DevOps is a powerful and crucial combination of software development and software operations that allows organizations like Aviture to deliver better products, faster. As we get closer towards the end of the year, many organizations like ours are taking a step back to look at the data surrounding DevOps from not only our company but others in our space. DevOps Research and Assessment (DORA), a long-running research program recently acquired by Google, has produced the Accelerate State of DevOps report for the past twelve years — gathering data and testimony from over 36,000 technology professionals from around the world. The 2023 State of DevOps report was recently released, and after reading through the 95-page document, I am excited to share some of my takeaways from their major findings this year.

So, what are the key factors that truly make DevOps successful? Let’s start pulling this report apart to identify the current landscape of DevOps globally but also within our company.

Healthy Culture: The Foundation of DevOps Success

Organizations that have some of the most successful and productive DevOps strategies share a common trait: a healthy culture. By prioritizing and creating a collaborative and inclusive environment where everyone is encouraged to contribute, it naturally generates better performance and job satisfaction. To some people, this might sound like a no-brainer — having a solid and inviting culture creates a better place to work, and there is a lot of research that backs this sentiment up. Although, there’s one specific type of organizational culture that shines above the rest: generative culture.

Some of the behaviors seen in generative cultures include high cooperation, sharing risks, and encouraging bridging. This structure is built around trust and improving how effectively information is shared around a team and the greater company. Companies that have successfully implemented a generative culture have a 30% higher organizational performance than those who haven’t.

The success seen from generative culture is something we can speak to personally. Aviture is structured to encourage cross-pollination of ideas across our projects, giving our engineers various opportunities to discuss historical problems and solutions, brainstorm new approaches, and collaborate to solve unique challenges. Castaway Week is another example of encouraging our company to come together, an event where we fly in our remote employees for a week of in-house teambuilding and workshopping.

Ultimately, culture is something that shouldn’t be underestimated. A siloed work environment will slow down operations around the company. Prioritizing knocking down bureaucratic walls first and you’ll start to see improvements.

Don't Let Code Review Put the Brakes on Delivery

A slow code review can have a hefty toll on time when it comes to delivering software. DORA’s results showed that teams with faster reviews saw a whopping 50% higher software delivery performance. This is a big part of why we’ve been promoting practices like pair programming, trunk-based development, and continuous integration for several years now.

When we first started testing out pair programming, there was a bit of hesitation amongst the engineers — to start thinking about not having tasking that they solely owned. As our engineers got more into it, they really liked the fact that there was someone on the line with them that they could bounce ideas and solutions off of, as well as opportunities to provide mentorship in the moment.

The direct result was us seeing faster deliveries, but more importantly, greater value. The software being developed was of higher quality with fewer bugs, and we saw a better dispersion of knowledge amongst the team. I would encourage other organizations to give it a try to see if these approaches have the same positive effect that we’re seeing at Aviture.

AI and DevOps: Is Now the Time?

There was a little call out in the report about the impact of AI on DevOps that I thought was particularly interesting. What DORA found was a slight bump in improvement to metrics like performance and job satisfaction when used by an individual, but at a group level, AI had a net negative impact — creating a substantial decrease to operational performance and a greater sense of burnout as people are struggling to get these tools into their workflows.

Could that change in the future? Absolutely — AI is still an emerging space — but companies are still trying to figure out how to use AI at a team and organizational level and are assessing whether these tools even make sense as a part of their workflows right now.

Like many others in our industry, we’re also testing and figuring out how to implement AI tools into our workflows while keeping a close eye on developing technology and models as they are improving frequently.

Documentation Deserves More Attention

While documentation hasn’t been a focus of the report in the past, there were some important takeaways in this year’s findings. If you focus on ensuring that you have high-quality documentation that is both useful and meaningful, it forces the team to come together to make decisions about the way that they work ahead of time, sparking conversations and getting more people engaged in the process. High-quality documentation can amplify the impact that DevOps has on organizational performance — an estimated 12 to 13 times more impact in fact.

