sales leaders who mistake their own skill issues for tech that’s ‘out of control because it’s open’.”
imo a lot of this is also just to placate sales leaders who mistake their own skill issues for tech that’s “out of control because it’s open”
— Phillip Carter (@_cartermp) August 26, 2024
Stephens’ findings drew applause from Joseph Jacks, founder of the commercial open source VC firm, OSS Capital. “100% spot on!” Jacks posted. “I have been saying the exact same thing for years now.”
And more tech-industry figures reacted on LinkedIn, including Rich Burroughs, host/creator of the Kube Cuddle Podcast. “I would argue that if the positive impact isn’t there, that the cost of alienating a bunch of community members is definitely too high.”
Taylor Dolezal, head of ecosystem at the Cloud Native Computing Community, even argued on LinkedIn that “This data reinforces putting in the effort and doing the difficult thing.”
Where’s the Profit?
Stephens’ analysis adds that companies moving to proprietary licenses “is largely attributable to pressure from investors.” But after Stephens’ analysis, MongoDB released new financial results at the end of August — still showing the company reporting a net loss of $54.5 million, and in fact a net less that’s larger than one year ago.
Stephens’ analysis drew a thoughtful response on LinkedIn, from Josh Lee, developer advocate at the real-time analytics provider Altinity. “If the provider is going to be unprofitable either way, I’d definitely prefer the technology be open.”
Ironically, on Aug. 29, Elasticsearch announced it was adding an OSI-approved license for Elasticsearch. I asked Stephens if this lack of better financial outcomes might have influenced Elastic’s decision, and she started by saying “Financials certainly may have been a part of it.
“That said, the timing between Elasticsearch deciding to switch back to an OSI-approved license falling so close to AWS’ decision to move the OpenSearch project under the Linux Foundation feels unlikely to be purely coincidental.
“That said, I have no insights into the internal motivations that drove Elasticsearch’s decisions.”
Reached for comment, Elastic’s senior director of global PR Alexia Russell pointed us to a new line in their FAQ “to understand Elastic’s motivations and successes regarding our license changes” — which in turn links to their announcement that “Elasticsearch is Open Source, Again.”
Yet MongoDB’s executive communications specialist would say much the same thing. He first stressed that their company “believes an open development approach leads to more valuable, robust, and secure software, and it directly enables a stronger community and better products.” So despite licensing that targets competition from large cloud vendors, moving from an AGPL to an SSPL license “allows us to maintain the same freedoms the community has always enjoyed—they remain free to use, review, modify, and redistribute the source code.
“The only changes from AGPL are additional terms that make explicit the conditions for offering a publicly available MongoDB as a service. For the vast majority of our developers and customers, nothing has changed.” (They also insist that “The continued satisfaction of over 50,000 customers and the millions of developers using MongoDB are a testament to the innovative value that MongoDB continues to offer.”)
But in a more skeptical response, TNS analyst Lawrence Hecht summarized Stephens’ findings as “these companies are nowhere closer to being profitable than before.” So for those companies, “This is a problem… The assumption has been that closing a company’s license will allow the companies to increase their margins among their existing customers.”
And Hecht notes that the absence of new money is a problem specifically in areas where “the percentage of companies using a given technology is not changing.” You can see that in statistics from elsewhere. “For example, in the Stack Overflow survey, the percentage of devs using MongoDB dropped slightly from 2018 to 2024 (26% vs 25%). In the same time period, Elasticsearch fell from 14% to 13%.”
Elastic’s late-August statement does say its SIEM security product “is one of the fastest growing in the market.” But as Hecht sees it, “It seems like none of these companies used their IPO money to increase the size of their potential market outside of current users of the technology. Instead, they try to find companies that are managing an open source app by themselves and want to offload that responsibility.”
So while these companies used their IPO money to build out their enterprise sales forces, “I don’t think they will have a lot of success upselling services to these users. A key point to consider is how price-sensitive users are. It is the top criteria for choosing a commodity product, which by definition has low margins.”
Some conventional wisdom has suggested proprietary re-licensing doesn’t improve a businesses’ fortunes. But maybe we’re now seeing that backed by industry analysts. An earlier post in June from RedMonk industry analyst Stephen O’Grady examined the upcoming relicensing of CockroachDB, arguing that “licensing-led attempts” at protecting a business will “probably not” be successful. “If history is any guide…”
“It is a simple enough task to draft a license that says this potential competitor may not use a given piece of software. Ensuring that the new license is more benefit than cost is another matter entirely, however…
“[V]irtually all of the available evidence suggests first that concerns may be overblown, and second that even if it is assumed that they are taken at face value, that licensing mechanisms are not the solution many vendors want them to be.”
