Is silo mentality stifling your experimentation success?
Is your experimentation program experiencing push-back from other departments? Marketers and designers who own the brand? Developers with myriad other priorities? Product owners who’ve spent months developing a new feature?
The reality is that experimentation programs often lose steam because they are operating within a silo.
Problems arise when people outside of the optimization team don’t understand the why behind experimentation. When test goals aren’t aligned with other teams’ KPIs. When experimentation successes aren’t celebrated within the organization-at-large.
Optimization champions can struggle to scale their experimentation programs from department initiatives to an organizational strategy. Because to scale, you need buy-in.
Most companies have a few people who are optimizers by nature, interest, or experience. Some may even have a “growth team.” But what really moves the dial is when everyone in the company is on board and thinks this way reflexively, with full support from C-level leaders.– Krista Seiden, Global Analytics Education Lead, Google
But getting buy-in for any initiative – especially one that challenges norms, like experimentation – is no easy task. Particularly if your organization suffers from silo mentality.
In this post, we outline a 5-step process for blasting through the silo-mentality blocks in your organization to create a culture of experimentation.
Our 5-step process for destroying silos so you can scale your experimentation program:
First things first: What is silo mentality?
At WiderFunnel, we often hear questions like: How can I get other people on board with our experimentation program? How can I create an organizational culture of experimentation, when our team seems to be working in a bubble?
When a company operates in silos, people have fewer opportunities to understand the priorities of other departments. Teams can become more insular. They may place greater emphasis on their own KPIs, rather than working with the team-at-large towards the organization’s business goals.
But it’s not silos that are necessarily the problem, it’s silo mentality.
And when an experimentation mindset is only adopted by the optimization team, silo mentality can be a major barrier to scaling your program to create a culture of experimentation.
Silo mentality causes people to withhold information, ignore external priorities, and delay processes where other teams are involved. All in an effort to ensure their team’s success over another’s.
Within a silo, members can suffer from groupthink, emphasizing conformity over confrontation and allowing weak ideas or processes to go unchallenged. They rely on intuition to guide their practices, and resist change because it’s new, uncomfortable, and different.
At its worst, silo mentality can point to adversarial dynamics between teams and their leads. It points to internal conflict, either between management as they fight over limited resources or compete to rise to the upper echelons of your organization.
Teams follow the leader
If you don’t have the right support at the very top, you’ll never be able to take your program beyond a certain point. Approaching your business in a different way requires a lot of risk and willingness to change, and changing process is hard. In the past, I ran into the roadblock of not having the right people evangelizing and supporting testing. Whereas at MailChimp, I’ve had that support. And I think that will define the success of experimentation at any organization.– Lauren Schuman, Director of Growth at MailChimp
Silo mentality often comes down to leaders, who are creating the goals and priorities for their teams. If team leads experience conflict, this us-against-them mentality can trickle down to their reports.
Managers, particularly under stress, may feel threatened by another manager’s initiatives. This is because silos often form in organizations where leaders are competing. Or, they appear in organizations where there is a clear divide between upper management and employees.
Unfortunately, silo mentality is a pain point for many optimization champions. But every department is a stakeholder in your organization’s growth. And to enable a strong organizational culture of experimentation, every department needs to understand the value of testing—the why.
So, let’s dive in and explore our 5-step process for breaking down silo mentality. At the heart of this process is creating an understanding of what experimentation can do for the individual, the department, and the organization-at-large.
Step 1: Align your stakeholders on a vision for experimentation.
You may be thinking: What does a “culture of experimentation” even look like?
That’s a great question.
A culture of experimentation is about humility and fearlessness. It means that your organization will use testing as a way to let your customers guide your decision making.
But you should clarify the vision for experimentation at your organization. Your vision should be high-level and aspirational. You’ll want to inspire others with the possibilities offered by experimentation.
Ask yourself these questions to create a vision for your experimentation program:
- What business problem(s) do we hope to address with experimentation?
- Why is experimentation the best solution to our business problem?
- What would test-and-learn values add to our individual team members, departments, and organization?
- What do we envision for our organization’s “culture of experimentation”?
In traditional business settings, leadership often takes a top-down approach to communication. But experimentation flips this dynamic on its head. Instead of the HiPPO (highest paid person’s opinion) calling all the shots, all ideas must be tested before initiatives can be implemented.
