This interview at Atlassian Team 26 features Tanguy Crusson of Atlassian, product lead for Jira Product Discovery, discussing product prioritization, acceleration driven by artificial intelligence and the Jira Product Discovery product collection and feedback app. Crusson outlines how their team approaches prioritization, roadmapping and connecting customer signals to delivery work, and explains how Atlassian integrates qualitative feedback with product analytics. The interview is part of theCUBE Research and features Alison Kosik of theCUBE Research and Christophe Bertrand of SiliconANGLE and theCUBE.
Key takeaways include the increased importance of choosing what to build and the need to start small, test and validate before scaling. Crusson recommends a bucketed investment model covering customer wants, product usability and reliability. They emphasize tying feedback directly to insights and work via Atlassian's platform. Analysts observe that AI accelerates development while raising the risk of the build trap, making accessible data and iterative validation essential for effective roadmapping and product strategy.
Watch for practical guidance on product prioritization, feedback management and using product analytics to improve decision making in SaaS platforms and product management practice.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
Atlassian Team '26. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open the link to automatically sign into the site.
Register for Atlassian Team '26
Please fill out the information below. You will receive an email with a verification link confirming your registration. Click the link to automatically sign into the site.
You’re almost there!
We just sent you a verification email. Please click the verification button in the email. Once your email address is verified, you will have full access to all event content for Atlassian Team '26.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
OR
Connect via Twitter
Connect via Linkedin
EDIT PASSWORD
Share
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
Atlassian Team '26. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open the link to automatically sign into the site.
Sign in to gain access to Atlassian Team '26
Please sign in with LinkedIn to continue to Atlassian Team '26. Signing in with LinkedIn ensures a professional environment.
Are you sure you want to remove access rights for this user?
Details
Manage Access
email address
Community Invitation
Tanguy Crusson, Atlassian
This interview at Atlassian Team 26 features Tanguy Crusson of Atlassian, product lead for Jira Product Discovery, discussing product prioritization, acceleration driven by artificial intelligence and the Jira Product Discovery product collection and feedback app. Crusson outlines how their team approaches prioritization, roadmapping and connecting customer signals to delivery work, and explains how Atlassian integrates qualitative feedback with product analytics. The interview is part of theCUBE Research and features Alison Kosik of theCUBE Research and Christophe Bertrand of SiliconANGLE and theCUBE.
Key takeaways include the increased importance of choosing what to build and the need to start small, test and validate before scaling. Crusson recommends a bucketed investment model covering customer wants, product usability and reliability. They emphasize tying feedback directly to insights and work via Atlassian's platform. Analysts observe that AI accelerates development while raising the risk of the build trap, making accessible data and iterative validation essential for effective roadmapping and product strategy.
Watch for practical guidance on product prioritization, feedback management and using product analytics to improve decision making in SaaS platforms and product management practice.
play_circle_outlineInside Jira Product Discovery: Tanguy Crusson on Prioritization, Roadmapping, and Serving 25,000+ Customers
replyShare Clip
play_circle_outlineFrom MRDs/PRDs to Dynamic Product Operations: AI-Driven Customer Feedback and the End of HiPPO Decisions
replyShare Clip
play_circle_outlineAnnouncing Discovery Expansion: Feedback App Aggregates Signals into a Data Mart to Connect Qualitative Feedback and Product Analytics
replyShare Clip
play_circle_outlineAtlassian’s unfair advantages: large distribution network and extensible platform.
In this interview from Atlassian Team '26 in Anaheim, Tanguy Crusson, product lead of Jira Product Discovery at Atlassian, joins theCUBE's Alison Kosik and Christophe Bertrand to discuss how AI is reshaping the discipline of product management and why choosing what to build has never mattered more. Crusson, who oversees a prioritization and roadmapping tool used by more than 25,000 customers, announces the new Atlassian product collection at the event. He warns that AI's speed advantage has made the "build trap" — shipping feature after feature that fails to ...Read more
exploreKeep Exploring
What do you do in your role at Atlassian?add
Has the growing ability to collect direct customer feedback and analyze product interaction/support data — especially with AI — changed how product teams make decisions and reduced their reliance on opinions and secondhand information?add
How is Atlassian helping product teams use customer feedback and product analytics to prioritize product decisions?add
How does a platform that ties customer feedback to insights, ideas, roadmaps, and work (and supports agentic workflows) help teams scale, differentiate, and follow up with customers?add
>> Welcome back to Atlassian Team '26. We're live in Anaheim, streaming live right here, right near Disneyland. I'm Alison Kosik alongside Christophe Bertrand, and we're about to get a little technical.
