I’ll say this. If this project moves away from open source, like this change does, I will reccomend that we migrate away to my team.
What? We haven’t been OSS in three years. This license update doesn’t change that at all.
If it stays on the current licence I will (just as I was doing) continue developing, open sourcing extensions, recommending to other people, and donating.
We’re not staying on the current license, for all the reasons in these comments. But you’re welcome to stay on previous versions of our software… either GPL or BSL. That’s part of the protections of those licenses.
I might suggest reviewing the new license first though. It helps to read read something before saying it doesn’t work.
I am even willing to be included in the paid tier, depending on pricing. I am all for paying people a fair wage. I am not for being lured into a (great!) tool and then having the licence changed.
Appreciate that, but you don’t need to pay us. Directus is free for ALL individuals and orgs under $5M/year + 25 headcount. We’re not changing that in our license… our focus is ensuring people actually follow the license by adding a key registration that can’t be stripped out.
I challenge the Directus team: if you want to “democratize data” like you state in big bold letters on your page, then be transparent. How much money are you making, and how much money do you need? And how much money do you want?
EDIT: Just two years ago the team was talking about “200% growth (3x) in annual recurring revenue”. How much money is does this project need to be sustainable?
Sure, let’s go. Let’s ballpark here…
Yes, years ago we had 200% growth. That means we 3x’d in a year. So that might mean we went from $1M to $3M in annual revenue. Cool. Exciting numbers.
Now let’s just do some simple (but fake math):
50 team members salaries (use whatever average you want)
Plus all the hardware/software we have to use
Plus 6-figures / month in infra
=============================
~$10M+ / year of expenses, just to sustain
The point is, you can have good growth, a great product, awesome clients/logos, etc. But for a product at our stage, you need to find ways to generate a massive amount of revenue just to keep the lights on.
How much money do we want? Enough to keep Directus existing as a modern data platform without selling out. That’s it.
Does something we’ve done scream corporate greed to you?
So if Monospace discovers that a company lied on the attestation, they have a signed document to use in a legal proceeding. Plus every self-hosted instance phones home with an organization name, contact email, and size attestation. So now Monospace has a directory of every organization using their software, what size they “say” they are, and how to contact them. So when a company that registered at under 25 headcount posts a job listing saying they have 26 employees, Directus enterprise sales team can reach out. When a company raises a Series B that gets announced on TechCrunch, Directus can check the registry and send an invoice.
Sure. Well, reach out to discuss… not send an invoice. But yes, what’s wrong with this? Not everyone has to register, but you do for certain usage.
Small teams and organizations will be disproportionately affected by this change, as icouto suggested.
How are they affected?
I work with some small government contractors who operate in environments with strict data governance and network security requirements. A self-hosted tool that phones home with organizational details to a third-party vendor is a non-starter in those environments. The whole point of self-hosting for these organizations is that nothing leaves their network.
Correct. The paid enterprise tier is for SCIM, Invoicing, and Offline Mode.
So if you need “offline mode” then you would need to pay for the enterprise version of the software when you move into production and need to air-gap.
Governments aren’t covered by our free Innovation Grant. But you can reach out if doing work for an NGO, non-profit, OSS project, etc.
Forced registration turns a self-hosted product into a product with a SaaS dependency, which is exactly what these users were trying to avoid by self-hosting in the first place.
A bit reductive to say that the only reason for self-hosting it to avoid software registration.
We have a free tier that will not require registration. And there will be a trial and grace period too.
If you want to use our full software for free, then yes, there is a requirement to attest you’re following the terms to get a free key… and basic product telemetry (the same as now).
The concern for me isn’t the thresholds, it’s predictability. As a small SaaS startup building on Directus, I’m not trying to avoid paying. When we reach the scale where a commercial license makes sense, we’ll pay it. Frustrating if some enterprises are currently not paying.
Awesome, thanks!