That’s no small change, so how does high-quality documentation lead to a 1200%+ increase? When high-quality documentation is in place, your team knows where to go for small questions that would’ve ended up slowing down other folk’s processes otherwise. If there's a proven way to solve a particular kind of problem, your team is empowered to consume that information themselves, and don’t need to have multiple people in a meeting to have that conversation. Compared to the AI takeaways above, quality documentation had a net positive impact at a group level.

There was one important caveat to the documentation piece, however, which surrounds the distribution of work within teams/companies.

The Importance of Distributing Work Fairly

The findings reveal a significant gender disparity, with women and underrepresented groups reporting higher levels of burnout in their work environments and teams when there's higher levels of quality documentation — and DORA was trying to figure out why that might be. It comes back to an incredibly important detail: distributing work fairly.

What the data suggests is that those same groups are the ones that are allocated more of the tasking to produce this documentation. They’re sharing an unnecessary amount of burden, being assigned between 29% to 40% more repetitive tasks, so that everyone else can benefit, causing them to feel more burnout. If you’re going to invest in producing better documentation, ensure that everyone, regardless of their background, plays a part in its creation.

Increasing Infrastructure Flexibility with the Cloud

We’ve invested a lot in the cloud, and we’re seeing that investment pay off across the industry. Metrics show that at first there’s a dip in personal productivity as you begin working in the cloud, but over time it really starts to accelerate what teams are capable of as they start to see the benefits of flexibility. This translates to a 22% increase in infrastructure flexibility and a 30% higher organizational performance compared to companies not using the cloud.

The cloud continues to be a powerful tool for companies of all sizes, and as an Amazon-certified partner, Aviture remains positioned as experts for our customers.

Balancing Delivery Speed, Operational Performance and User Focus

The report underscores the importance of balance within DevOps. No organization should focus solely on one aspect; it's a balance between technical concerns and user-centric yields the best results. This balance not only results in a superior product but also a more productive workforce. The report identifies four distinct team types that are accounted for: User Centric, Developing, Future Driven, and Balanced — each with its own set of characteristics and performance indicators.

User-Centric Teams

These teams focus most of their effort on meeting user needs. When combined with modern DevOps practices, they consistently achieve high levels of organizational performance. However, an overemphasis on this approach can lead to higher burnout rates, highlighting the importance of balance.

Developing Teams

 Developing teams are a group I wouldn’t actually consider a team yet. They’re still in the process of finding their footing and market fit – something we see happen in startups, small organizations, emerging companies, or a team that just hasn’t graduated to that next step. They often demonstrate lower delivery performance compared to teams with a more user-centric or balanced mindset and report higher levels of burnout than other types of teams.

Feature-Driven Teams

These groups concentrate on adding as many features as quickly as possible, they have a relentless focus on delivery, which in turn can distract the team from meeting users’ needs. Often, this process leads to overlooking the user's needs and a mismatch between product capabilities and user demands.

Balanced Teams

This team is just that — a team that demonstrates a balanced, suitable approach that pulls on the best parts of the previous teams. They use technology in a sustainable way to achieve solid organizational performance, personal performance, and high job satisfaction. These teams also report the lowest level of burnout. This is what we’ve been driving towards at Aviture and what a lot of our talks on things like the product mindset have been focused on.

Tying it All Back Together

So ultimately, why are these findings important? There are a lot of developments happening in not only the DevOps market but the technology space as a whole. Understanding what is and isn’t working across hundreds of thousands of companies helps you and companies like our own position themselves to make important decisions year after year.

At Aviture, we strive to be at the forefront of these developments to deliver faster value and create more reliable products for our customers. The 2023 State of DevOps Report has helped us identify what areas we need to prioritize but also has allowed us to celebrate all the boxes we’ve been checking and exceeding with flying colors.

We’re constantly working to improve our processes, and I hope these findings help to support your own business decisions and pivots as well. After all, teams that learn the most, improve the most. By embracing a healthy culture, fostering a user-centric approach, promoting flexibility, and maintaining a balance in all aspects, organizations can continue to drive innovation, productivity, and success in the world of DevOps and beyond.

I’d love to discuss and analyze the impact your DevOps strategy is having on your organization, and if you need a partner to help identify your problem areas — contact us today so we can build something great together.