To me, a culture of experimentation is simply measured by the number of people in an organization who are able to admit, ‘I don’t know the answer to that question, but I know how to get it’.
If people within your organization are telling you ‘This is what our customers want’ (without the data to back it up) then you have a problem. Organizations that excel at experimenting aren’t better at knowing what customers want, they are just better at knowing how to find out.–Mike St, Laurent, WiderFunnel Senior Optimization Strategist
The most effective way to persuade others to adopt an experimentation mindset is to subscribe to your vision. You need to demonstrate the test-and-learn values of an Optimization Champion. Values like:
We listen to our gut, then test what it says.
We gather market research, then test it.
We create best practices, then test them.
We listen to our opinions, then test them.
We hear the advice of others, then test it.
We hear the advice of experts, then test it.
We believe in art and science, creativity and discipline, intuition and evidence, and continuous improvement.
We aim for marketing insights.
We aim to improve business results.
We test because it works.
Scientific testing is our crucible for decision-making.
– Chris Goward in “The Optimization Champion’s Handbook”
Once you have clarified your vision, write it down in a declarative statement. And be ready to communicate your vision over. And over. And over.
Step 2: Get buy-in from stakeholders and leaders.
You can’t achieve a culture of experimentation by yourself. You need testing allies.
Other department leads can help create momentum, acting as internal influencers, inspiring others to adopt an experimentation mindset in their workflow. They can help spread the gospel of a test-and-learn culture.
When executives embrace failure as an inherent and integral part of the learning process, there is a trickle-down effect on the entire enterprise. Increasingly, more employees from more departments are eager to learn about the customer experience and contribute their ideas. With more individuals invested and involved, it’s easier for a company to gain a deeper understanding of its customer base and make informed decisions that drive business value.– Optimizely’s “Creating a Culture of Experimentation”
To do this, you need to understand what will motivate stakeholders to fully adopt an experimentation mindset; how to incentivize them to champion the cause of experimentation. And of course, not everyone will prescribe to your vision.
At least not right away. It may take some finesse. In her Growth & Conversion Virtual Summit presentation, Gretchen Gary, Product Manager at HP, outlined three different types of stakeholders that may have difficulty engaging in a testing culture.
- New employees. Recent hires may have yet to experience a test-and-learn culture in their work histories. When their ideas are then positioned as hypotheses, they may feel threatened by the idea of testing and results.
- Senior-level managers. They have been hired for their expertise and granted the authority to lead projects. They may feel challenged by the need to test their ideas, rather than using their executive judgment.
- Long-term employees. Sometimes the most difficult stakeholders to convince are the employees who have been with your organization for decades and have set workflows. To them, they’ve found a proven process that works and they’re (perhaps) uncomfortable with change.
The underlying emotions for all three types of stakeholders are the same:
- they may feel threatened by new ideas and concepts;
- they may feel challenged by their reports and/or leaders;
- or they may fear failure, being proved wrong.
Your job is to inspire them to overcome these emotions. You need to communicate the possibilities of experimentation to each department in a way that makes sense for them – particularly in terms of their own performance.
What’s in it for your stakeholders?
You, the Optimization Champion, will need to mitigate different perspectives, opinions, and knowledge levels of testing. You’ll want to:
- Address the questions, “Why should I care?” Or, “What’s in it for me?”
- Tailor your message to each individual leader’s concerns.
The best thing you can do is try to familiarize yourself with [other team’s] KPIs so you can speak their language and know what might drive them to be more involved with your program.– Gretchen Gary
Support your vision of experimentation by building a business case. Leverage existing case studies to demonstrate how similar organizations have achieved growth. And show, through real-world examples, how different internal teams — from product development to marketing, from branding to IT — have incorporated experimentation into their workflows.
Step 3: Clarify your experimentation protocol.
It’s important to create an experimentation protocol so that people across your organization understand how and when they can contribute to the experimentation program.
Remove bottlenecks and unify siloed and understaffed teams by formalizing an optimization methodology, empowering individuals and groups to take ownership to execute it.– Hudson Arnold, Senior Strategy Consultant at Optimizely
A standard process enables any employee to know when they can take ownership over a test and when they’ll need to collaborate with other stakeholders.
