AMA on Tuesday, April 7th, with Anu Bharadwaj: Atlassian's Head of Product Management

[Bridget]

Hello Community members!

On Tuesday, April 7th, from 3-4 PM PST, Anu Bharadwaj, Atlassian's Head of Product Management will be LIVE on Community to answer your questions.

In case you missed it, you can watch her Remote Summit 2020 session recording here: Atlassian Enterprise Cloud - Scaling the Future of Cloud

Feel free to submit your questions before the AMA goes live - Anu and her team will be sure to address them when the time comes. 

We look forward to reading your questions - don't hold back! 

[Russell Zera]

Can you speak to the roadmap of Jira Next-Gen and Jira Classic as it pertains to Cloud customers? Presently it seems as though there are multiple offerings with no plans to sunset Classic in favor of the Next-Gen, but at the same token, the Next-Gen does not have the capabilities necessary for some of your customers (like myself) to be able to use it on the enterprise level to fit their needs, So I would like to know if there is a roadmap that maps the MVP features of Jira Next Gen that Atlassian feels must be in place prior to sunsetting Jira Classic?

[anu]

Good question Russell. We are impatient to move quicker to bring more capabilities to next-gen projects too. The best place to see what we’ve shipped for next-gen so far and what’s in the works is our public next-gen roadmap: https://www.atlassian.com/software/jira/whats-new/next-gen#overview

That said, we’re painfully aware that having this bifurcated experience is making it more difficult for many teams to grow on Jira Software Cloud. While we’ve done our best to ensure that the newest capabilities for enterprise customers in Jira Software Cloud, like automation and advanced roadmaps, work across both classic and next-gen projects, we also know that it’s bound to create a sense of unease for Jira administrators like yourself.

Let me be clear: we have no plans to sunset Jira’s classic projects at this time. We’re really proud of the progress we’ve made in simplifying Jira with next-gen projects, but we’ve heard loud and clear that they are not yet fully ready for enterprise customers. We are working through the best path to reconcile both project types and make next gen more powerful.

So to directly answer your question, the link above is the best resource on what features are coming to Jira next-gen, but there is no “MVP” set of features that we believe we need to have to sunset classic projects because as of now, we’re not planning to sunset classic projects.

[Russell Zera]

Love this response and thank you so very much! That is what I was hoping would be the answer and that we can continue to leverage some amazing things from both Classic and Next Gen capabilities. For what it's worth, I always imagined a world (insert your own movie trailer voice overlay here saying "In a world...") where you could leverage a base standard workflow as a template for projects (say "To Do", "In Progress", "Code Review", "UAT", "Done") for standardized reporting and alignment, but then embrace the Next-Gen team empowerment mechanism to allow teams to add custom steps on top of that template however they want, so long as they don't violate the enterprise underlayment (so if a team wanted to add "UX Review" between "Code Review" and "UAT" to better define the way they work, to the intent of JNG, they could do that and it wouldn't impact the larger alignment and ability for team members to join a new team quickly). This is something I would love to eventually have, because I truly want to have squads harness the power of JNG, but we do have to have some general commonalities and structure to enable us to operate at scale.

Thanks so much!

[anu]

Thank you, Russell. Yes, our goal is to have our project types be both simple and powerful. Tall order, taking us time to get there. Thank you for your patience

[Kat Warner]

I hope I am not too late for this question.

Could you follow up on the integration between Confluence Server and Trello?

https://jira.atlassian.com/browse/CONFSERVER-45779

This is a significant usability issue for many of our customers (as a Solution Partner) and for our team communication (Confluence is where we could like all out streams of work to converge from Marketing's Trello Boards, to Dev's Jira backlogs, to Support's customer project tracking).

[Alison Huselid]

Kat - Thanks for calling out this issue as one that needs follow up.

Right now we don't have plans to add a Trello macro in Confluence Server/Data Center. We are instead first focused on improving the general usability and intuitiveness of existing macros based on feedback we’ve received. Once that effort is further along, we will re-evaluate introducing additional macros, including one for Trello. 