Christophe Bertrand
>> Yes, we're going to talk product with another Frenchman.
Alison Kosik
>> Let's bring him in and bring in Tanguy Crusson. He's the Product Lead, Jira Product Discovery, here at Atlassian. Welcome to theCUBE.
Tanguy Crusson
>> Hello. Thanks for having me.
Alison Kosik
>> So talk us through what you do in your role at Atlassian.
Tanguy Crusson
>> Sure. I lead product for Jira Product Discovery, which is a prioritization and roadmapping tool that's been used by more than 25,000 customers today. And today, we're announcing the product collection, which is an extension of this with a few more apps. So basically, I focus on that, the impact of AI on product teams, and how to link that to the delivery work that engineers do.
Alison Kosik
>> Congratulations on your product announcement.
Christophe Bertrand
>> So needless to say that the old days of market requirements, documents, and product requirement documents, those days are gone. We don't do that anymore. How has it changed? And do we even need those since we have now what is a more dynamic way of approaching products?
Tanguy Crusson
>> Yeah. So what we've seen is that the maturity of the craft tends to differ between customers and industries, but it really goes from the world that you described, but mostly, transitioning to the product operating model, which was first initiated by startups, tech companies, and this way of working, which is, like you said, a lot more dynamic, is really transitioning to be adopted as the standard way of working across all product teams out there. We've come a long way. It's a lot more dynamic. We still have a long way to go. Product teams still tend to rely on a lot of opinions, gut feel, listening to more senior stakeholders or people with, like we said, the highest paid person in the room. And so that's the kind of stuff that we're trying to help change.
Christophe Bertrand
>> Would you say that if you think about where we're at now, the ability to go collect a lot more customer feedback, and also, look in what, in existing products, is going well, not going well, I mean, you have all of these logs, all sorts of interactions happen with tech support, et cetera, we essentially sometimes walking away from the easy path, but just looking at what the market is telling us?
Tanguy Crusson
>> So product is an incredibly hard job in that, whatever you do, you'll never know if you got it right after a lot of analysis and a lot of iterations. What used to happen is that product managers had to rely on secondhand information-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> from people who could summarize it for them. So you could not know what every support ticket would be. So you would go to a customer success manager to understand what pains the customers were going through. You would go to someone who could summarize what's happening in sales calls with customers, but product had to fly a bit blind and rely on secondhand information and then fill in the blanks to make product decisions. Now, where the craft evolved a lot before AI was the fact that since we know that we are always having complete information-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> we need to test and validate more. So-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> we do try something which might be only 10% of what we're going for, but then we keep testing and iterating and validating to keep informing investments. And now with AI, we're getting to the stage where it's actually possible to go straight to the source and here is straight from customers, which is where we're currently going.
Alison Kosik
>> So has AI made product development easier or is there a good bit of risk in this as well?
Tanguy Crusson
>> So product development itself definitely got faster. You can ship features way faster. And what, product teams, a lot of them have done is revert back to this whole habit of what we call the build trap. Melissa Perry coined that term and describes the fact that you're trying to achieve a goal. And for that, you ship a feature. This feature doesn't work. What you do naturally is to try and ship more features to try and close the gap. And eventually, you land with a product that's bloated that no one uses. That's the build trap. And so what we've realized is that now with AI, it's easier than ever to do that and to land with a massive build from your AI provider. So really where the product teams need to go now is to focus a lot more on choosing what to build, and deciding what to build becomes even more important now than before AI. It's always been important, but now it's even more important.
Christophe Bertrand
>> Yeah. It's interesting that it would be so emphasized because, to me, that was always the rule. And then, of course, the risk of you listening to your large customers and you custom build a feature for them, but it's not really for the market. So how do you bring the market back into the conversation? Beyond, again, looking at feedback from customers and primary interaction with them, there are competitors out there in any product, any industry. How do you reconcile everything?
Tanguy Crusson
>> Yeah. So for us, being at Atlassian, we started from developers because we do know a lot from how software gets built and what works well there. And we've progressively moved more and more to the left in that we started with developers and then we moved to how to help people get together and agree on what the next priority should be initially with more of a collaborative way of working-
Christophe Bertrand
>> ....