But the combination of a custom, unreviewed license and mandatory registration keys makes it harder to confidently build a long-term product on this foundation. Potentially having some features behind a paywall is scary too.
The license was reviewed by us and license attorneys.
The BSL had a community behind it. A bespoke license means the terms are whatever Monospace decides they are, now and in future revisions. That’s a different kind of risk than the financial thresholds themselves.
Curious, I know it helps to have so many people who have reviewed the license, but what does a community for a license do?
I said this several times, but being able to “change our license now or in the future”… that’s the same for every piece of software out there. The copyright holder can change license at any moment, so there’s no risk that’s any different here.
Most importantly: Publishing the actual MSCL text before asking for feedback would go a long way here. It’s hard to evaluate a structure without seeing the language. You’re just saying it’s ”better”.
Carwei and others building “small” SaaS products on Directus: Ben posted a detailed response on a Reddit thread about this announcement that included specifics on feature gating. Two items in particular that anyone building multi-tenant applications should be aware of:
* Custom RBAC filters will be gated to paid tiers. I’m sure you’re aware, but this is the mechanism you’d use to scope data per-tenant.
Nope. Please re-read this comment.
You get unlimited roles, policies, and custom RBAC on/off per collection in the free tier.
Custom RBAC Filters (defining more specific rules) is also free in the grant… but if you are above $5M/year and 25 headcount, then yes, you have to pay to unlock that.
…
Additionally, seats and collections will be capped in the free tier.
So they’re knee-capping any meaningful implementation of multi-tenant architecture.
Yes. Go read the comment above.
There’s a cap of a few HUNDRED collections per project in our free grant. If you have more than that, talk to us about a paid license.
Do you think that’s unfair of us?
These details were not in the original announcement. They were mentioned in a single sentence on Reddit, buried in a much longer post emphasizing how everything is free with the Innovation Grant.
I would link to the thread directly but I’m not sure if it’s against forum rules. Search Reddit for the announcement title and you’ll find it.
Nothing is being buried. This post is about the license and keys. NOT tiers, pricing, etc. You’ll have to be more patient. Not everything is locked in, which is why we were only looking for general thoughts on the license for now.
Ok, I don’t want to say that all here is bad, but what could be kind of a deal breaker is the pricing. This leaves me quite unsure right now if we are still able to count on Directus. Don’t get me wrong, I love Directus and what possibilities it creates. But I think it will be very hard to justify to most of our customers paying 1k/month for only the license + infrastructure.
So what I would like to know: will there be a pricing adaptation here as well? Because even the most discounted licenses will not work with the majority of our customer base.
Thanks Ben for taking the time to respond to so many questions, including several that were on my mind too.
For me personally, the past 24 hours since I discovered this thread have been a bit of a rollercoaster of speculation. I think the key issue was that it wasn’t 100% clear from the Reddit thread whether the Innovation Grant included feature gating or not. I read it as:
“If you make less than $5M per year, you can use Directus completely free”
“As for feature gating. Yeah, it’s a scary idea.”
“We plan on gating features that are very much “Enterprise” within paid tiers, such as: SCIM, Invoicing, Offline Mode, custom LLMs for our copilot, custom RBAC filters, and SSO. That’s it.” (which is an issue for us trying to set up a multi-tenant SaaS)
In the wall of text, I missed that you explicitly wrote:
BUT most importantly… All this is still free with our grant. (except paid support! obvi)
So again, if you want to use Directus as an individual or small org, you don’t need to worry about paying even for these features EVER, unless you’re above $5M/year or 25 headcount.
Missing that drove a lot of the anxiety for me (honestly, my anxiety probably made me miss it). You perhaps should have opened with that, with empasis on “all features” and not just “free” But I don’t blame you, I missed reading that part.
The confirmation that the grant is free from feature gating is the most important thing for me. That was what I needed to hear.
One suggestion for next time: publishing the MSCL text upfront (again, makes sense if you’re asking for feedback) alongside a simple flowchart showing who lands where — paid license, free tier (with feature gating), or Innovation Grant (without) — would have short-circuited a lot of the speculation. The structure makes sense once you understand it, but it wasn’t obvious from the initial announcement.
