What do people get wrong when trying to get different teams to run experiments together?
The 2025 Experimentation-led Growth Report is clear: experimentation is at its most effective when it happens across teams.
Shared metrics, shared goals, shared outcomes.
Shifting from being an organization with siloed experimentation to one with unified teams can pose challenges. So we asked the Hall of Fame honorees in our 2025 Who’s Who in Experimentation report what people often get wrong when trying to get different teams to run experiments together.
Here’s what they had to say.
1. Don’t chase perfection straightaway
One mistake companies often make is trying to align everything perfectly and straightaway. Ton Wesseling, the founder of Online Dialogue, notes that you don’t need everything aligned perfectly upfront, and that trying to achieve this often slows down experimentation teams who are trying to improve their programs.
“Too many meetings, not enough risk, not enough pace,” he notes. His advice? “Ensure that each team can run experiments without worrying too much about cooperation, conflicts, and overlaps.”
This makes sense; allowing different teams to run experiments independently while building guardrails for collaboration is one of the best ways to set up a culture of experimentation that delivers business results.
Kameleoon allows each team to experiment using their preferred tools without overstepping or getting in each other’s way.
A good tool is not a substitute for a good strategy
Kameleoon enables experimentation across teams. But without a set strategy, grounded hypotheses, and overarching goals, experimentation is a distraction.
Jonny Longden, Chief Growth Officer at Speero, agrees. According to him, people often “focus on the tool or the tactic of experimentation … rather than the overarching strategic objective and the human decision-making process it's meant to inform.”
Experimentation must be intentional and aligned. It’s easy to get distracted by the allure of quick, easy changes that boost your revenue, but these changes are almost always the result of careful planning and targeted insights.
Most teams are just moving stuff around on a page. They focus on execution, not the solution, or indeed, the problem they're trying to solve. Are you experimenting to try to increase revenue for a certain behavior, or are you genuinely doing it because it's required for evidence? Where does it sit? Marketing versus development? That’s a great indicator of purpose.
Founder, Made With Intent
Kameleoon’s platform is designed to tie experiments to shared goals and give teams strategic visibility into each others’ work.
Fear is a bigger blocker than process
Having the right tools, teams, and strategies is a crucial starting point, but company culture also plays a big role in experimentation across teams. This is especially true when technical and non-technical teams are both involved.
Erin Weigel, author of “Design for Impact: Your Guide to Designing Effective Product Experiments,” describes this as “an approach to testing that’s too academic.” She goes on to tell a cautionary tale about how the wrong culture and poor communication can lead to roadblocks because multiple teams are running experiments:
"When data science teams own the tools and processes, they sometimes unintentionally discourage people from testing with big words, fear of doing it wrong, and over-complicated tooling.”
This is a key idea that gets at the heart of experimentation: the best way to learn is by doing. Embrace the idea that every test is an opportunity for learning and don’t allow fear to become part of your experimentation culture.
Kameleoon is built for both technical and non-technical teams, and applies guardrails that automatically ensure your tests are rigorous and statistically significant.
Prioritize purpose ahead of process
Many Hall of Fame honorees agree: when experimentation scaling fails, it is often because of culture, communication, and misalignment. Tech, tools, and methodologies can only take you so far.
It’s true that great teams test a lot. But those great teams are also testing in a system that allows them to work together, build off of each other’s learnings, and test in the ways that work best for them.
If your teams do run their experiments in silos, ask yourself: do they share their goals? Are their insights connected? What is stopping them from working together?
The tools already exist. What matters is how you use them.
