Lean practice.Performance optimization.Digital Transformation.
There’s something else that is beginning to be oft-sought-after like this: DORA metrics.
Did that make you think of that cartoon girl with a pink T-shirt? Maybe you already knew about DORA metrics, so it didn’t. Or maybe you aren’t that well-versed with DevOps, so it did.
Well, we breathe(d) DevOps and that was still what came to our minds when we first came across the term xD
So either way — DORA metrics are related to DevOps — now you know. And not just related, they are essential to move towards the three aspects mentioned at the beginning of this post. How so, you might ask?
Hm, let’s see.
Before that, what exactly is DevOps?
Background: DevOps
For the uninitiated, ‘DevOps’ is an amalgamation of ‘Development’ and ‘Operations’, literally and functionally. In simple terms, it refers to creating and delivering software applications and services. Good DevOps delivers value to its end users and is speedy and stable.
Can speed and stability be achieved simultaneously?
Here is the relevance of DORA metrics.
Back to DORA
To start, DORA makes the results of DevOps processes tangible and visible. Noticing something is the prerequisite for improving it. How will you better something that you can’t see or measure? Progress begins with acknowledging the present state, and understanding where you’re starting from and where you want to go.
DORA’s research shows that it is possible to optimise for stability without sacrificing speed.
There are four DORA metrics; their names are pretty self-explanatory:
- Deployment Frequency (How often the organization successfully releases to production)
- Lead Time for Changes (Time taken for changes to get into production)
- Change Failure Rate (Percentage of deployments causing a failure in production), and
- Mean Time to Recover (Time taken for the organization to recover from a failure in production)
Observing the nature of DevOps processes in an organization, each of these parameters is given a score. Depending on the score for each of these metrics, organizations can be bucketed into ‘elite’, ‘high’, ‘medium’, and ‘low’ performers.
But it’s not like a school report card where you get grades at the end of the year, get a medal (or none), and what’s done is done.
While the tag of ‘elite’ or ‘low’ serves as a good indicator of organizational performance, especially in comparison with competitors, their purpose is to reflect the current state of DevOps SO THAT that state can be improved.
This data from DORA metrics serves as a base for you to think of measures to become better at delivering value, improving the effort-to-output ratio, and aligning your processes with the vision that you work towards. Each team has its own aims, visions, and business values that it works to deliver. Don’t mistake DORA metrics to be ends and aims in themselves.
‘When a measure becomes a target, it ceases to be a good measure’.
Apart from this, DORA inherently also has a few challenges.
Challenges
These are easy to find. Just Google ‘DORA challenges’ and go to ‘Images’.Very simple challenges, right?And you thought DevOps could be difficult…
Joking. Keeping aside the fact that the challenges of the cartoon DORA are pictorially evidently more popular than DevOps DORA’s.
The data that DORA metrics intend to measure is scattered across varied departments and stages of DevOps.
Additionally, that data needs to be presented WITH proper context for it to make sense. For example, what will you think about seeing ‘40 deployments’ on the metrics dashboard?
40 deployments in what? One second? One month? Year?Deployments of what? Which project? How big? What if you want DORA metrics on a specific project?
This logistical barrier, once crossed, can be the beginning of your experience with DORA.
Closing remarks
We had mentioned three terms at the beginning of this post: Lean practice, performance optimization, and digital transformation. With context about DORA metrics now, you can see how it matters to each of them, which warrants posts of their own.
(No, we didn’t just assume that you would be familiar with them. There are some links at the end for those of you that aren’t.)
From now on, hopefully, a cartoon girl with a pink T-shirt won’t be the only thought you have when you come across ‘DORA’.
Read more about…
Calculating DORA metrics here and here
Doing DevOps here
Challenges that engineering leaders face, here