The risk that remains is future risk in you changing the license again (always a risk), or restricting the usage grant (new risk). The latter could cause massive headache since “just staying on the current version until we figure out a plan” won’t work in that situation, we don’t know how much of a headsup we’ll have in case of a usage grant change. Again, our goal as a startup is to be able to grow and pay for a license. Only talking about predictability here.
Our customers might fall into either category. The thing is, this uncertainty makes it quite hard to plan ahead. If the license cost turns out to be a deal breaker for some customers — for example those above the headcount threshold — we might need to consider switching to another system.
We’d want to avoid a situation where we have to maintain different backends depending on the customer’s license situation, especially when in most of our cases Directus is only responsible for a small part of their ecosystem.
As a smaller agency, that complexity adds up quickly. I just find it hard to justify the current license cost to certain customers.
Without getting into the discussion of whether software licenses match the OSS-ethos, I want to be very pragmatic: you are a company trying to provide a great product, and ensuring people pay for it and don’t steal. All good.
But please don’t say that “you’re not changing that in your license” if having a headcount for the free-tier is a completely new thing. Can you clarify (pardon if you have done so) who counts towards this number? If a company has 1000 people, but only 3 are using Directus, what is the number? If I use Directus as a member platform where 1 or 2 devs are working there, but 10k people are using it as a frontend for updating their personal details, what is the number?
Can you please clarify the different between the “free tier” and the “full software for free”, and when each applies (according to the scale metrics), please? I think that might be the main source of confusion around here?
No, not at all! But I wish businesses and software teams would appreciate that “growing” is a means, not an end. And if your expenses are greater than your revenue, you need to re-organize. In other words, if you get all of this new income from enterprises that stop stealing, but just spend it all so that expenses are higher than revenue again…
I mean “move away” as in “move further away”, not as in a binary thing
That’s very good to know
Not technically, but they can influence decisions that promote stupid strategy and short-term gains, then exit. You must know that that’s how it works in tons of VC-backed startups and scaleups, right?
And finally:
To “keep” existing and to be “sustainable” is opposite, antithetic to growing. You either exist sustainably or you grow. Not both. Monospace need to decide what you want. Look around you: how many companies have ruined a great product and a committed community because they “grew” out of bounds? Because they lost track of what makes them get up in the morning?
Crazy idea (I am truly not an expert): shouldn’t the cost in infra and a big chunk of salaries be more than covered by paying Cloud customers? If that’s not the case, isn’t something wrong? Wouldn’t that mean that Directus would not be financially sustainable even if it was closed-source and Cloud-only? The extra salaries/software/etc. that are mainly spent on self-hosted clients are still not covered by the paying customers? Do you have that much stealing?! Honest question! The world is crazy…
I appreciate the metaphor, but, as you know, software is not a consumable. When you write Directus, me using it does not prevent your paying customers from using it. A better example would be street artists:
If I’m a street artist, I need to get in “donations”/tips at least X amount for my performance. But I am not forced, at all, to prevent new people from watching for free, after that threshold is reached. One might even say that, after making, say, 100€/hour, a true artist would incentivise as many people to watch, whether they’re paying or not, since a true artist cares about spreading art, not making money.
Of course, for street artist (like open-source software companies), you never know what tomorrow holds. You can’t rely on the future that much. You can’t pretend that making 30€/hour (which sounds good in Portugal, for example) is enough, because maybe next week it rains and you have no revenue. So you need to adjust for uncertainty. But that doesn’t change the fact that, after a certain “sustainable threshold”, more people enjoying the fruits of your labour does not hurt you or your other customers, and, in fact, only helps (directly, by spreading the word, and indirectly, e.g. by making the economy more digital which will boost the demand for products like Directus in general).
