(Editor’s note: This post is the second in a four part series that \u00a0that discusses experimentation at GoDaddy. <\/em>You can read part one\u00a0here<\/a>.)<\/em><\/em><\/p>\n GoDaddy teams ran a record-breaking 2,000+ A\/B experiments in 2024, more than in any year before. In our first post in this four part series,\u00a0we looked at how A\/B testing changed the way we build products<\/a>\u00a0– moving from opinions to evidence. But what really surprised us wasn\u2019t just better outcomes – it was better teamwork.<\/p>\n Experimentation didn\u2019t just change what we shipped, it changed how we worked. It brought product, design, engineering, and analysts into closer collaboration, created a shared language, and gave everyone a common purpose: learn fast, together.<\/p>\n This article explores how A\/B testing became a unifying force, reshaping our teams and laying the foundation for a lasting culture of experimentation.<\/p>\n The transformation started at the top. Leadership committed to a culture where learning matters more than certainty. Instead of long roadmaps built on assumptions, we began treating each initiative as a hypothesis – something to test, not just execute. As\u00a0GoDaddy CTO Charles Beadnall explained it<\/a>:<\/p>\n We, in some ways, acknowledge that we don\u2019t know the future, and so we are going to experiment our way into that.<\/p>\n<\/div>\n<\/blockquote>\n A failed test isn\u2019t wasted work, it\u2019s a new data point. After every test, we ask:<\/p>\n This approach builds psychological safety. Teams feel free to experiment boldly, knowing they\u2019ll be celebrated for the insight, not just the win (though we avoid framing outcomes as wins or losses). Leadership reinforces that every result is a learning opportunity, keeping teams motivated and open to feedback.<\/p>\n At GoDaddy, A\/B experimentation is driven by passionate people focused on solving customer problems. These experiments can originate from anywhere in the organization, especially from frontline insights. When a promising hypothesis emerges, we bring together a cross-functional \u201cexperiment squad\u201d to explore it. These squads rally around the hypothesis like a shared mission, then disband after the question is answered. It\u2019s not uncommon for the squad to stay together through a series of experiments as they develop a deeper knowledge. Each squad includes:<\/p>\n This model makes experimentation fast, focused, and collaborative. It also builds networks across the org \u2013\u00a0connections that carry into future work and speed up future learning<\/a>.<\/p>\n The following diagram depicts a typical cross-functional experimentation squad:<\/p>\n A shared language helps everyone stay on the same page. But with a company as large as GoDaddy, acronyms and team, product, and organization-specific terminology can provide challenges to alignment. Since these squads work closely together, sometimes over multiple experiments, this shared language is built quickly. When a product manager says “iGCR” (incremental gross cash receipt), the whole squad knows what that means. This kind of alignment shows up in other ways, too, like having a common understanding of:<\/p>\n This shared fluency helps teams move faster and make decisions with confidence.<\/p>\n To scale experimentation, we invested in the right culture, processes, and tools \u2014 across three levels of the organization:<\/p>\n Leadership sets the tone by valuing learning over certainty. Regular company-wide showcases celebrate top experiments (regardless of outcome), reinforcing that every outcome has value. Learning happens not just through training programs and playbooks, but also from each other, through shared stories, informal exchanges, and everyday collaboration. Best-practice rituals help build a consistent, data-driven mindset across the company. We make learning enjoyable because growth sticks when it\u2019s engaging.<\/p>\n Experimentation has become a shared pursuit of understanding. We show up at the whiteboard with purpose and curiosity, co-owning ideas and staying engaged throughout. It\u2019s deepening our insights and strengthening how we grow together. – Fiona Guan, Business Analytics Analyst, GoDaddy<\/p>\n<\/div>\n<\/blockquote>\n GoDaddy leans on videos by Daniel Hedblom (Dr. Dan) to provide entertaining and informative snippets around experimentation that are easily digestible for folks who might not yet be as familiar with experimentation best practices. The following image shows\u00a0Big Swings with Dr. Dan<\/a>:<\/p>\n Teams use a shared language, consistent tools, and a unified metrics framework. Experiments are tied to company-wide\u00a0Objectives and Key Results (OKRs)<\/a>\u00a0and\u00a0Opportunity Solution Trees (OSTs)<\/a>, ensuring that insights don\u2019t stay isolated. Knowledge is reused and shared in visible places.