We’ll go ahead and provide an update on the ticket with this current status and you can also watch our current near term efforts on our public backlog for Confluence here: https://jira.atlassian.com/secure/RapidBoard.jspa?rapidView=2529 

[John Funk]

Well, you said don’t hold back, so here we go. I know this will probably not make me very popular in the Community or in Atlassian ranks. But I feel it is important to represent the users to some degree.

This is my third attempt at this (and my last 😊 ), but honestly, I was very disappointed with the Product Leadership AMA on Monday a week ago. My question, which I paste again below, received the most votes, but it was well after the first half hour of the AMA before it was even addressed. And then it was clear that the question made people uncomfortable and the full question was not even read. And as a result, in my opinion, it received corporate doublespeak and was never truly addressed. I hope that can happen today.

In the meantime, I have seen several emails with comments on open requests that validate my concerns even further. I include some of them here – simply read some of the latest comments on each request.

https://jira.atlassian.com/browse/ID-6677

https://jira.atlassian.com/browse/JRACLOUD-17783

https://jira.atlassian.com/browse/JRACLOUD-67563

https://jira.atlassian.com/browse/CLOUD-6999

https://jira.atlassian.com/browse/JRACLOUD-70555

The first two above have more than 1000 votes and have been in the Gathering Interest state for 17 years and 11 years respectively. Frankly, that’s insulting to users.

As others have commented before, it seems that Atlassian has decided certain paths and is pursuing them regardless of customers’ requests/comments – like the move to Cloud, New Issue View, Next-gen projects, etc. While outstanding requests for years get ignored.

One of the links above is aimed at the New Issue View and has many comments from users about how they do not like the interface. AND there are already more than 100 sub-tasks linked to it for bugs and changes. That means that dozens, probably thousands, of other requested features have gone undone while the new interface was developed, put in place and corrected.

Granted, there are lots of grumblers out there and always will be. But I don’t think the grumblings related to these types of things is unfounded. And as stated above, this will be my last time to grumble about it.

So, I guess my question would be “Is Atlassian Leadership aware of these problems, aware of the lack of communication on these requests by Atlassian personnel and is there any plan to change how the requests are handled.

Below is a part of my original question for that AMA last week that was not addressed (again in my opinion):

My question is around selection of features to implement. And, I guess I have two different areas to explore.

1. There are requests that have hundreds or even thousands of votes yet are ignored. I say ignored because some have been open for YEARS. I know of at least 2 that celebrated their 17th birthday this month. That’s incredible! To not have a policy to quickly deal with the requests – either do them or scrap them – is very baffling to users. It’s also not good customer service. And often, there are no comments at all from Atlassian personnel for months or years about what is going on with requests. This is extremely frustrating for customers.
   
   
2. I know you have a document that details what process you go through for feature selection, but frankly, the end result that I get from the document is that you pretty much do what you want to do regardless of what customers are asking for. That’s another reference to requests that have hundreds or thousands of votes. But then things like the new issue view or next-gen comes along with no apparent supporting requests for those – certainly not thousands asking for those. And the amount of negative comments about those products/features shows the frustration of customers.

[John Funk]

I should have also included this link above - sorry for the oversight:

https://community.atlassian.com/t5/Atlassian-Answerers/Atlassian-backlog-management/gpm-p/1336963#M275%3Futm_source=atlcomm&utm_medium=email&utm_campaign=immediate_general_post&utm_content=topic

[Darin Klein]

@John Funk - THIS -  "As others have commented before, it seems that Atlassian has decided certain paths and is pursuing them regardless of customers’ requests/comments – like the move to Cloud, New Issue View, Next-gen projects, etc. While outstanding requests for years get ignored.

One of the links above is aimed at the New Issue View and has many comments from users about how they do not like the interface. AND there are already more than 100 sub-tasks linked to it for bugs and changes. That means that dozens, probably thousands, of other requested features have gone undone while the new interface was developed, put in place and corrected."