Tanguy Crusson
>> and progressively more and closer to the data, which is why now we're announcing a bunch of initiatives with our new product collection, one which is a feedback app that helps you collect signals from every customer conversation, whether they were sales, support-
Christophe Bertrand
>> ....
Tanguy Crusson
>> customer success straight from the source, any support ticket, anything happening online when people talk about your products and bring that as a data mart that product managers can now query. And we're marrying that with product analytics so that you can compare that with how people actually use your products-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> because the last thing you want to do is to focus your time on a feature that's in your free plan that a few vocal customers are using-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> when what you're trying to do is to improve the experience of enterprise customers, right? So we're basically bringing all that together, which is qualitative and quantitative data, to make it accessible when you're making product decisions.
Christophe Bertrand
>> All right.
Alison Kosik
>> So you're literally separating the signal from the noise in the most practical way.
Tanguy Crusson
>> That's the idea. And what AI made possible now is to make all of this data accessible to everyone. So to give you an example, I recently did a roadmap for all this. And before, what I would do is to have a data scientist work with me and I would ask questions about the market. I would ask questions about customers. And this data analyst would basically run very complex queries to give me answers so I could make decisions. Now, we've used the tools themselves that we've built to do that. So sorry, that's a bit meta, but basically, this idea that I can use this data, it's at my fingertips when I have questions, but it's the same for my stakeholders. So when I have a discussion with my stakeholder, instead of just basing it off information that's been filtered down via multiple levels, we can go straight to the source when queries, 30 seconds later, we have answers-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> and it makes for much richer conversations.
Christophe Bertrand
>> So I have a question about features that may or may not be used. Obviously, yes, you never want to build features or expand on features that are not generating either use or income. But in a competitive world, sometimes you have to have those check boxes, right? And if you don't have them, you're not even going to be in the running. Again, whatever industry you're in, but let's take software services generally in tech, some in area I'm very familiar with as a reformed product person. Sometimes you got to have it. It doesn't matter. How do you account for that? Are there ways that you can maybe say, "Well, look, hang on. We understand. These are really the things... These are must haves that you're not going to invest too much." Is there that level of intelligence already?
Tanguy Crusson
>> So basically, that's where the approach that we recommend to our customers is to think of it as buckets.
Christophe Bertrand
>> Okay.
Tanguy Crusson
>> So you have a team, you've got resources, you've got investments, and what we really need to do is to invest those smartly in different buckets, depending on the goals that you're trying to achieve. So we tell our customers, you can start with three buckets. The first one is what your customers want, which is new that you don't have in your products yet. The second one is the product as it is today. And is it usable? Are people actually adopting the features that you've already got-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> before you even build new ones? And the third one is reliability. Is your Product always up? And can people really achieve what they need to do with it without bugs and things like that?
Christophe Bertrand
>> Right.
Tanguy Crusson
>> And so you start with something like this and you progressively expand. And so what you need is just to be very intentional in how you allocate your resources to these different buckets. So it might be that initially, you spend most of your time implementing new features, and over time, you allocate more resources to just improving the existing ones. There comes a moment when you start a new product where, let's say that you start by selling to small and medium businesses. So you don't really need to have all the checkbox features yet.
Christophe Bertrand
>> Right.
Tanguy Crusson
>> As you move up market is when you start to allocate a bucket for that. So this type of prioritization that tries to be not one size fits all, but instead, tries to capture the different nuances of like... There's so many things you can do, but you need to find ways to be intentional about it, which is what we try to help with-
Christophe Bertrand
>> Yeah. I would just argue that this example is interesting of medium-sized businesses maybe going into enterprise. I would argue, well, you can want to build scalability from a design standpoint, don't necessarily want to invest all of your money yet, but everything is fun and games until you are at scale in any industry whether it's about data, consumption, compute, you name it, right?
Tanguy Crusson
>> Yeah.
Christophe Bertrand
>> So I'm curious about that because we live in a world where everything's accelerating, everything is consuming a lot more data. Hardwired products are not really hardwired anymore. I mean, I was using the example of a video camera. Is it a video camera or is it an AI tool now, especially for private home cameras and things like that? So how do you account for that in a way that really allows you to differentiate now your product in the market? Because at the end of the day, you want to achieve sustainable differentiation in your products, right?
Tanguy Crusson
>> Yeah. So that's where you need to play to your unfair advantages.
Christophe Bertrand
>> Right.