OSS companies just need to find a strategy to ensure they guarantee the sustainable financial income threshold. If this was Monospace’s strategy, I hope it works! But it is genuinely sad to see that everyone must suffer because some corporations lie, and because Monospace couldn’t find another way to prevent the lies from happening.
But, honestly, I appreciate your and the rest of the team’s work, you have an amazing product here. And I can only imagine what challenges a OSS/source-available company faces in the current capitalist/technofeudalist socioeconomy. I just hope that you don’t fall into the traps that so many have fallen into.
I totally get that. I was trying to get some feedback on the license update (that feels big enough to warrant its own discussion), but it’s all so tied together, that I think that partitioning led to confusion… since y’all didn’t have the full picture: license + grant + tiers + pricing + discounts
I understand the concern, and we’re working to get pricing to a place where we can chat about that too.
Where we are now, is what I’ve posted elsewhere:
There’s a free tier that is meant for enterprise evaluation of the whole platform. There’s no time limit so you can really kick the tires and even develop your full system with no costs. But we’ve tried to find some limits that limit you from using it for big/prod use cases. Notably, it’s missing SSO, and has a low cap on users.
Then there’s the main tier. This is free under the grant, and paid for larger orgs. Everyone has different expectations on price… but this tier is roughly in the “hundreds of dollars / month” range (not $1k/mo). Beefy limits here are really just to avoid edge-cases. Notably, this tier is missing SCIM, Offline Mode, and annual invoicing. And again, this is where almost all users will go… and it’s 100% free for all individuals and small orgs.
Then we have our enterprise tier. It’s what you think… a price point based on scale. Basic support included, opt-out of all analytics, no limits, etc.
And it’s worth mentioning that for any paid tiers… we always work with non-profits, NGOs, OSS projects, educational institutions, etc to give discounts needed to make things work.
I hope this info helps… and stay tuned for an official announcement on all of this.
One more bit of extra context there as well: The BSL has a flexible additional usage grant meaning that every usage of the BSL license is different from the next, meaning that every individual one has to be reviewed separately. Us using a BSL w/ additional usage grant over a FCL derived license makes no meaningful practical difference for legal reasons as they both have to be reviewed with equal scrutiny.
For me personally, the past 24 hours since I discovered this thread have been a bit of a rollercoaster of speculation.
I get that… me too, haha. I wish we had waited to release ALL the info at once, but I made the call to start the license chat earlier so we had more time to actually discuss with the community. In hindsight, I wish we had framed the license alongside more insight into tiers/pricing. I (incorrectly) thought our community would assume we would still be doing the right thing… so a lil bummer to see so many people assume the worst in our intentions/plans. But it is what it is.
Missing that drove a lot of the anxiety for me (honestly, my anxiety probably made me miss it). You perhaps should have opened with that, with empasis on “all features” and not just “free” But I don’t blame you, I missed reading that part.
Yeah, totally my bad. I was too focused on the license part… and have been too close to all this, having worked on it for almost a year now.
Everything you would need to run a multitenant SaaS on Directus should be included in the grant. SSO, full custom RBAC+filters, etc.
The main thing is this:
We’ve learned that big companies will ignore important features if it means they can just use it all for free. We were hoping that SSO, Offline Mode, Premium Support, etc would be enough for large companies to pay for the software. But in reality, they just adapt and use free tiers.
That’s why we have had to really figure out this $5M + 25 HC threshold. It’s the only thing that we can use to force the larger companies to fund the project so we can keep it free for everyone else.
I think $5M is a good number… we’ve had it for 3 years, and it works. I understand why adding ANOTHER threshold is making people nervous, but it’s needed based on what we’ve seen. But, that’s exactly why we’re here discussing… as I think we might look at what that HC number is.
One suggestion for next time: publishing the MSCL text upfront (again, makes sense if you’re asking for feedback) alongside a simple flowchart showing who lands where — paid license, free tier (with feature gating), or Innovation Grant (without) — would have short-circuited a lot of the speculation. The structure makes sense once you understand it, but it wasn’t obvious from the initial announcement.