Building a test protocol is essential. If I’ve learned anything over the last six years, it is that you really have to have formal test protocol so everyone is aware of how the testing tool works, how a test is operated and performed, and then how you’re reading out your results. You will always get questions about the credibility of the result, so the more education you can do there, the better.– Gretchen Gary
First, evaluate how your experimentation program is currently structurally organized. And think about the ideal structure for your organization and business needs.
Experimentation programs often fall into one of the following organizational structures:
- All-owned testing: Optimizers exist in different teams across the organization. Each strategizes and executes experiments according to their own KPIs, as they are positioned to reach larger organizational goals. This model does not have the oversight of a central team.
- Center of excellence: Often in companies where individual teams have their own sites or domains. A central body that enables independent teams to run in parallel, encouraging the growth of skills and expanding experimentation.
- A hybrid version: When there is shared ownership of digital initiatives, a central body works with all relevant stakeholders to prioritize and plan for experimentation. But individual teams have the ownership to execute tests.
Regardless of how you structure your program, education is a major part of ensuring success when experimentation is a company-wide initiative. Anyone involved in testing should understand the ultimate goals, the experimentation methodology, and how to properly design tests to reveal growth and insights.
When clarifying your organization’s experimentation methodology, you should:
- Share your organization’s SMART goals for the experimentation program;
- Clearly communicate expectations for experimentation, including defined key deliverables;
- Outline the steps involved in an experiment, noting when approvals are needed and when team contributions are welcome;
- Educate your team(s) in experimentation, including the tools, frameworks, and technologies, and how to properly structure experiments;
- And, finally, ensure you have a proper feedback loop to determine how and why contributions are prioritized internally. When your team understands that you can’t test everything at once, they’ll understand that every idea is vetted through a prioritization framework.
“Every department should have complete access to and be encouraged to submit ideas for experimentation. But this should only be done when the company is also confident it can complete the feedback loop and provide explanation as to the acceptance or rejection of every single idea,” Mike St. Laurent explains.
“An incomplete feedback loop – where people’s ideas get lost in a black hole – is one of the most detrimental things that can happen to the testing culture. Until a feedback loop can be established, it is better for a more isolated testing team to prove the value of the program, without the stressors caused by other parts of the organization getting involved.”
Different departments in your organization offer unique insight, experience, and expertise that can lead to experiment hypotheses. Experimentation protocol should communicate why your organization is testing, and how and when people can contribute.
Step 4: Get strategic with cross-functional teams.
If silo mentality is limiting your experimentation program, cross-functional teams may be an ideal solution. On cross-functional teams, each member has a different area of expertise and can bring a unique perspective to testing.
Eliminate the territoriality of small teams,” advises Deborah Wahl, CMO of Cadillac and former CMO of McDonald’s. “[Leverage] small, cross-functional teams rather than teams at-large and really get people committed to working towards the same goal.
When you form cross-functional teams, everyone benefits by gaining a deeper understanding of what drives other teams, what KPIs measure their success, and how experimentation can help everyone solve real business problems. They can also generate a wealth of experiment ideas.
Hypothesis volume is (after all) one of the biggest roadblocks that organizations experience in their optimization programs.
Cross-functional teams can channel the conflict associated with silo mentality toward innovative solutions since they help to break down the silo characteristic of groupthink.
How to move from groupthink to think tank
Bruce Tuckman’s theory of group development provides a unique lens for the problem of collaboration within teams. He breaks down the four stages of group development:
In the first stage, forming, a team comes together to learn about the goals of other team members and they become acquainted with the goals of the group. In this case, the goal is growth through experimentation.
Everyone is more polite in this stage, but they are primarily still oriented toward their own desires for an outcome. They are invested in their own KPIs, rather than aligning on a common goal. And that’s fine, because they’re just getting to know each other.
In the second stage, storming, the group learns to trust each other. And conflict starts to rear its head in group discussions, either when members offer different perspectives or when different members make power plays based on title, role, or status within the organization.
But for the team to work, people need to work outside the norms of hierarchy and seniority in favor of collaboration.
