An Open Letter: Atlassian's Forgotten-Middle re:The Discontinuation of Server Licenses

[#1]

October 19, 2020

Dear Mr. Cannon-Brookes and Mr. Farquhar:

I am writing to you in response to the announcement that Atlassian will discontinue selling sever licenses.

The largest of your customers will see no impact from this change, the smallest of your customers are nimble, and hungry and ready to change, that leaves the middle where migration efforts are highest, budgets are thinnest, and the most straightforward way forward, data center licensing, is priced out of reach.

The forgotten-middle deserve a better option. I implore you to reopen the discussion on a smaller user data center license tier.  I say "reopen the discussion" because I know that your actions, much like all actions taken at Atlassian, are carefully considered and debated. I ask you to re-consider the impact on the people that have invested trust in you, grew your reputation, and grew your company.  Remember the creative adventurous people who took risks by believing in your company values, and your products when they were unknown.
These are the forgotten middle and are some of your most loyal fans and advocates.

Atlassian software enables organizations to do more, it builds momentum in companies. Companies have their own north star, their own core competency and Atlassian software accelerates its progress. Migration for them to the cloud needs to be on their terms
and as part of their journey, it's not a matter of if, but when. Cloud is the future, and Atlassian is staying true to its values by boldly moving into the era of cloud. But please do not forget the middle and move on without them. They are so much of your history.

Allow the middle to continue to license the products they depend on at a reasonable price.  The forgotten-middle are not looking to remain in the past, they want to move to the future, they are the people who first took the risk on you and saw a better way. They want to move, but on their own schedule.  Reward them for their hard work and years of struggle who have allowed you to now take this momentous step forward.

As an admirer of Atlassian and its values and as a champion of the incredible products you produce, please allow us, the forgotten-middle to license what we already have and love at a reasonable price.  A company of 60 should not be forced to buy a 500 user license if they need to stay on premises for now.

Be kind with the great power you wield and offer an affordable data center license.

The middle care about all things Atlassian, they will continue to be your closest ally. Please take action so they will not be forgotten.

David Willson
Community Leader
Ottawa Ontario
Canada

[Michael Woffenden]

This also applies 100% to the "forgotten" very small businesses.  We need an option at the low end as we currently have with Server products.

[M Amine]

agree 

[Cameron Deatsch]

David, thank you for your post and your continued involvement in our community. As our customer base digests this latest news we are actively capturing feedback from everywhere and posts like this are read and discussed by many within Atlassian.

You are not alone in the ask for lower Data Center tiers and we will continue to gather data and feedback in this area. 

However, I visited your company's website and it appears that your organization provides cloud technology in the cybersecurity space. What keeps your org from adopting Atlassian's cloud? Is it apps? Is it security?

[Andrew Pane]

I will echo David's concern that the lack of a lower pricing tier for Data Center products really means Atlassian has decided to abandon small- and medium-sized businesses who made the apparent mistake of committing to the Atlassian product line.

In 2015 we started a business using Atlassian Server products for non-software workflows for clients in the energy sector, after attending that year's Atlassian conference where they were heavily pushing using their products outside of the software development space. We've built our own custom add-ons for integrating with energy sector applications, and for regulatory reasons these products *cannot* be housed in the Atlassian cloud, and furthermore they *must* be co-located with other systems they communicate with. 

Since then we've grown two clients from 10-user starter licenses to 250-user licenses, with three more growing along the same path, and numerous sales leads just getting started. Now I had to tell the two big customers they have to absorb a 50-60% increase in licensing fees just to stay where they are today, will have to tell the three smaller customers that we are sunsetting their products in 2024, and have no growth path to start new customers along this path.

("Plan B" is we spend the next three years attempting to re-write our software for Cloud with no guarantee of success and no additional revenue as a result, all while attempting to navigate our small business through the COVID-19 pandemic, all just to get back to where we were last Thursday!)

Please Atlassian, at minimum give us smaller pricing tiers for the DC products. Anything short of that is a massive betrayal of your existing customers.

[Alex Janes]

@Cameron Deatsch We are a non-profit that handles medical data. We are required to maintain HIPAA compliance. Even though a handful of users are in the application, we are now forced to go from $0 in licensing costs, to $20k/year. I can't justify that to the CFO.

EDIT: Just to clarify, I'm not looking for the free ride to continue. I get it. You have costs. But we don't always have the money or resources that these for profit orgs have. We have a team just large enough to justify the use of this type of workflow automation software. But not the resources to buy the software for $20k/year.

We just need a better pricing option, or we need to ditch the software NOW to prevent any future compliance/pricing headaches.

[Cameron Deatsch]

Alex, thanks for your feedback. We've received plenty of feedback related to our cloud and academic licensing and are working on programs to address this. Please stay tuned.