100% — I appreciate you saying that. I’ve posted a FCL/MSCL diff now, but I agree that a full view of the license and the tiers would have been better. We’ll adjust for this in our main release of info.
I guess this was also to get feedback on how we explain the change… so in that way, this discussion has helped a lot.
The risk that remains is future risk in you changing the license again (always a risk), or restricting the usage grant (new risk). The latter could cause massive headache since “just staying on the current version until we figure out a plan” won’t work in that situation, we don’t know how much of a headsup we’ll have in case of a usage grant change. Again, our goal as a startup is to be able to grow and pay for a license. Only talking about predictability here.
This is VERY valid. It’s frustrating that everyone is talking about how risky it is that we might change our license… because this is always true, for us and everyone else. But I hear you on the grant side. Let me talk with some folks on our side to see if there are any options to make this more predictable long-term.
Again, I sincerely appreciate the constructive chat. Thanks for being part of our community <3
I find the current licensing model really hard to justify from a practical standpoint.
In our experience, many customers end up choosing alternative systems - not because Directus isn’t great (it absolutely is), but because the annual licensing cost alone is simply too high. Even for companies with around $5M in funding per year, the jump in pricing feels disproportionate.
As a result, we’ve ended up using Directus primarily for smaller projects that fall well below the threshold. Which is a shame, because we genuinely like the product and would prefer to use it more broadly.
The core issue, in my opinion, is how pricing scales.
There are plenty of other open-source CMS options with paid models, but they typically scale in gradual, accessible steps. Paying $50 to $100 per month for a solid product is easy to justify. If usage increases, paying more also makes sense. But with Directus, the jump is from effectively nothing to $1k+ per month once a certain threshold is crossed - that’s a huge leap and a major barrier and deal breaker for most potential customers.
I understand that there are customers who are willing to pay at that level. But I’d argue that the majority of potential users - especially those running smaller or mid-sized projects - are not.
Another concern is tying the license cost to company funding or revenue. That model feels disconnected from actual product usage. In most software ecosystems, pricing is tied to how much you use the product - not how much money your company makes.
Think of tools like Adobe Creative Cloud, Microsoft 365, or GitHub: You don’t pay more because your company earns more - you pay more because you use more seats, storage or features.
A usage-based or seat-based model would feel much more natural, predictable and scalable - and furthermost more predictable for a potential customer. It would also lower the barrier for teams who want to adopt Directus more widely, instead of limiting it to edge cases or smaller projects.
Right now, the pricing structure makes it difficult to advocate for Directus internally, even when it’s technically the best fit - because other options currently are just way cheaper.
Appreciate the organized response… I’ll cut too it, since I have comment fatigue from the past 48 hours, haha
Can you clarify (pardon if you have done so) who counts towards this number? If a company has 1000 people, but only 3 are using Directus, what is the number? If I use Directus as a member platform where 1 or 2 devs are working there, but 10k people are using it as a frontend for updating their personal details, what is the number?
It’s the headcount of your legal entity. For obvious reasons, we can’t use “team/dept size”, as you can say whatever you want. I’m sorry if this comes off as insensitive, but I’ve had 3 years of HUGE multi-billion dollar companies using our software, and saying their “team” has an annual budget under $5M… so they should get it free.
Basically, every loophole you try to give the community will unfortunately get exploited by a much larger number of people. So the revenue and headcount numbers are for the legal org… but NOT for volunteers or their end customers or external contributors outside the org.
That said, we don’t charge a small project/team crazy money just because they are at a large company. That’s literally why were adding these tiers, haha. Those larger companies can use the free tier, or the main tier (hundreds / month)… or talk to sales and discuss any/all discounts.
Can you please clarify the different between the “free tier” and the “full software for free”, and when each applies (according to the scale metrics), please? I think that might be the main source of confusion around here?