In this stage, members feel the excitement of pursuing the goals of the team, but they also may feel suspicion or anxiety over other member’s contributions. You want to make sure this stage happens so that people feel comfortable raising unconventional or even controversial perspectives.
In the context of experimentation, one person’s opinion won’t win out over another person’s opinion. Rather both opinions can be put to the test.
“I find [experimentation] has been a great way to settle disputes over experience and priorities. Basically you just need to find out what people want to know, and offer answers via testing. And that in itself is gaining trust through collaboration. And to do so you need to deliver value to all KPIs, not just the KPIs that your program will be measured on. Aligning on common goals for design, support, operations, and others will really help to drive relevancy of your program,” explains Gretchen Gary.
It’s important to enable the right kind of conflict—the kind that can propel your experimentation program toward new ideas and solutions.
The third stage, norming, is when members of the group start to accept the challenge of meeting their shared goal. They understand the perspectives of others and become tolerant of different working or communication styles. They start to find momentum in the ideation process, and start working out solutions to the problems that arise.
And the last stage, performing, is when the team becomes self-sufficient. They are now committed to the goal and competent at decision-making. And conflict, when it arises, is effectively channeled through to workable solutions.
Teams may go through these stages again and again. And it’s necessary that they do so.
Because you want weak ideas to be challenged. And you want innovative ideas to be applied in your experimentation program.
Step 5: Let the internal communication flow.
Free-flowing internal communication is essential in maintaining and scaling experimentation at your organization.
You should be spreading experiment research, results, and learnings across departments. Those learnings can inform other team’s initiatives, and plant the seed for further marketing hypotheses.
Information has a shelf-life in this era of rapid change. So, the more fluid your internal communication, the more central and accessible your data, the more likely it will be put to use.
How are customer learnings and insights shared at your organization?
One method for sharing information is to create an “intelligence report.”
An intelligence report combines data generated from your organization and data derived from external sources. Paired with stories and insights about experimentation, an intelligence report can be a helpful tool for inciting creativity and generating experimentation ideas.
Another method is to provide regular company-wide results presentations. This creates an opportunity for team members and leaders to hear recent results and customer insights, and be inspired to adopt the test-and-learn mindset. It also provides a space for individuals to express their objections, which is essential in breaking down the silo mindset.
But sharing insights can also be more informal.
WiderFunnel Strategist Dennis Pavlina shares how one of his clients posts recent test results in the bathroom stalls of their office building to encourage engagement.
A new idea doesn’t get anywhere unless someone champions it, but it’s championship without ownership. Keep it fun and find a way to celebrate the failures. Every failure has a great nugget in it, so how do you pull those out and show people what they gain from it, because that’s what makes the next phase successful.– Deborah Wahl
Whatever tactic you find most effective for your organization, information dissemination is key. As is giving credit for experiment wins! At WiderFunnel, we credit every single contributor – ideator, strategist, designer, developer, project manager, and more – when we share our experiment results. Because it takes a team to make truly drive growth with testing.
Build trust in a culture of experimentation…
A lot of what we talked about in this post is about building trust.
People need to trust systems, procedures and methodologies for them to work. And every initiative in breaking down silos should be geared towards earning that trust.
Because trust is buy-in. It’s a commitment to the process.
…but creating a culture of experimentation is an iterative process.
Creating and maintaining a culture of experimentation doesn’t happen in a straightforward, sequential manner. It’s an iterative process. For example, you’ll want to:
- Revisit your vision after six months. Has it had wide-spread adoption? Is your organization aligned on the vision? Do your stakeholders thoroughly understand the benefits of a culture of experimentation?
- Evaluate your experimentation program for efficiency. Are team members communicating any constraints, any roadblocks, any redundancies in the process? Can you make it more agile, more flexible?
- Apply your learnings to accelerate the progress of your cross-functional teams. Are your teams forming, storming, norming, and most importantly, performing? What can be improved in terms of collaboration and knowledge-sharing?
Because a culture of experimentation is about continuous exploration and validation. And it’s about testing and optimizing what you’ve learned as an organization. Which means you’ll need to apply these concepts over and over.
Make the terms a part of your vocabulary. Make the steps a part of your routine. Day in and day out.
What does a culture of experimentation look like for your organization? Let us know in the comments!