[M Amine]

@Cameron Deatsch We really want to migrate our clients to Atlassian Cloud but after a thorough analysis, it is not possible because Atlassian Cloud is missing a lot of features that can be found in Server/DC. Here are some features missing to name a few : 

• The User Experience cannot be customized as desired. In server we can use : Scriptrunner behaviors, Dynamic forms, SIL Live fields ...etc. Unfortunately, there is no Cloud app offering these features right now
• Postfunctions are all async in the Cloud. So the ordering of post functions is of no use. Besides, every issue update from the post function will not update the issue immediately. We have to hit the refresh button in order to see the update. Lately we had to demo this to a client and he didn't accepted and asked to migrate to Server
• Lately we had to use Insight AM Cloud and we have noticed that a lot of features are missing : Custom fields, User experience in the portal, Integration ...etc. 
• Data residency. We know there is a roadmap (https://www.atlassian.com/roadmap/cloud) for that, but a lot of things have to be done in order to be compatible with what clients ask for
• Domaine names are not customizable. Although there is a workaround for JSD users if they use refined app but Jira for agents will still use the *.atlassian.net
• ....

We really want our clients to be part of "Cloud first" strategy but they feel that Cloud is not really yet ready and they do not accept to go back or have regressions. 

[Ryan Carpenter]

Thank you @David Willson for making such an considerate appeal, giving voice to those of us feeling speechless. @Cameron Deatsch thanks for doing the thankless job of responding to our reactions.

I share @Michael Woffenden's sentiment. I am thinking of medical-device companies, about 80% of which are small. Very small. These are not government, education or non-profit companies, but a handful of innovators risking their retirement (in some cases) to introduce new devices in an uphill regulatory environment. They don't buy from Master Control, because it is cost prohibitive, while smaller competitors enjoy 45% growth in employee head count (or did before the pandemic).

Atlassian has an incredible value proposition among this audience, but only a small band of torch-bearers has known it, until recently. They have advanced the case on Atlassian's behalf, while Atlassian struggled to find the vocabulary to participate in the conversation. In addition to SoftComply (@Matteo Gubellini _SoftComply_), Matrix Requirements, Radbee, and prominent vendors like Comalatech, there are others off the public's radar that have gone about quietly attracting a new customer segment to Atlassian. Atlassian's appearance of being unaware or unconcerned with these medical and compliance businesses may have frightened some away, just when momentum had begun to build.

On a personal note: I am not a developer, but have used Atlassian software for more than a decade. The awesome starter license (Server) made it possible for me to build some basic server chops (if I am generous), explore more apps than I care to admit, and configure a client portal for a business that did not yet exist. I may not be a valuable source of revenue, but I have returned the goodwill in the form of demonstration, promotion, engagement in the community, interviews with Atlassian product managers, participation in local meetings, and so on. Along with your programs for education and non-profits, I hope you will include opportunities for sandbox creations.

[Normann P_ Nielsen _Netic_]

I agree - both as a vendor, partner and end-user, both private and companywise.

[Jörg Godau]

A powerful statement and one we can sign in to.

@Cameron Deatsch Atlassian chooses to lose all the smaller and mid sized customers who are unable to go to cloud or unwilling to pay for Data Center, how much of that loss can you absorb? And who will we grow into new big customers?

Follow the logic to EOL for DC:

Without server to get them started on the Atlassian journey, I don't see many organisations starting on Data Center. Those that start on cloud don't need DC. 

This logically means only large existing customers can or will be on Data Center.

This means few customers for a large cost (on Atlassian side). That will justify price increases and that will justify companies leaving DC.

Result very few customers, High price, high costs. Business/Share holder value is low, so DC will also be EOL.

My prediction 3 to 5 years max before this is announced.

Prove me wrong Atlassian please! I love your products, your ethos, your spirit and drive. Keep me?

[Kenneth Juul Wannebo]

Edit

[Jeff Tillett]

I'll throw in my name to led to the "terrible decision" camp. I have loads of interviews with various teams around Atlassian where I have given all of my reasons. Atlassian has clearly told me that I don't matter. 

The icing on the cake is that one of the big concerns I raised was the certified admin community who had built their careers on Atlassian, and invested their own time and money into doing it. It seems that the response to my feedback was to nullify all the certs for everyone globally. Way to live up to your values of not effing the customers. 

Thanks to these two announcements, I now know where Atlassian's loyalty is, and it's not to their customers, it's to their shareholders. Even though I am a Data Center customer, and can afford the price hikes, I'll be looking to remove any trace of Atlassian from my company. 

[Taylor Huston]

My thoughts on the matter:

https://medium.com/@taylorhuston/atlassian-is-killing-their-server-line-fe3921ca905

[AndersS]

My 2-cents

As an owner of a mid-sized business, I agree witht this post - this is a terrible decission by Atlassian.

