This Fire Is Out Of Control

Sometimes, despite all of your best efforts, your team combusts.

Not literally of course, but when things aren't going well, it's common to say that everything is on fire.

An unhealthy team is terrible for the organisation, because it represents a growing risk against whatever the team is responsible for.

It's even worse for the people in the team, because an unhealthy team will slowly but surely chip away at all that is good and right in the world, leading to dissatisfaction and exhaustion.

So, when you see a fire, you really should try to put it out.

Now There Is A Fire In Me

Not too long ago, my team was on fire.

Our operational load was high, and the effort required to keep up with it was exhausting the team and making it difficult to keep up with all of the other work that we needed or wanted to do.

Speaking of which, lots of other teams and business units were asking things of us, which made it difficult to prioritise maintenance and improvement work. This fed back into the first problem (high operational load) in a wonderfully negative feedback loop.

Finally, our services and processes were not as reliable as they needed to be, which also contributed to the high operational load. They weren't bad enough to be non-functional, that would have actually been better, because we would have stopped to fix them. They were just bad enough to constantly cause a little bit of extra pain on a week-by-week basis.

Now, the fire had been smouldering away for a while and we'd been doing a variety of things to try and manage it. For example, we had reduced the number of things we were tackling in parallel and increased the amount of effort allocated to our Business as Usual (BAU) pipeline (i.e. non-project work).

It helped, but it wasn't enough.

At the end of last year, when we looked forward and saw that the asks of the organisation would increase in both quantity and importance, we knew we needed to do something drastic.

We needed to pull the fire alarm.

A Fire That Burns

Atlassian has a concept called the X-Alarm Fire (XAF).

It's very similar to an actual fire alarm, being a thing that you trigger in order to provoke an immediate response. Ideally before said fire burns out of control and does a bunch of damage.

X-Alarm Fires come in 5 delicious flavours:

• 1-Alarm Fire (1AF) - A part of a single team is focusing on the fire
• 2-Alarm Fire (2AF) - A whole team is focusing on the fire
• 3-Alarm Fire (3AF) - A set of teams are focusing on the fire
• 4-Alarm Fire (4AF) - An organisational unit is focusing on the fire
• 5-Alarm Fire (5AF) - The entire organisation is focusing on the fire

When you call an XAF, you're essentially saying:

All work that is unrelated to putting the fire out is stopping. Said work will not resume until the fire is extinguished to the our satisfaction.

It's a clear and unambiguous signal to the organisation that there is a problem, and that any previous organisational priorities are temporarily suspended until the problem is resolved.

As you can imagine, that's a pretty big decision to make, so it comes with one big condition; you must provide a clear set of exit criteria.

These criteria are generally measures of some sort, and you are required to track and regularly report on them, so that interested parties can understand the state of the fire and whether or not you are on-track to extinguish it. They might be quantitative, or they might be qualitative, but you need something.

So, with some light prompting from those around me, in late 2022, I called a 2-Alarm Fire (2AF) for my team.

I'm Gonna Burn This City

I'm not going to go into too much detail about the process of calling the actual 2AF, but I will say that I did a bunch of analysis, surveyed the team, decided on some exit criteria, wrote it all up and published it internally.

There was some pushback from our stakeholders, who were worried about how our 2AF would affect their priorities, but one of the beautiful things about this mechanism is that people understand and respect it.

Looking back now, I can say that calling a 2AF provided a number of benefits.

The first benefit was a quarter of focus. In the end we used that quarter as best we could, but I wouldn't say it was a smashing success. We had some non-discretionary work that we needed to do which complicated things, but we did make a positive impact on a few of our exit criteria.

The second benefit was increased engagement and morale. Acknowledging the problems that the team was experiencing and prioritizing solutions to those problems reinforced that we actually did care about the experience of the people in the team.

The third benefit was that we got additional funding. I wasn't necessarily expecting this as an outcome, but by being very clear with the organisation about the trouble we were in and the risks that imposed on other initiatives, they were able to make decisions about people that were more in our favour.

A single quarter wasn't enough to fix all of the issues that led to the fire in the first place, so at the end of March 2023, we did a bit of a re-evaluation and decided that yes, we're still on fire, but we're getting better. Since then, we've started executing on some external asks, but we're still allocating at least 60% of our effort to fiery projects.

Our current projections say that we will have extinguished the fire by the end of this quarter (aka June 2023).

Moving forward, we'll aim to better balance external asks against internal maintenance and improvement initiatives. I'm thinking something like a 40/60 (internal/external) ratio is probably a good balance to aim for. We might not hit it exactly every quarter, but we can certainly try.

Oh, How I Burn For You

The concept of an X-Alarm Fire is a useful one for communicating to those around you that you need to stop, take a breath, and fix some problems before everything comes crashing down around you.

If you find yourself in a similar situation, regardless of whether or not you work at Atlassian, you should definitely consider pulling the fire alarm.

It's all well and good to continue to soldier on and suffer, but it's much better for the business and for the people in your team if you take a stand and fight for a more sustainable state of affairs.

If I've learned anything, it's that I probably should have done this sooner.