Tanguy Crusson
>> If I take the example of Atlassian and what we are shipping with this product collection, what we are relying in is on two things. The first one is Atlassian's unfair distribution advantage. We've got more than 300,000 customers we can distribute any new product to. And the second one is our platform. So we have a large team at Atlassian building our platform so that when we create the product connection on Atlassian, we build on top of this platform, which unlocks so many things for us. For example, we're adding new apps, and the feedback app is one of them. With the feedback app, we're introducing two nouns, two objects in our teamwork graphs. So we're adding feedback and insights. That means that now that whenever someone needs a piece of feedback, it gets tied to an insight, which is tied to an idea, which is tied to a roadmap, which is tied to a piece of work. So whenever we ship something in Jira, we know which customers ask for it. We can get back to them. We can send them a personalized message, that kind of stuff. So that's the differentiation that you get by building on a platform that gives you the scale. And then your workflows, basically, anytime you add one more piece to it, you add a lot more in terms of capabilities, especially now that you can create agentic workflows on top. So whatever the products don't do, you create an agentic workflow. It's based on objects on that graph-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> and so on and so forth.
Alison Kosik
>> I'm wondering, with AI helping teams move faster, how much does this increase the likelihood of just making the wrong decisions, even having a human element involved?
Tanguy Crusson
>> Yeah. So the AI moving faster means that you can probably go a lot further into focusing on features that are not going to work. And the problem is that even if you listen to your customers, you've got all this data about what they need. You think, "We know what they need. We know what features to build. It's going to work," right? Well, in most cases, there's a truism in product, which is 50% of what you try is going to fail, regardless of the validation that you had upfront. So the way we try to go around that is, instead of just shipping features based on upfront analysis, we basically start small and we test with customers and we test whether the feature actually resonates with them. Are they going to use it? Do they get the value from it? Is it usable? Can we afford to keep investing in those ideas because things always turn out more expensive than what you thought they would be? And you do that by progressively developing your appetite based on the validation that you have. So it's a bit of a... Initially, your appetite is small because the validation is small. And as you get more and more validation, your appetite can grow with it. And if it doesn't work, you just shut it down, right?
Christophe Bertrand
>> Let me ask you a question about why... It's a bit of a trick question, not an existential why, I promise. But what's the difference between a product and a platform in this day and age?
Tanguy Crusson
>> Okay. That's a very good one because honestly, a platform is a product used by products .
Christophe Bertrand
>> Exactly.
Tanguy Crusson
>> So basically, every platform team this day, it's a product team. The main difference is in the number of customers that you have for it.
Christophe Bertrand
>> Right.
Tanguy Crusson
>> So if you have an internal platform that just serves internal customers, the part of the craft that you're going to use is very different. So if you build, say, a platform that's used by millions of developers out there, like what we built at Atlassian. So often the line between those is like, I'd say platform is a product function, but the part of the craft that you use might be a bit smaller to the one that you would use if you actually focus on a customer facing platform.
Christophe Bertrand
>> Right. Yeah, because over time, what happens is platforms create friction, right? So how do you manage that sort of operational potential waste that happens where you have all of these parts, right? That are supposed to work together yet don't necessarily always do that, right?
Tanguy Crusson
>> Yeah. So that's where this... It started many years ago. You build services. The services get better as you get more and more customers that use them. The fact that we are building a platform that is used not only by our own teams internally, but by thousands of developers out there building businesses on top-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> of Atlassian means that we've got benefits of scale by doing this in that the platform itself gets much better by having to support all these very wide use cases across-
Christophe Bertrand
>> Right....
Tanguy Crusson
>> a wide variety of workflows. And that's why when we build products on Atlassian, we can go so much faster.
Alison Kosik
>> If you've had to simplify it, what's the one mistake product teams need to stop making?
Christophe Bertrand
>> Gut.
Tanguy Crusson
>> Trust that gut. Validate it. We tend to go with gut, and it's a good signal for whether something can work or not. As soon as you've got it, you start testing. You don't start shipping.
Alison Kosik
>> watching.
Christophe Bertrand
>> . Yes.
Alison Kosik
>> Thanks so much for your time. Thanks for stopping by theCUBE.
Christophe Bertrand
>> It was great. Thank you for your time.
Tanguy Crusson
>> Thank you.
Alison Kosik
>> Thank you. And you've been watching theCUBE, the leader in live technology coverage and enterprise tech analysis. We'll be right back.
Alison Kosik
Christophe Bertrand