AI Is Making Us Faster. So Why Aren’t Businesses More Productive?

Paid partnership with Atlassian

AI is becoming remarkably good at helping us complete individual tasks faster. AI can now write code, generate reports, analyze data, automate workflows and complete tasks that previously took hours in a matter of seconds.

Surely, this should mean we’re all becoming dramatically more productive?

Well, not quite.

Atlassian’s new State of Developer Experience 2026 report, based on a global survey of more than 1,560 developers and engineering leaders, reveals a fascinating contradiction.

87% of developers say AI helps them write code faster, yet only 28% report experiencing less friction in their work.

For me, this highlights one of the biggest challenges facing businesses today. We’re getting very good at making individual tasks faster, but we’re much less successful at transforming how entire organizations operate.

And this research reveals four important lessons that extend far beyond software development.

1. The AI Productivity Paradox

Let's start with the good news. AI is clearly making developers faster.

But the report reveals that many of the problems slowing them down haven't disappeared.

Developers still struggle with finding information (51%), meetings (50%), switching between tools (45%), technical debt (44%) and unclear requirements (40%).

In fact, only 26% of developers now report saving more than ten hours per week using AI, compared with 68% in Atlassian's 2025 research.

This reinforces something I've been arguing for years: Simply giving employees powerful AI tools won't automatically transform productivity.

Imagine using AI to write a complex piece of software in minutes, only to spend the next three days waiting for approvals, searching for documentation and navigating internal bureaucracy.

It's like putting a Ferrari engine into a car with square wheels. Impressive horsepower, but you're unlikely to win any races.

I believe the real opportunity lies in fundamentally rethinking business processes, connecting fragmented systems and redesigning workflows around what AI can now do.

And this applies just as much to marketing, finance, HR and customer service as it does to software development.

2. Are We Creating A Culture Of Performative AI?

Perhaps the most worrying finding is what Atlassian calls "performative AI."

According to the research, 60% of developers say AI usage is treated as a signal of productivity, while 41% admit to using AI more than necessary simply because it's expected.

Meanwhile, 59% of engineering leaders are concerned about performative AI, and 81% agree that measuring AI usage can create bad incentives.

I find this particularly frustrating because it reminds me of some of the worst habits of traditional corporate management.

We introduce a new technology, create dashboards to track adoption and congratulate ourselves when the numbers go up.

But does any of this mean we're delivering more value?

If employees start using AI simply to demonstrate they're embracing innovation, we've completely missed the point.

And measuring productivity by counting AI prompts makes about as much sense as measuring intelligence by counting emails.

We should be measuring business outcomes, not technology consumption.

Are we delivering better products? Improving quality? Reducing costs? Creating happier customers and employees?

Those are the metrics that matter.

3. Giving Everyone AI Tools Isn't An AI Strategy

Another striking finding is that only 17% of developers and 23% of engineering leaders say their organization has a clear, active AI strategy across the software development lifecycle.

And the difference appears significant.

Developers in organizations with a clear AI strategy are nearly twice as likely to report noticeably less friction, 41% compared with 22%.

Of course, correlation doesn't automatically mean causation, but the findings reinforce an important point.

Buying AI subscriptions for everyone is a procurement decision, not an AI strategy.

A meaningful strategy requires understanding where AI can create genuine business value and then redesigning processes, systems and ways of working to capture that value.

In software development, Atlassian advocates moving toward an AI-native software development lifecycle (SDLC).

This means looking beyond individual coding assistants and using AI across the entire development process, from requirements and planning to coding, testing, deployment and maintenance.

I believe this is exactly how businesses should approach AI transformation more broadly.

Instead of asking which individual tasks we can automate, we should be asking: If we were designing this entire process from scratch today, knowing what AI can do, would we build it the same way?

In most cases, I suspect the answer would be a resounding no.

And as agentic AI becomes increasingly capable of executing entire workflows autonomously, this question will become even more important.

4. AI Is Changing The Role Of Humans

The final finding highlights another important transformation.

As AI takes on more coding responsibilities, developers are increasingly becoming architects, problem-solvers and supervisors of AI-generated work.

According to Atlassian, 75% of developers report greater responsibility for validating the safety and quality of AI outputs.

And that responsibility comes with challenges.

Some 65% spend time correcting AI-generated work that missed the original intent, while 49% spend at least four hours every week managing AI tools and reviewing code.

This is a fascinating development because it challenges the simplistic assumption that automating work always reduces human effort.

Sometimes AI creates entirely new work that needs to be managed.

We've all experienced AI producing something that looks impressive until we examine it closely and discover that it's confidently misunderstood what we wanted.

Suddenly, the time savings don't look quite so spectacular.

But I see this as part of a much bigger shift in the nature of work.

As AI becomes increasingly capable, human skills such as critical thinking, judgement, creativity, communication and systems thinking will become more important.

We'll increasingly move from performing individual tasks to defining objectives, orchestrating AI systems and evaluating outcomes.

And I think preparing people for that transition should be a priority for every business leader.

The Real AI Transformation Is Just Beginning

What I find most interesting about Atlassian's research is how clearly it reflects the next phase of enterprise AI adoption.

We've spent years being amazed by what AI can do. Now comes the harder task of making it deliver meaningful results.

That means rethinking workflows, developing coherent strategies, measuring genuine business value and ensuring people have the skills to work effectively alongside increasingly intelligent machines.

We can make individual tasks ten times faster with AI, but unless we fix the processes connecting them, we'll simply create bottlenecks at record speed.

Atlassian's research offers valuable insights into the challenges and opportunities ahead.

You can explore the full State of Developer Experience 2026 report here.

And I'd love to hear your perspective. Is AI genuinely making your organization more productive, or are you simply getting faster at doing things that perhaps shouldn't be done in the first place?