Sure… check this comment I made out here:
No, not at all! But I wish businesses and software teams would appreciate that “growing” is a means, not an end. And if your expenses are greater than your revenue, you need to re-organize. In other words, if you get all of this new income from enterprises that stop stealing, but just spend it all so that expenses are higher than revenue again…
Valid. But I would hope that after 22 years of running the org the way we have, that we’d get a little grace form the community we’re building for and trying to protect from any/all costs.
The short answer is: we’re at a great place in terms of our team and expenses, so we don’t need a lot more internal growth. But we do need to generate more revenue than expenses (profitability), and until that happens, we are beholden to investors (not a good thing) and could risk running our of runway/cash.
So our main goal is just to cover our costs… and also have enough revenue padding to protect against AI or anything else that might kill paying projects/contracts.
Yes, I’d love to hire more ENG and go faster (for everything, I’m not talking ENT-only features)… but our focus is profitability so we can steer our own ship.
To “keep” existing and to be “sustainable” is opposite, antithetic to growing. You either exist sustainably or you grow. Not both. Monospace need to decide what you want. Look around you: how many companies have ruined a great product and a committed community because they “grew” out of bounds? Because they lost track of what makes them get up in the morning?
Welp, it depends on how you look at those words. Here’s how I use them:
Growth — growing our team as needed to handle our current/future workloads while avoiding individual burnout. And growing revenue/business to profitability.
Sustainable — the project continues to exist. Profitable against the team expenses needed to maintain an operable business.
Could we just be a smaller company and only focus on security patches? Sure, that works short-term, but we’d be gone in a year. We also have to keep the platform modern enough to maintain our current customers/revenue.
Crazy idea (I am truly not an expert): shouldn’t the cost in infra and a big chunk of salaries be more than covered by paying Cloud customers? If that’s not the case, isn’t something wrong? Wouldn’t that mean that Directus would not be financially sustainable even if it was closed-source and Cloud-only? The extra salaries/software/etc. that are mainly spent on self-hosted clients are still not covered by the paying customers? Do you have that much stealing?! Honest question! The world is crazy…
Without getting into it all too deeply, here’s how it works:
OSS project maintained by Rijk and Ben ($0 revenue), we can only invest free time, since we need jobs to pay the bills
Get VC funding, gives us millions of dollars for a core team, start a SaaS, etc. ($0 revenue, BIG expenses).
Try to grow revenue quickly to cover our expenses before the VC money runs out (runway)… get more VC funds if you have an opportunity to grow (good) or are running low on cash (bad). Repeat step 2+3 as needed, until the project fails, or…
We become profitable. Now revenue is greater than expenses and things are sustainable.
We’re almost at step 4, but also have opportunities to grow (in a good way) and are looking at what makes the most sense.
In short, cloud revenue does not magically cover our team/infra costs. We have 50 people and a VERY big AWS bill. We’re doing awesome, but are still growing towards profitability. Maybe because Rijk and I are so passionate about giving things away for free as much as possible, haha
I mean “move away” as in “move further away”, not as in a binary thing
Well, sure… haha, but I don’t think BSL → MSCL is a big leap here. Things are being restructured and clarified. I’m not trying to minimize the changes, but we’re keeping the grant and oss conversion and source available and ability fork, etc… while making some structural changes to close exploited loopholes.
Not technically, but they can influence decisions that promote stupid strategy and short-term gains, then exit. You must know that that’s how it works in tons of VC-backed startups and scaleups, right?
Haha, welp, of course! The investors, team, community, customers, market, and everyone in between can INFLUENCE decisions. That’s what I call “feedback”… just like this.
I should mention that our investors are amazing, well aware of the OSS/PLG approach (big investors in key OSS orgs), and have never pressured us to make bad decisions for short-term gains. In fact, they were the FIRST to question my decision to move from GPL to BSL 3 years ago.
More importantly, when it really comes down to it, I have made sure we’re protected from these risks (or any enshitification) as long as I’m CEO and on the board.