We use our Jira and confluence servers for many purposes, including retaining information aout our employees and customers (and other sensitive information).

Releasing this to the "cloud" is not an option for us, both due to GDPR and due to our customers.

And migrating to a data-center is way to expensive, for our current 50-user licenser tier.

I deaply implore Atlassian to reconsider this, as the only way for us, that I can see, is migrating to a different product.

[Normann P_ Nielsen _Netic_]

I think this is somewhat "terrible" - although I do understand it from a business perspective.

But why does nobody look into the Cloud and Cloud Connect and the distributed way that Add-Ons Works - and the data are moved outside Atlassian Cloud (I guess), No SLA exists etc etc

Ref: https://community.atlassian.com/t5/Jira-Core-questions/SLA-for-Addons-in-JIRA-Cloud/qaq-p/1332913

I think that is a HUGE problem for those customers that may go towards the cloud.

BR,

Normann

[David]

I agree with the sentiment or all the comments here.

We're increasing leveraging Jira etc in our business for departments outside of IT, with theses departments we've been using Jira Cloud with its many limitations.

However our IT helpdesk is ran on Jira server for numerous reasons such as product integration or leveraging Apps that only work on Servers (such as your own Insight Discovery).

Cloud currently doesn't offer anywhere near the same flexibility or customization as on Prem with many Apps never to likely work.

For Data center to work for us as a business I would have to migrate hundreds of users from Cloud to Data center to even start to justify the cost all with the potential threat of not knowing how long will Data Center be around. 

I can only sympathise for even smaller companies losing out from the discontinuation of Server, I foresee a lot of unsupported, unpatched Servers remaining until they move to another platform.

[Thomas Prescott]

We've been an Atlassian shop at my workplace for a long time (about 10 years now).  I've personally championed an increase in usage of other products Atlassian's server line of products, including Jira Software, Jira Service Desk, Bitbucket Server, Confluence, Crowd, etc.

We've spent a lot of time and money on customizing our usage of Jira, using software and features that simply aren't available in your SaaS offerings.  We also had enjoyed having complete control over where our data is located, how it's secured, who we know can access that data, and most importantly we have control over when the applications are patched so we can provide documentation to our end users and also minimize disruption to our users.  Our company simply isn't big enough to justify the expense of Data Center the way it's currently priced.

If this continues as planned, I will have lost all faith in Atlassian.  This decision may force us into spending time and money again in finding and implementing a competing product.  The SaaS offering really isn't going to be a good choice for us.

It really feels like Atlassian has forgotten about their middle-sized customers that were loyal to them for many years.  I really feel like the trust I've placed in Atlassian was misplaced.

Please don't forget about the customers that gambled on you, that took a chance on you.  Without them, you'd still be at the bottom of a google search result.  You have some incredible on-prem products in an era where everyone wants to be in the cloud.  If the issue is the fact you want a subscription instead of perpetual licensing, then switch the server-teir to a subscription model, or provide better terms for your data center products for smaller installations.  There are more options than just leaving hundreds or even thousands of mid-sized organizations in a lurch for no reason other than to boost revenue.

[SK]

As a sysadmin I've been responsible for maintaining company Jira and Confluence instances since 2013. So I remember when maintenance for 500 user Confluence was $2000 and 500 user Jira + Agile was $3000. Comparing these old prices to the new price list boggles the mind. The improvements the products received in that time-frame nowhere near justify the increase, which have mostly been UI facelifts and quality of life improvements, not life changing new features.

Despite this, I too, have championed the use of Atlassian products on many occasions. This won't likely be the case anymore. Either because of cost, because cloud is not an option, or simply because the trust has been broken one time too many. There is zero reason to believe that prices won't continue to increase, especially now that it became abundantly clear the strategy is to milk fewer enterprise customers more. The old company motto is long dead. I can only urge other customers considering the cloud offering to rethink if they really want to willingly give their data to be held hostage by such a company.

I'll leave links to two discussion threads in relevant communities, with many accounts of user impact and eloquent reactions:
https://news.ycombinator.com/item?id=24805748
https://www.reddit.com/r/sysadmin/comments/jcd1v1/atlassian_will_stop_selling_and_supporting_server/

[h_mueller_imp-ag_de]

As a small company with fewer then 50 people your decision to discontinue server would force us to leave your product line. Datacenter with 500 users is to expensive for us, cloud premium is not legal in the European Union and cloud enterprise with 1000 users is to expensive too.

Also moving to cloud would hamper our installation because of the missing functions that  Mohammed Amine has mentioned above.

We just have started with Jira/Confluence one year ago, if you stay with this decision we and a lot of middled sized companies in the EU will be forced to leave Atlassian.