I cannot agree more and have also voiced my opinions about the Cloud push, the new issue view, next-gen, etc. and no one seems to want to listen. 

Thank you for bringing this up again.  

I love the Atlassian DC and Server products, but just can't get behind Cloud. I run a relatively large instance of it and we're pulling 80% back to on-premise (Large DC environment) this year.  We'll still use Cloud, but it's just not right for the majority of our work. 

[Russell Zera]

I will say, as a Server to Cloud convert, that we have been finding immense success at large enterprise cloud adoption in my company. It took a little getting used to, but I honestly can say I PREFER Cloud to On-Prem now! What I will say is that I do understand and share the concern from @John Funk about not understanding the "Gathering Interest" status and the thresholds for customer votes to get a feature into a state where it is actively worked... It does FEEL sometimes that maybe the "Vote" functionality is a mechanism that is utilized to evaluate "Gathering Interest" but I don't know if there are thresholds. I also know that there are things that are much more complex to implement than to type, so I also give a lot of patience and understanding if something that is important to me, might not be a "stop the presses, guy here has a need" type effort. I do wish for more communication and transparency into what each status means and how the request and ticketing functionality works so that I know EXACTLY what each status means, how they progress, and have more clarity that I'm heard.

For what it's worth, I've been very lucky to have seen, on multiple occasions just how much Atlassian cares about their customers and the feedback they receive, which is why I try to share as much as possible that experience or what I have witnessed with others so that people can fully see just how much Atlassian creates as a result of customer feedback on issues like these! Maybe one idea could be to share that kind of data more regularly? Something like "% of features delivered this month that were customer requested/driven" or "Customer votes fulfilled this period" or something like that? Also communicating more about the UNBELIEVABLE amount of user testing that Atlassian does, and how many features they implement from those user research sessions?

TL/DR: Just spit-balling, but I think this thread may be more about possible improvements in customer communication on requests and tickets (Expectation Managemenrt), and less about the number of issues that Atlassian is releasing.

[anu]

Hey John, thank you for the hard questions. I’m sorry about my late reply and I’ll try to address your points directly and honestly. I know my answer might be disappointing, but to directly answer your question as to whether Atlassian leadership is aware of the issues you raised, and aware that we haven’t done a consistent job communicating updates, the answer is an unequivocal yes.

I know that we say this a lot, but it bears repeating: feature requests on jira.atlassian.com are an important component of our product roadmap, but far from the only one. Our responsibility as product team is to look at the needs of our past, current, future, and potential customers from many different angles, and aggregate this into the best product roadmap we can. We conduct market research, we interview customers in-depth, we visit them in their offices, we talk to them at Summit (and Remote Summit), we collect feedback directly in our products, and so on. We even analyze feature requests on jira.atlassian.com in ways you might not think of! For example, we don’t just analyze total votes but also “vote velocity” (how quickly requests accumulate new votes) and we perform sentiment and keyword analysis on the comments themselves. These inputs holistically influence our strategic decisions.

In addition, our strategic focus on SaaS and examples you raised like Next Gen are driven by market waves as well as customer feedback. These are decisions we make collectively as a company to ensure our competitiveness in the future.

All that said, we accept that we will never satisfy every single request. Actually, it’s worse than that. There will always be more requests created and ideas generated than we could ever implement. And the implication of this is most of the time, we won’t be working on most feature requests. Even ones that are unequivocal improvements to our products. Even the ones that have been open for 17 years. If we ran our backlog purely on a first in-first out model, I truly believe we would be doing a disservice to our customers.

In my view, there are three classes of requests on JAC that have been around for forever:

1. Ideas that we recognize customers want, but don't align with our product philosophy or vision. Or we believe implementing will detriment the product in some other way. Call these the "don't think we should"s. These ones usually get closed as “won’t do” pretty quickly.

2. Ideas that we believe absolutely have merit and would make the product better for most or all customers. But the value doesn't justify the cost to implement the feature (not just the immediate dev cost, but the opportunity cost of other features we would not be able to work on). Call these the "wish we coulds".