<\/p>\n Experimentation is a natural extension of the design process. With each test, we\u2019re just not validating ideas of improvement, we\u2019re learning more about our users and listening to what they are saying. Inside this process we test and refine, always with the goal of creating thoughtful, user-centered experiences. – Laura Oggioni, UX Designer, GoDaddy<\/p>\n<\/div>\n<\/blockquote>\n All teams are required to report on the outcomes of their experiments in a standardized way to make sure insights are accessible, comparable, and easily understood across teams, regardless of the specific context or product surface. The following image shows a simplified version of the GoDaddy experimentation template:<\/p>\n Product squads drive their own experiments. They move fast, test frequently, and learn in short cycles. Tools like\u00a0Hivemind<\/a>, our internal experimentation platform, provide everything they need (test setup, randomization, data integrity, and real-time results) so they can focus on learning, not logistics.<\/p>\n Engineering through experimentation has transformed how we build software. It’s about intentional design and validated decisions. Every feature we develop is a hypothesis we can test, measure, and learn from. This data-driven approach has made us more precise in our solutions and more impactful in our delivery. – Sandeep Ramamoorthy, Software Development Engineer, GoDaddy<\/p>\n<\/div>\n<\/blockquote>\n For example, Hivemind includes a wide range of capabilities and built-in guardrails that help teams apply consistent experimentation standards, like a Minimum Detectable Effect (MDE) calculator, to properly estimate the required sample size and test duration before launching an experiment. The image below shows GoDaddy\u2019s internal experimentation platform Hivemind:<\/p>\n The result? Hundreds of experiments run simultaneously, all grounded in shared strategy and systems.<\/p>\n The following sections explain a couple of the challenges we faced building our culture of experimentation.<\/p>\n Early on, teams ran A\/B tests independently, meaning learnings stayed isolated. That led to duplicated work and missed opportunities to build on each other\u2019s results. To change that, we created team-level and company-level operating mechanisms to break down silos and create deeper collaborations. Not only within squads, but also across teams with shared dependencies and goals. Aligning experiments around OKRs or key initiatives and clearly stated impact helped prioritize requests and work more effectively within other teams\u2019 roadmaps and bandwidth. These habits made it easier to spread desired patterns, reuse ideas, and avoid common mistakes. Over time, this built trust and helped us move faster as a company.<\/p>\n As experimentation scaled, we saw a wide range in test quality. Some had clear goals and strong design. Others didn\u2019t. Our rule of thumb is to prioritize quantity first to build the experimentation muscle, then improve the quality for greater impact. We wanted to keep up the pace, but also raise the bar. So we introduced lightweight peer reviews and a shared set of criteria for what \u201cgood\u201d looks like. It wasn\u2019t about policing quality, but to help experimenters learn and level up. Each team keeps a mix of fast, low-effort tests and slower, high-impact ones in their backlog. Over time, this helped teams run better tests with more clarity and shared standards.<\/p>\n A\/B testing has evolved from a validation tool to the backbone of GoDaddy\u2019s product development process. But beyond the tests themselves, the experiment-first mindset has transformed how we work together. We\u2019ve built a culture where teams move faster by learning together, and the results show.<\/p>\n In 2024 alone, GoDaddy teams ran over 2,000 controlled experiments<\/a> across our products and customer experiences. Since Aman Bhutani became CEO in 2019 he has championed a culture of experimentation and contributed to $1.6 billion in revenue\u00a0growth<\/a>.<\/p>\nShift in mindset<\/h2>\n
\n
\n
Building the squad model<\/h2>\n
\n
![\"Diagram](\"https:\/\/www.godaddy.com\/resources\/wp-content\/uploads\/2025\/06\/experimentation_squad.jpg?size=3840x0\"\/)
<\/figure>\n<\/div>\n\n
Scaling with systems<\/h2>\n
1. Company-wide culture<\/h3>\n
\n
![\"Big](\"https:\/\/www.godaddy.com\/resources\/wp-content\/uploads\/2025\/06\/dr_dan.jpg?size=3840x0\"\/)
<\/figure>\n<\/div>\n2. Cross-team alignment<\/h3>\n
\n
![\"Image](\"https:\/\/www.godaddy.com\/resources\/wp-content\/uploads\/2025\/06\/gd_experiment_slide_template.jpg?size=3840x0\"\/)
<\/figure>\n<\/div>\n3. Individual team execution<\/h3>\n
\n
![\"Image](\"https:\/\/www.godaddy.com\/resources\/wp-content\/uploads\/2025\/06\/gd_hivemind.jpg?size=3840x0\"\/)
<\/figure>\n<\/div>\nChallenges<\/h2>\n
Making collaboration consistent<\/h3>\n
Balancing speed and quality<\/h3>\n
Conclusion<\/h2>\n
![\"Image](\"https:\/\/www.godaddy.com\/resources\/wp-content\/uploads\/2025\/06\/gd_sustainability_2024.jpg?size=3840x0\"\/)
<\/figure>\n<\/div>\n