Measuring the Impact of AI Coding Assistants

We’re sick of hearing these outlandish, unfounded misconceptions about the implications and the impacts of generative AI on the software development profession, and you probably are, too. So we wrote Measuring the Impact of AI Code Assistants to compare the hype and the reality.

It’s clear that AI tools augment, automate, and accelerate software development tasks in a way that can feel like tapping into “dev beast mode.” Experiencing that whole new level of flow has excited the development community and propelled a wave of AI adoption and retooling for developers — a change at least on par with the shift to agile and cloud development a decade ago.

Software engineering and product development have truly “eaten the world,” as Marc Andreessen predicted, and software development teams are now a crucial function of every business. In even the most traditional of firms, the software engineering teams are quite large, with financial services institutions like JPMorgan Chase having more than 50,000 technologists and retailers like Walmart employing more than 20,000 software developers.

Software engineering teams represent one of the largest and most costly departments in enterprises. At the start of 2024, the average salary of a software engineer in the US is $118,000. With large team sizes, annual costs run into billions. So helping those developers be as productive and efficient as possible is crucial. For software engineering departments of any size, even small increases in developer productivity can have enormous impacts on delivery time, quality, and developer happiness, all of which help enterprises be more competitive, create better products, and retain their talent longer.

Numerous independent and vendor-generated studies show that developers are more satisfied and can be as much as 45% more productive with AI coding assistants. But you may be wondering, “How much productivity can my team specifically expect from AI? What’s the measurable business impact of adopting these tools for our situation?”

This report provides answers to both questions through a grounded, balanced, and data-driven approach. It aggregates and analyzes the most significant research available and provides a simple, logical framework for you to calculate the value for your software engineering team specifically.