3. The ones where we're genuinely unsure how much customer demand there is and we are actively seeking out feedback and use cases. This is what JAC should primarily be doing. Call these the "maybe we woulds". Usually we leave these open to track the number of votes and watchers

I think most of the vitriol comes around tickets in the 2nd category. Because over a sufficiently long time horizon (i.e 17 years) the argument that the dev cost of the feature appears weak. Let's say renaming groups would take, based on a reasonable level of investment, 18 months to implement. But that's not actually how decisions are made. We don't plan our roadmap on 17 year time horizons. What actually happened is that at many points over a 17 year period, we looked at it and said "Is this the best way engineers could spend 18 months to maximize value for customers?" and the answer at each point in time has been "no" And this discrepancy lies at the heart of a lot of the mismatched expectations that drive negative sentiment on JAC.

Finally, I absolutely agree and hear your frustrations about how we have handled our responses on JAC. I think we have lacked the discipline to update the tickets as often our customers would like, and for that I apologize on behalf of my team. We’ll work towards improving our process so that we’re up to date and being transparent about the feature requests that we do choose to work on or reject. I think the Server team in particular sets the bar for where we want to get to: https://jira.atlassian.com/secure/Dashboard.jspa?selectPageId=85890. I’ll make sure that we leverage JAC in a way that is productive and honest for both Atlassian and our customers.

Thank you again for your passion and honesty.

[Russell Zera]

Very nice response @anu! I really found your definitions around the "Don't Think We Should", "Wish We Could", "Maybe We Would" categorizations helpful and interesting (Very similar to the MoSCoW (Must have/Should have/Could have/Won't have) prioritization method). I'm curious if leveraging your categorizations more in your statuses or more appropriately resolutions would be an improvement for some customers? I think a blanket "Gathering Interest" that hasn't changed in years is misleading and could be a challenge, but if it was much more clear that it was just some word or two that embraced things like "Great idea/feedback, we would love to implement this but it doesn't align right now" and giving a bit more transparency in the comment might help? When I look at most Jira instances I have encountered over the years, the first thing I look for is "Do the tickets in this instance provide me a complete picture? If I were to pick this ticket up right now, would I understand where it stands and know what's next?" I think that's where my feedback comes from on the JAC tickets, is I seldom see that in JAC tickets when I look at them, but I'm also on the customer side, not the dev side, so I might not be seeing all the info anyway, which is totally ok.

One thing I LOVED was your description of vote velocity which is incredibly valuable... can you tell me how you view vote velocity, and when Jira Cloud customers might be able to get the VOTE feature returned to our product? It was a great feature that I was completely shocked disappeared, and would love to have back. Seems like such an integral function for a team to express interest in a particular feature/story/task/etc that Cloud users presently don't have. 

https://jira.atlassian.com/browse/JRACLOUD-71251

Final note, the ticket referenced above has a great example of better communication on what is happening on a ticket (Bravo!), I'd just like to see it a touch more frequently ;-)

Thank you so much for your great insights and wonderful transparency.

[John Funk]

Hi @anu  - Thank you for your very open and honest reply. Overall I was very pleased with the response. I also found the three classes of requests very insightful and would echo Russell's response that if those could somehow make it into the tickets if might give the customer a better idea of what might happen with it. 

I also would like any insight you can provide into how you handle vote velocity. I understand that you have lots of different influences in how you will deal with a JAC request, but the customer is just not going to understand those as a whole. All he or she sees is 1,000 votes and zero movement or comments. I REALLY appreciate your commitment to changing that moving forward - that means a great deal. One suggestion would be that if you move the ticket to a disposition much quicker, there will be less time for votes to pile up on something you probably are not going to do. 

Thank you again for the opportunity to even pose such questions for you, and for taking the time to thoughtfully answer them. I know that I have been heard.  :-)

Finally, what can the Community do to help with responses when people go over the deep end with comments on tickets? Anger feeds anger and if there is anything we can do to clarify a process or provide some insight, I would be happy to do that. But I don't speak for Atlassian, and never want for it to even appear as such.