I appreciate the metaphor, but, as you know, software is not a consumable. When you write Directus, me using it does not prevent your paying customers from using it. A better example would be street artists: […]
Love this (and all metaphors). I’ve already kinda answered this one above, but will say that OSS is indeed like a street artist… and we’ve lived that life for over a decade with Directus.
But donations rarely work, unfortunately. We were not able to find a path to profitability/sustainability under that model. Which is why we updated to BSL.
OSS companies just need to find a strategy to ensure they guarantee the sustainable financial income threshold. If this was Monospace’s strategy, I hope it works! But it is genuinely sad to see that everyone must suffer because some corporations lie, and because Monospace couldn’t find another way to prevent the lies from happening.
Thanks! I hope this works too. And to be clear, it’s not JUST about companies lying… it’s about any ways to generate recurring revenue to be profitable. If we set our grant to $20M and 200 headcount, we could lose enough of our current paying customers to this free option where we can’t get to profitability.
Anywho… I guess a lot of this is about trust. I hope that you and the rest of our community have at least a little trust in how we’re operating. And with that trust, we will continue to offer as MUCH as we can for free, as permissively as possible, and without the assumed corporate greed once we’re profitable.
Appreciate the chat! And pardon MY wall of text… <3
In our experience, many customers end up choosing alternative systems - not because Directus isn’t great (it absolutely is), but because the annual licensing cost alone is simply too high. Even for companies with around $5M in funding per year, the jump in pricing feels disproportionate.
Yes, that’s a big reason for this boarder update.
We’ve introduced a free tier (beyond the grant) that gives bigger projects/teams the ability to use Directus for free (mostly eval and dev).
And we’ve added a tier between Free and Enterprise to have a lower price point than before (think: a few hundred dollars a month, instead of $1k+).
Another concern is tying the license cost to company funding or revenue. That model feels disconnected from actual product usage. In most software ecosystems, pricing is tied to how much you use the product - not how much money your company makes.
Yes, that’s pretty common. However, we felt strongly that the majority of our users should get the software completely free and with very few barriers to entry. So instead of trying to charge everyone $20-100/month (annoying for individuals, and way too cheap for big companies)… we opted for this threshold approach.
Also, usage-based models CAN work, but aren’t a silver bullet. If we charge on Collections, that might lead to bad schemas to save money and feels icky to devs/us. We tried it for years, it didn’t work.
Charge on API requests/bandwidth? Hmm, that feel WRONG when people are self-hosting. And predicting that usage can be REALLY tough. We tried it for years, it didn’t work.
Charging on seats? Cool, but some HUGE mission critical projects just keep a few admins in the studio and pay almost nothing (not to mention how AI might change “seats”).
Also, that doesn’t mean just because you pass $5M or 25HC that costs should be crazy. We still want to scale fees to value… which is part of this pricing change.
It would also lower the barrier for teams who want to adopt Directus more widely, instead of limiting it to edge cases or smaller projects.
But again, double edged sword… usage-based might be way cheaper for enterprises, but even a $5/month usage cost could stop a HUGE amount of individuals from using the software. The credit card alone is a BIG blocker, regardless of cost.
Right now, the pricing structure makes it difficult to advocate for Directus internally, even when it’s technically the best fit - because other options currently are just way cheaper.
I get that, and I’m glad there are options out there. We’re just working towards a balanced system that will work for MOST of our users. Also, I think you’d be surprised by how much we give away for free… so I don’t know about cheaper options unless someone is paying you to use their software, haha
We’ve introduced a free tier (beyond the grant) that gives bigger projects/teams the ability to use Directus for free (mostly eval and dev).
this is a strawman - evaluation and developement for free is not a “tier” that is sensible for a decision beforehand - if the final product would be too expensive to run, why even invest time and money in the developing a system? this is what is happening right now for us: we simply go a different path
And we’ve added a tier between Free and Enterprise to have a lower price point than before (think: a few hundred dollars a month, instead of $1k+).
