How to scale Agile marketing with Jira Software and Structure PPM

[#1]

This is an excerpt from a speech I delivered at the Jira Day 2023 conference in Krakow, Poland.

Read the story of Agile marketing implementation at Deviniti for more historical context.

---

Within 3 years, we grew as a marketing department and matured our project management approach.

Here we are - double the size, with a lot of experience under the belt, organized to deliver results together with our stakeholders. Step by step, we went our way from request-based planning and sprint-based execution to a hybrid project management approach, tailored to business needs and marketing processes.

We took the best from both worlds - Waterfall and Agile - and made them work together. The next step was to adjust the tools.

And there was much work to do in this regard.

The project documentation was everywhere across Confluence and Google Sheets. We had to update all the plans and roadmaps manually, create every task during sprint planning, and merge a couple of spreadsheets every month to properly report the cost of labor.

This could easily be a more productive time at work.

The way our Jira project was built didn’t suit our operations anymore. We needed a more refined workflow and issue structure, as well as a couple of new features that would make our lives a lot easier and more transparent.

So we asked our almighty Atlassian Services team for help once again. After hours of consultations, they offered us an improved solution for marketing project portfolio management that exactly mirrored our existing processes.

Deviniti's marketing project portfolio management setup

This is our current basic project tool stack. We call it ‚the trio’ because it serves 3 basic purposes:

• planning and tracking;
• doing and storing the actual work;
• and fostering ongoing communication.

Of course, there are many more tools used in modern marketing, but let's take a closer look at the Atlassian part of things.

One Structure to rule 'em all

Previously, our Epics tended to grow way too much and become senseless sacks of hundreds of tasks. Now, we’ve set up a 4-level issue type structure:

1. Projects and Standing Orders
2. Stages
3. Epics
4. Tasks and Stories

A Standing Order has no End date - an example is the ongoing maintenance of our CRM system.

A Stage can be a literal stage of the project (like the promotion of an event) or a timeframe, like a quarter.

An Epic is a full single deliverable, like a landing page.

We don't use the default Scrum definition of a Story. For us, it’s a Story because it's neverending (like daily standups or weekly synchro meetings with a stakeholder).

Because they are never Done, we can exclude them from sprint reports and not overload Jira with issue volume by recreating them every sprint.

To visualize and manipulate this structure, we used the Structure for Jira app, currently owned by Tempo Software.

While roadmaps and ideas still exist elsewhere, the actual planned work is seen here along a Gantt timeline. The customizable columns show the status and the project cost in real time. With transformations, we make it easier to further break down our projects, and find unplanned work or issues with incorrectly filled reporting custom fields.

We configured the Structure based on custom Jira issue links and added a couple of interesting quirks to the view.

Displaying linked Confluence pages on the same screen is one of the coolest features I've ever used! Now we have our drafts and ready deliverables under each Epic, so the review process is seamless. We can also edit these materials right away, and it respects the space permissions set up in Confluence.

Jira project configuration

The new workflow looks like this:

We've added a separate status for new tasks created in advance that haven’t been planned yet. Also, the On Hold status is a game changer, because we have so many dependencies and often have to wait for something or someone.

To make new issue creation faster, we added a workflow screen that allows us to lay out the basic fields and then transitions the issue back to the current status.

Then we've added a couple of more custom fields, useful for grouping tasks and reporting.

• Beneficiary and Cost type (Select lists) help us break down the cost of labor and properly track billable utilization which is one of our KPIs.
• Stakeholders and Consultants (User pickers) are used for notifying relevant people about the latest changes and informing the assignee about who they can contact regarding the particular task.

We’ve integrated our external partners and collaborators tighter into our processes as well. We have a group picker that enables us to share only the selected issues in the project. These user groups also belong to corresponding Tempo teams so they can estimate and log time just like us.

While they have an account in our Jira, we can also give them limited Confluence access to work together. We also maintain ongoing communication on Slack using Connect, so both project channels and DMs are available to us.

Zoom in: sprint planning and reporting

With Structure for Jira, we can group Epics per sprint. This means no more Confluence tables and Google Sheets for planning!

We release a typical Epic within 1-2 sprints, so we plan it as a whole or split it in half. We don't plan the Stories, though, as they never close. Instead, 20% of our time is always allocated to them as a buffer.

We’ve started estimating Epics in working hours, for which we collected reference historical data from Tempo Timesheets for similar Epics.

For optimal workload, we check the sum of estimated time per assignee and daily distribution on the Resource Usage view. At the end of the sprint, we verify the estimates vs. the actual time spent and talk through them during the retro.

To make issue management even faster, we started automating the most tedious actions, most notably status change which is a nightmare at scale when done manually.

And when it comes to reporting:

• Both project and sprint views are public so our stakeholders can take a look at the work in progress anytime. 
• We've set up a custom time and cost dashboard in Tableau, which aggregates data from Tempo Timesheets each month automatically. We don't have to even think about reporting it! This is the favorite feature of our CMO who can forget about merging multiple Excel spreadsheets anymore.

Project portfolio management implementation results

What have we achieved with the new toolset?

We know exactly how much work we have at hand, which is 44 projects for a 20-something team. 13 of them are already finished.

We saw 20% billable utilization growth and shortened delivery times by 59% on average almost instantly, but this time we didn’t change the way we work to achieve this goal.

Instead, we changed the way to track and measure it.

For our CMO, the best things about the new Jira are:

• the ability to zoom in and out on the project portfolio;
• ongoing status and cost control without asking the teams;
• the automatic reporting.

The Project Managers and Coordinators in our department highlighted the increased flexibility, project transparency, and ease of use at scale.

During this project, we learned that Jira Software can be a surprisingly good fit for non-technical teams. It's most beneficial when the team is scaling up, the processes are complex, and those who we collaborate with also work in the same environment.

---

We couldn’t get this thing right without the help of the Deviniti Atlassian Services team, who are always helpful and responsive. This time, we were the internal customers, and we can recommend these guys with confidence! Check out what else they can do for you.

Have a similar story or challenge? Share your experience in the thread, or find me on LinkedIn to connect.