Deployment Frequency

Introduction

Today, there is no successful company without DevOps as an approach to software development and operations. This set of processes allows companies to coordinate development teams within a software project. This enables products to be created and improved in a faster, more agile , flexible and successful way than traditional software development methods.

The goal of DevOps is the awareness of the team to follow a path in order to apply the necessary automation and to implement the automation and methodologies necessary to save time for both, the company and the individual members of the development team. This frees up time from mechanical work and allows team members to spend their time on really important tasks.

DevOps has many benefits, it reduces costs, improves productivity, facilitates scalability, in addition to improving communication between the team and the end customer, better quality software …

But let’s focus on one of the great benefits of DevOps: CI/CD (Continuous Integration/ Continuous delivery). In this software development practice, members of a development team integrate code in the same location, which allows to automate delivery of new versions.

Continuous Deployment is the automation of these continuous deliveries to a production environment. The metric we’re going to talk about is related exactly on how often we do these deployments.

Data-driven Metrics

Measuring the effectiveness of a team within a project is the path to improvement and ultimately success. Data-driven metrics help us to measure and monitor this effectiveness.

There are 4 key interrelated metrics for assessing performance within a team, according to DORA (DevOps Research and Assessment):

• Mean Lead Time for Changes (MLT)
• Deployment frequency (DF)
• Mean Time to Recover (MTTR)
• Change Failure Rate (CFR)

We will focus on the second one, Deployment frequency.

Deployment Frequency

It is one of the main (data-driven) metrics of the Agile methodology and represents one of the main goals of DevOps effectiveness, CI/CD.

This metric function is to measure the frequency which we deploy code to development, test and production environments. The frequency of code deployment can include functional pieces of code (incremental value, new features), enhancements or bugs.

One of the DevOps main goals is to make continuous code deliveries smaller, faster and more testable. Deployment frequency is thus closely related to the DevOps continuous delivery philosophy, hence the importance of this metric.

It also provides an opportunity to understand how well your processes are working. Monitoring it can help us identify wider issues, such as lack of professionals among the team or inefficient processes.

How can the frequency of deployment be improved?

As we have said before, measuring the effectiveness of a team within a project is the path to improvement and ultimately success. If our goal is to be effective and we intend to be successful in a project, we must improve the deployment frequency.

Within a software development team, many actions can be taken to improve this metric.

For example, let’s say we want to get a new feature into production in a mobile application.

If a development team that is working on that functionality does not use agile methodologies, it complicates CI/CD. Deployments are less automated, tasks are not broken down into mvps… and therefore it is much more difficult to improve this metric. We might even think that it doesn’t make much sense to talk about deployment frequency. So, the first step to improve it, or start using it, is to be part of an Agile team.

Let’s go back to the example. Let’s say we are a development team using Agile.

When it comes to breaking down the functionality into tasks, an exhaustive work must be done in the refinement of each one of them. This is where the awareness of the whole team comes in and it must be very clear what we want to achieve and understand the path to follow, in this case, to follow the ways of DevOps with CI/CD.

Therefore, if a team is aware that this is their path, they will be more aware when it comes to refining the tasks. As a result, smaller and smaller tasks will appear from the initial tasks, and each one of them will contribute value, and that is, reaching the minimum value product (MVP).

In this way we get directly, that the general functionality is addressed in a totally particular way with each task, and we can deploy each MVP in production continuously as the project progresses.

This is one of the ways to improve deployment frequency within a software development team.

In addition, another major pillar of CI/CD, which helps to improve deployment frequency, is to have a test suite that therefore helps to improve deployment frequency is to have a reliable test suite. What does this mean? Every new feature, no matter how small must have its unit, integration and functional tests. With this, we ensure that the continuous integration process will have a high reliability and we can be sure that we do not break anything, every time we make a new deployment that we make a new deployment, and at the end of all this, we will be able to make deployments with greater deployments with greater security, reliability and ease.

Accelerate State of DevOps

Each year Google Cloud’s DevOps Research and Assessment (DORA) publishes its “Accelerate State of DevOps” report. This report provides data-driven industry insights that examine the capabilities and practices that drive software delivery, as well as operational and organisational performance.

So, as we can see in the table below, they rank teams based on their software delivery, if we focus on deployment frequency:

• Elite teams perform on-demand code deployments multiple times a day.
• High-level software delivery teams perform deployments once a week or once a month.
• Mid-level once a month or every 6 months.
• And at low levels less than once every 6 months.

Conclusions

Deployment frequency is one of the most important data-driven metrics.

Smaller, continuous daily deliveries mean easier testing, deployment, and rollback if something goes wrong, so the Change Failure Rate and Mean Time to Recover also drop, increasing availability.

A high deployment frequency, daily on-demand deployments, indicates a high level of Agile maturity.

This is closely related to the main objective of DevOps and its respective benefits. Therefore, if a company wants to achieve data-driven goals, it is very important to measure this metric among others.