this could be interresting: it depends on the features it has - the current dealbreaker for the cheaper options (especially the cloud version) is first privacy and compliance issues from the customer and second, that the cloud versions can’t use custom endpoints
just bind it to sensible features, that can be added (modular) at a cost - i’m not talking about 5 $ per month tiers, i’m talking about the current huge gab between 0 and 1k+ per month - if there would be a sensible tier in the middle, that is actually useful, this might be the solution
Also, I think you’d be surprised by how much we give away for free…
You would be surprised by how much money you are throwing away because you have no valid options for customers that are willing to pay a sensible amount of money but can’t and then decide to use a different product
as i said: in virtually any project we use directus with, the customers pay zero money, because they fall way below the treshold - and in the past 3 years were not able to make projects with directus happen for customers that are above this threshold, with a more sensible pricing structure, you would have a great potential for a lot of smaller customers
Thanks Ben, really appreciate the thorough response. Great to know we’re covered for our use case. We’ll keep building, and look forward to crossing the 25 headcount line eventually I was literally in the Directus+ program just so I could chip in a bit. Happy to be able to continue to advocate for Directus whenever it’s possible.
I think the headcount limit should be increased. Most startups today cross 25 employees fairly early, especially after initial funding, but that doesn’t necessarily mean they are financially stable. Many of these teams are still in a growth and experimentation phase, and taking on new recurring costs at that stage can feel risky.
I’m also concerned that this could discourage startups from using Directus altogether. Many teams know that crossing 25 employees is relatively easy, and during their struggling period they may avoid adopting tools that could later introduce licensing costs or compliance concerns. This could unintentionally reduce adoption among startups, which are often key contributors to community growth and long-term ecosystem success.
Additionally, headcount doesn’t necessarily reflect usage. In a company with 25+ employees, only a small subset of teams may actually use Directus. There are multiple teams — frontend, backend, mobile, design, product, marketing, etc. — and not everyone interacts with the CMS directly. Charging based purely on company size rather than actual value received may feel disproportionate.
There’s also a broader ecosystem consideration. If CMS tools, frontend frameworks, and other core infrastructure tools start charging at very small company sizes, it could create friction for startups trying to grow. The open-source ecosystem thrives because many foundational tools — like Node.js packages and npm libraries — are freely accessible, even though they require significant effort to build and maintain. That accessibility is what enables innovation and growth.
To be clear, I’m not against monetization — everyone deserves to earn and sustain their work. But to earn, companies should also be given a fair opportunity to adopt, grow, and derive value first.
Instead, I think lowering the revenue threshold from $5 million to around $1 million could be a more balanced approach. Companies generating ~$1M revenue are typically in a better position to contribute financially, while still keeping Directus accessible to early-stage startups.
Alternatively, if the revenue threshold remains the same, increasing the headcount limit to at least 75 employees would better reflect how startups scale today.
Another approach worth considering is a time-based model. For example, allowing companies to use Directus freely for the first 1–2 years, and then applying licensing requirements afterward. This would allow teams to experiment, adopt, and validate Directus internally before committing financially. Many companies follow similar approaches by offering generous startup-friendly periods to encourage adoption and long-term loyalty.
Ultimately, I believe a slightly more startup-friendly approach would help maintain strong adoption, grow the community, and still ensure that financially capable companies contribute back over time.
From my experience, startups with 25 people have funding (as you said). Not financially stable? Maybe not. But it should be included in their budget when funding. You can’t expect a one-size-fits-all solution to be free, but they still said there will be some free tier (besides the grant) and pricing tiers that are not 1k USD per month.
75 headcount sounds silly to me. I’d rather have a stable grant than something that needs to be adjusted again. Predictability + sustainability. Fee for 1-2 first year is not stable, that sounds more like a very long trial. Or a loophole.
Ben already explained why income/budget is hard to enforce.
free 
But I don’t blame you, I missed reading that part.