How to Evaluate Jira Apps: Advice for Solution-Hunting

[#1]

Nightmare scenario: Picking a Jira app, getting your team onboarded, and then realizing that it doesn't solve your problem. On the other hand, you might end up being overwhelmed with Atlassian's options and never end up picking something in the first place. Here's a suggested process — based on conversations with Jira admins and users — for app vetting, testing, and selecting solutions.

---

When you're struggling to accomplish something you want to do in Jira, you go to the obvious place: The Atlassian Marketplace. But it's teeming with options, and you can spend hours browsing — not great when you have more important things to do.

Here's some suggestions for an efficient, effective process that lands you with the right solution.

Define your problem & get stakeholder buy-in

Before you find the right solution, you have to define the problem. This step is important to provide clarity — both for yourself as you search, and to other stakeholders who may not all be on the same page.

Even if you can't answer all these questions in their entirety just yet, it's important for you and your fellow stakeholders to start by asking:

• What problem are we looking to solve?
• Who does the problem affect?
• Would this just be a tooling change, or a process and administrative one as well?
• What parts of the existing system will play a role?
• Will other systems be impacted?
• What are we measuring after we introduce these changes?
• What is our approximate timeline for finding and implementing a solution?

It sounds like a pain to go through all these questions, but it's an important early part of the process. Putting in the time now will save you a lot of hassle later.

Create your shortlist of apps or solutions

Spend a little time perusing the marketplace category you want to explore, and search (or ask) here on the Atlassian Community to see where the topics/apps you have in mind have been discussed. This website contains a wealth of information from your peers: A question like "Can you do X with Y app?" from previous posts — or that you ask on the forum — can serve up answers. A quick check of app reviews can provide red flags: A review that says "It can't do X" or "doesn't integrate with Y" will also save you from going down a dead end.

Lean on your Atlassian Solution Partner as well. Discuss your use-case and ask about apps that you're considering. If they have experience with your app, they can advise you on whether to go with it, and even help with onboarding.

Finally, it’s often a good idea to reach out to app vendors you already trust and simply ask them for recommendations — they may know which other apps play well with your current setup and which ones may not, even if your problem is something they don't deal with directly. Take it from someone who knows: Vendors like getting these types of questions — they're a chance for us to prove how useful we are!

Start digging into app(s) details

Buying an app is a lot more like starting a relationship than many people realize, so make sure the app vendor won't desert you in your hour of need. Clear communication and being supported are important pillars for success.

See how active the vendor is in the Atlassian ecosystem. What kind of onboarding materials exist? Do they have governance/permissioning recommendations? Do they have badges that indicate they are constantly working on the app’s reliability and performance?

If you don’t feel like you have all the answers you want, submit a support request to the app, asking about the support they offer and how you can best evaluate their product. Odds are they’ll be happy to answer all your questions, but if you don’t hear back in a week… well, maybe they’re just not that into you.

Guard against unintended consequences

Before you add an app to your Jira instance, there are a few key steps to follow:  

• Ask "What changes will this app (or apps) make?" Some apps add new data fields, issue types, and/or link types that persist after you remove the evaluation version of the app. Make sure these changes are acceptable before you start an evaluation, and that you're comfortable reverting those changes if you decide not to proceed with that app.
  
  
• Install the apps in a test environment. While there likely won't be any issues if you directly evaluate in production, no one wants to be the (un)lucky 1 percent that impacts a mission-critical system when testing out a new app. (Note for those who are testing multiple solutions: While certain tests are best run in isolation, we generally recommend trying all your possible apps at the same time. Dedicate time to figuring out how they do things differently, or maybe how they could work in tandem to solve your problem – doing this allows you to compare and contrast apps more easily).
  
  
• Invite the end users to test out apps in a production-like environment. As all of us who have ever built software know, user interaction always leads to unexpected effects. As they test out the app and start to impact adjacent systems, more usage questions will arise, ranging from the simple (“what does this button do?”) to the incredibly complex. Again, loop in the vendor at this stage. They can answer all those questions as well as highlight any inefficient behaviors your users are exhibiting. With their help, you can skip the "trial and error" phase of evaluation by getting a quick primer.

If it all looks good, congratulations, you have a winner!

In all cases, remember what we said about app selection being like starting a relationship: The marketplace vendors are there to help you, so don't hesitate to ask if you want to answer a nagging difficulty or even suggest a new use-case and see if they have any pointers for you.

So, head on over to the Atlassian Marketplace to find the right apps. In addition to the search field, check out options on the left that enable you to specify:

1. That you are looking for a Jira app
2. That you are looking for cloud or data center apps (leave it blank if you want to cover both cases)
3. How to sort the list
4. Which filters you wish to use (check all that apply)
5. And the various categories you want to use to narrow the search (if any)

Happy hunting!

[Mykenna Cepek]

I find that doing a Proof Of Concept (POC) can be vital to selecting Apps which meet your needs.

A POC is usually focused on showing basic functionality and confirming requirements and assumptions. A POC should be very limited in time (as brief as possible), and focused on the bare minimum effort to gain confidence that the rest of the solution could be built-out. POCs are not pretty, they are minimally functional. Think of them as quick experiments.

You can use most Atlassian Marketplace apps for FREE for a few weeks or even a month. Usually this is plenty of time to complete a POC. Create a test project or space, and/or use a Sandbox.

The hands-on effort with a POC is invaluable for decision making. Were there problems installing it? Were you able to configure it as needed? Were there additional handy features discovered, or required features found to be missing? Walk through the MVP use cases and see what you learn.

If multiple marketplace Apps are being evaluated, a Confluence page can be super helpful to outline the requirements and use case results during the POC for each app.

I highly recommend using a Sandbox if possible, and starting fresh for each POC (with a new Sandbox for each App). I discovered the hard way that uninstalling or abandoning an App can leave changes in Jira (for example) which linger. You don't need that cruft in your production instances.

[Dave Rosenlund]

Thanks for sharing your thoughts, @Mykenna Cepek  👍🏻

[Fabian Lim]

Great article Dave!

[Astrid Sieben]

A great article @Dave Rosenlund