Interested in hearing app ideas from customers?

Atlassian uses https://jira.atlassian.com/secure/Dashboard.jspa for collecting feature requests from users. Inevitably, there are requests that product teams decide we’re not going to add/build. Some of these requests are likely great candidates for being built out via an app in the ecosystem.

We’re in the very early stages of thinking about how we can get these out to the developer community. So I’m looking for thoughts/feedback on what we can do to make this happen in a way that is useful and valuable for everyone involved. Some questions…

  • Would you be interested in seeing these?
  • How would you like them shared/delivered?
  • What context/info could we provide that would be helpful/useful?
23 Likes

Would you be interested in seeing these?
Yes

How would you like them shared/delivered?
I guess it does not make sense for multiple vendors to build the same idea/feature? So it should be some sort of time-limited posting where people can sign up for it or something? Not sure how that would work, but this should not just be some bulletin board type of thing.

What context/info could we provide that would be helpful/useful?
Bring us in contact with the customer requesting the idea/feature

3 Likes

Aren’t vendors already doing this by looking at votes etc?

The big thing you would need to do is a contractual guarantee you’re not going to sherlock us as has happened on multiple occasions.

22 Likes

I too would definitely be interested.

However, I think that it should be opened. Yes, chances are that many would be trying their hands on it but that’s life. Best execution and time to market wins :man_shrugging:

9 Likes

Hi @bentley,
Thank you for bringing up this topic. We would be really interested to see and understand what customers bring up.

I agree with @remie that just throwing out these would probably not work well.

Not sure what model would work.

It would definitely make sense for vendors to record their interest for each of the customer requests.

Some possibilities for recording that interest meaningfully:

  • Use a scale of likelihood to release. Something like may enter our roadmap / will probably work on it / have a plan to ship it
  • Will be part of an existing app / will create an app specifically for this
2 Likes

We are already using JAC to assess the business value of upcoming features (we look at votes & comments).
If you create something more structured, we’d be interested too!
Especially if we can use it to connect with potentials users.

1 Like

Good points, @JaimeCapitelresoluti. I do think having something like a “firmness of release plan” from a vendor would be an important filter. Otherwise you’d end up with well-meaning, but over-ambitious people squatting on an app request (and that’s charitably assuming that no one acts in bad faith). Criteria I would regard as legitimate for judging a vendor’s committment would be:

  1. A public Jira issue (bonus points if it’s assigned, “In Progress”, has a due date)
  2. How much it “makes sense” for an existing app to add a feature. This is a super-subjective, gut-feel sort of thing, but it’s something I would consider.

On the other hand, I don’t think it’s wise to send a message like, “This vendor has some sort of claim to this app idea before they’ve even released a thing.”

Likewise, I’m rather skeptical of any system in which Atlassian contractually agrees not to sherlock vendors. Don’t get me wrong: getting sherlocked sucks big time, but… I dunno. That just smells like a treatment that might be worse than the disease. I suppose the nightmare scenario is that it becomes an avenue to stifle innovation, entrench a guild of existing vendors, and enrich a bunch of IP troll lawyers. Maybe I’m just a little scarred by software patent laws. :slight_smile:

4 Likes

Interesting aspect, I didn’t follow that … do you think it could be a good idea to provide a few examples/clues to give an impression how big that risk is?

Sorry the off - @bentley is the current Trello REST API based development being downprioritised/retired perhaps? In my assessment or as much as I could check back (I may be getting things wrong!) we haven’t had an official reply for 14 days now, communication there feels/seems a bit unusually stuck.

(Tagging here as perhaps those topic-specific notifications are going astray for you? Wild guess I know…!)

I agree with @yvesriel . The web is at its best when it is open.

Limiting it to insiders would not produce results. Companies making money often don’t want to bother with niche use cases, no money to be made and better to focus on apps doing well, and often take a long time to release.

“Limiting to insiders would not produce results?”

Unlimited access can provide a lot of bad/incomplete results, half-baked solutions, abandoned repositories on the other hand…

I don’t mean to answer the question, but I don’t believe it to be that simple a question. Maybe not the best/only angle to examine?

Still, I’m not saying it isn’t the right answer, just that the reasoning feels a little incomplete.

Maybe you had some more in mind than what you wrote?

@Janca insiders often want things behind a password, locks out so called not serious people, called gatekeepers. I think better if public. That for me is what is core to success. It needs to be fully crawled by Google or it is like it barely exists.

If five work on same problem versus one, that’s even better. One many not see the light of day or be bad, likely actually.

The big one is Atlassian could give some money, front initial sales, for an app that is approved and solves a problem they haven’t been able to solve and want solved. Means they hold back initial sales, can be as little at 1k, something to help get over the children and egg problem.

We just launch an inline Word Count for Confluence Cloud because it’s a core feature we wanted and is somehow still missing. It’s pretty niche, not clear if it will make money, but the kind of thing Atlassian needs vendors to build as they don’t seem to be able to do so quick enough (Easy Word Count for Confluence Cloud | Atlassian Marketplace).

If something can be organized to structure ideas in an open fashion it would be awesome (ie public).

But to be honest - the biggest challenge with this is Atlassian. In order for something like this to be successful Atlassian would have to give some sort of guarantee that they wouldn’t change track and build the solution into the core product. Otherwise - vendors would have to be looking over their shoulders everyday(they already do - but right now there isn’t a “list” of suggested solutions).

2 Likes

This is exactly what I’ve done more then 10 years ago - I took the most voted https://jira.atlassian.com/browse/JRASERVER-1330 and implemented it as an addon. But the guys are right, the biggest challenge is Atlassian. My addon depends heavily on the knowledge of Jira internals and requires patching Jira, so access to the sources is crucial. I was always sort of ready for Atlassian to stop sharing Jira sources. Thankfully this never happened.
But what I was not ready for is cancelling Jira Server completely :sweat_smile:

Answering the questions:

  • Surely yes
  • I believe this should be public
  • Contact with the customers
2 Likes
  • Would you be interested in seeing these? - yes, definitely.
  • How would you like them shared/delivered? - I believe this should be open. To have it in a structured mode to be able to select relevant areas would be great.
  • What context/info could we provide that would be helpful/useful? - User expectations in terms of use case they’re trying to solve. And a clear understanding of Atlassian intentions in this direction :slight_smile:

I’m not sure I get it …
I am able to see the filter results of the links on this dashboard - so they are already out there.

  • Would you be interested in seeing these? : I can see them already - here is my favourite JQL (we :heart_eyes: Bitbucket) project = BSERV AND type = Suggestion AND statusCategory != Done AND labels in (plugin_opportunity) ORDER BY votes DESC
  • How would you like them shared/delivered? : Yes , is there a way to subscribe to a filter other than RSS?
  • What context/info could we provide that would be helpful/useful? : Vendors can contact reporter/commenters via a form.
1 Like

Thanks to all the folks who chimed in. This is all very helpful.

As I mentioned:

:sweat_smile: So I don’t expect us to do anything wild here anytime soon. But all of this feedback has helped us build a case for making sure we have a workflow/plan for all of those we don't plan to do that request. I’m attaching this topic to our internal planning pages around the changes and will certainly get looped back in once folks start making changes. More updates then! In the meantime, definitely share more ideas if you have.


Right, one of the plusses is that they are all already public and you can see what has been voted for, how many folks are commenting, etc.

I think a step further would be getting the PMs to share thoughts on the feature and why it doesn’t belong on their roadmap and trying to create some more context around their thinking on it.

A contractual obligation would certainly remove doubt.

3 Likes