An idea might be to add smaller and more affordable datacenter licenses (e.g. 25, 50, 100, 250) to your portfolio, so that we can continue working with your products.

We really like your software! Please make it possible that we can continue using it!

[Kimberly Deal _Columbus ACE_]

As a community leader I meet people from all industries and all company sizes.  How great is it that we all have access to such a wonderful product suite that really advances us all. The move, of retiring the On-Prem products with their easy entry level, moderate tiers and Non-Profit/Academic license, now really makes it only viable for those companies that have the revenue and the type of business that will afford them a cloud or Data Center solution.

I feel like Atlassian has really lost it's way from the company it once proclaimed to be and it makes me incredibility sad.  I have personally recommended the Atlassian products to just about anyone who will listen to me. I have helped more than a few people customize the Atlassian products to do what ever they needed. More than a few out of the box thinking sessions have been held.  I felt like I was part of a company, product, and social good in the world working with Atlassian, and they never needed to put me on the payroll to make me want to do that for them. 

I don't feel like that today.  I am watching Atlassian turn into any other company out there that values profit over all else.  I had hope that Atlassian would be able to show us a different way to be.  That they could have shareholders and still do great things for their customers and the world at large.  I wanted to see that happen, and did what I could as a Community Leader to help make it possible.  Now I see that maybe it isn't and we're just in a status quo world with a status quo company.  My disappointment in that is a very heavy thing to carry today.

[Jeff Tillett]

New mission statement time: "Unlocking the potential of all teams....who can afford to pay"

I'm totally with you @Kimberly Deal _Columbus ACE_ . Charlie Horse is sad too. 

[Marco Moraglia]

As another mid-sized company, we have been using Atlassian products such as Jira Software Server and Jira Service Desk Cloud for a few years now, even recently upgrading our license for Jira Software Server. We are currently evaluating Confluence, having considered buying Confluence Server, while testing the features on Confluence Cloud.

After our initial period of discovery, we have found that Confluence Cloud was not addressing all our requirements, for example, to meet our needs, we needed to buy and add apps, many of which are only available for the server version of Confluence.

Now that Atlassian has removed server products from the road map and that Data Center products have a minimum user requirement that is too big for most of our use cases, we are being forced to look elsewhere.

Dear Atlassian, thank you for this opportunity to look at other available solutions. We already found interesting alternatives, some of which are even Open Source.

[Axel Faustmann]

Historically Redmine users, we started our JIRA & Confluence journey with a slightly oversized 100 user license of the server product in June 2018. Since then, we have been growing the volume of business processes and users covered continuously, which triggered additional licensing both of Atlassian software and partner plugins.
We are now at standing at 250 Core & Software users each, as well as various plugins. We keep hiring a local Atlassian Partner as new customizations are required. Based on the positive experience so far we had plans to grow this further in the years to come. A licensing level of 500 users is not out of reach, but it will take time to get there.

The cloud version is a non-starter for us, as outage of internet may not affect the service. There are additional concerns in terms of customization, lacking api capabilities, data access and geographical storage of information.

Covid-19 is hitting us hard. Revenues are down by 75%, we lost more than a third of our colleagues already. To force customers to expand their license is never a helpful move, but budgets are substantially harder to access these days. It is certain that the extra licenses we'd be entitled to on paper for the smallest Data Center subscription would not serve any productive use in the next year or two.

As Taylor has noted above in the linked post, the move against the server product also raises serious concern about the longer term future (5-10 years) of the Data Center edition. Certainly it looks from the outside that it should commercially suffer long term, because companies like ours will lack the entry level licensing to grow to a size where Data Center makes commercial sense. Together with the fundamental problems posed by the Cloud, this puts the whole strategy for Atalssian products in the organization under question.

Please reconsider. We need a pricing model which allows gradual licensing cost increase also for an on-premise edition, whatever the name may be.

Thank you for listening.

[German Leontev]

Time to find an alternative.

[Sergei Titarenko]

It is a very sad and disappointing move from Atlassian and probably irreparably damaging client's trust in Atlassian's solutions. Some corporate clients of over 1k users suddenly discover that in 4 years from now they will have to pay almost 15 times of their current annual subscription for Confluence & Jira Cloud yet immature to their business needs (plugins, automation tools, test environments won't work out of the box and will require significant additional investment). How come only 15 times increase, why not 150 times then? I felt happy using Jira for 17 years, I don't feel like that anymore. 

"One of our core values is “Don’t #@!% the Customer,” and we will be sensitive to customers facing challenges due to COVID-19." from https://s2.q4cdn.com/141359120/files/doc_financials/2020/q4/TEAM-Q4-2020-Shareholder-Letter.pdf