Thanks again!

[anu]

Thank you for following up, @John Funk and @Russell Zera 

We talked about this specific question at the exec weekly at Atlassian and @Jurgen Spangl1 , our beloved Chief Customer Experience Officer has raised his hand to work on a proposal of how we integrate our customer feedback channels so that we more clearly communicate how we are closing feedback loops on customer raised requests. I would love to have you two review this proposal and provide feedback on what you think. Give him a couple of weeks to get back on this thread. 

You are right - anger feeds anger. Please point to this thread where useful and please be Atlassian's voice on how we deal with customer feedback when you see users get frustrated on tickets. Most of all, please continue to keep us honest when we are not doing a good job of managing expectation or delivering the right message. We are lucky to have passionate folks like you representing Atlassian as well as our customers! 

[Russell Zera]

It would be my pleasure @anu ! This has been an amazing experience even being able to pose these questions (as @John Funk stated above). This is the type of experience with Atlassian that I have grown to love. The openness to hearing from their customers if something they are doing is missing the mark, and the commitment to do their best to correct it. I have seen it countless times, and always love opportunities to share with others that experience, and that Atlassian is listening and acting upon their feedback, they just might not see it. :-) 

I have bookmarked this page for reference as I encounter posts or threads that it would be helpful to direct them to this to see. Thank you and God Bless!

[John Funk]

Hi @anu   - Absolutely! What an honor to be able to be a part of that. Thank you so much for raising that to the executive level and the work that is progressing. I couldn't be more pleased!  :-)

[John Funk]

Hi @anu - Here's a good example of one with several votes but has gathered some recent negative comments. How's the best way to see about getting a Product Manager to give an update?

I have added a comment to help drive some civility, but not sure how much that will help. 

https://jira.atlassian.com/browse/JRACLOUD-17783

[LarryBrock]

Hello @anu just having a chance to read through this thread and second all that @John Funk, @Darin Klein, and @Russell Zera had to say (both questions and comments about your responses).

One concern that I'll add is a perceived lack of participation over time by the Product team on Atlassian Community as a method of engaging with customers and steering feedback in constructive directions.  As an example of anger feeds anger there was, for several weeks recently, a raging dumpster fire Community thread about the new Confluence editor.  The Community team, bless their hearts, responded magnificently to the situation both engaging directly and asking us leaders to comment about appropriate behavior on the site and more importantly, justify the change(s).  What was glaringly missing was any engagement by the product team to address the issue.

As part of the review and proposal you mentioned Jurgen Spangl will be making, please consider incorporating more active Product team engagement on the Community site as a vehicle for closing those feedback loops and proving customers concerns are being heard.  I can speak from years of experience as both an AUG Leader, Community Champion, and now Community Leader that meaningful engagement by Atlassian staff, especially those in decision-making positions, makes very positive impressions.  (This post and the discussion therein is a great example! )

Thank you for not just hearing but for truly listening and respond to the concerns mentioned here.  Cheers!

[Darin Klein]

Thank you @anu for the response and all the excellent comments / discussion happening around this topic.  This is exactly what should be happening and I love the moves to make it easier for us users to understand why something is or is not getting picked up for a future release. 

[Leisa Reichelt]

hello everyone - thanks for the additional context you've been adding recently, pls feel free to keep the examples coming.

As @anu mentioned,  @Jurgen __  will be owning how we improve the way that Atlassian responds to feedback - I'm part of Jurgen's team and this work will fall into my area. We've been doing some work already to try to understand the problem and some potential solutions, and as Anu mentioned earlier would love to include some community participation in helping us with this. 

We have a few things we need to do first, but we hope to be back here really soon with a format for you to participate in this process. 

Look forward to working with you on this soon.

[John Funk]

Hi @Leisa Reichelt ,

It's been a little over a month since your last post. Is there any update you can share?

Thanks!

John

[Leisa Reichelt]

Thank you all for your comments on this AMA post. We want to keep the discussion going and understand how we might deliver a better feedback experience for everyone. Please join us here.