Enterprise AI adoption isn't about picking the right softwareâit's about building the right culture first. When automation platform Make set out to transform how its workforce uses artificial intelligence, leadership rejected the standard playbook: approve a tool, track logins, declare victory. Instead, they discovered that 96% of employees building AI automations came from something far more human: intentional discovery, embedded coaching, and an unwavering focus on measurable business outcomes. How Did Make Achieve 96% AI Adoption Without a Big Tool Rollout? When Sara Maldon joined Make two years ago as the company's AI adoption lead, there was no approved AI tool at all. Nobody could use ChatGPT or other models for work. But Make already had something more valuable: a culture built around automation. "Do it once, do it twice, automate it, never do it again" was already how teams operated. Rather than launching a company-wide mandate, Maldon started with discovery. In her first week, she organized a voluntary four-week bootcampâand a third of the company signed up without being asked. People wanted to understand AI but hadn't been given a pathway in. Before formally launching the "Miriwa" program (a Korean expression meaning "going into the future"), she conducted one-on-one conversations with over 160 team membersânearly half the company. She gathered data on individual knowledge levels, AI readiness, and where each department saw opportunities. This early engagement work meant the program felt relevant to the people it was designed for, not imposed from above. The result: by the time Miriwa launched in July 2025, organic adoption had already taken hold. By January 31st, 96% of employees had built AI automationsâroughly half agents, half AI workflows. The "Samurai" Model: Embedding Coaches in Every Department Make's breakthrough wasn't a toolâit was people. Maldon built a dedicated team she calls "samurai": full-time people embedded in every department dedicated to AI adoption. One in human resources, one in engineering, one in product, one in marketing, and so on. Each samurai operates as an embedded product manager who understands the department's actual problems, builds solutions alongside the teams, and coaches people through using them. The role is deliberately challenging. As one samurai told Maldon: "I'm obsessed with impact, but it doesn't depend 100% on me. I need to convince people and get them excited, and that doesn't always directly show." Maldon hires for resilience and curiosity partly for that reason. The samurai don't report to a central AI team aloneâthey report into both Maldon's team and the department VP, creating accountability to both the transformation and the business. Why Measuring Business Impact, Not Time Saved, Changed Everything Make deliberately rejected time savings as a success metric. If a project saves time but doesn't shift a core business number, it doesn't count. Maldon calls her targets "big fat hairy goals"âambitious enough to stretch but realistic enough that teams don't dismiss them. Against that bar, early pilots had a high failure rate. Maldon expected that, because the failed projects often produced more value than the ones that shipped. When a samurai overreached with an ambitious solution when the team actually just needed help with basics, that failure revealed where the real friction was. A department might have no process to act on what a tool produced, or data wasn't flowing where it needed to. Maldon used those discoveries to fix fundamentals rather than let her team keep pushing against resistance and risk losing buy-in. To balance ambition with momentum, each team now runs a tiered portfolio: - Strategic Projects: One or two major initiatives that could change how the department works - Procedural Improvements: Three to five smaller changes that support how teams operate day-to-day - Quick Wins: A handful of fast implementations to build confidence and momentum When Maldon announced Miriwa in July 2025, she targeted 90% AI and automation literacy across employees and 75% of projects moving core departmental metrics. The team cleared literacy at 96% of eligible employees (91% overall) and landed at 73% on business impact. The bar goes up again this year. Real Business Results: What the Numbers Actually Show Make's AI adoption strategy produced measurable outcomes across the organization. Annual recurring revenue (ARR) per full-time employee and average contract value both improved, and time to fill in hiring came down. While Maldon won't claim direct cause and effect, the improvements tracked to the areas where Miriwa invested most. Specific department wins included: - Product Team: Built an Insights Hub that pulls themes from thousands of user research data points into summaries via Slack - Marketing: Uses Make to update outdated articles, transcribe video, generate SEO fields, and monitor monthly performance - Engineering: Automated SOC compliance and change management processes - People and Culture: Launched a fully automated merch store where employees earn points for learning and redeem them for rewards All samurai projects must already move core business metrics. Employee builds don't have to show that yet, but the program is heading in that direction. How to Build Enterprise AI Adoption Without Top-Down Mandates - Start with Culture, Not Tools: Before approving software, understand your organization's existing relationship with automation and change. Make's success built on a foundation of automation-first thinking that already existed. - Conduct Discovery Before Launch: Talk to nearly half your organization one-on-one. Gather data on knowledge levels, readiness, and departmental opportunities. This makes the program feel relevant, not imposed. - Embed Coaches in Every Department: Hire full-time people who understand each department's actual problems and can build solutions alongside teams. They report to both the transformation lead and the department VP. - Measure Business Impact, Not Usage: Reject time savings as a metric. Focus on whether projects move core business numbers. This forces teams to solve real problems, not just adopt tools. - Embrace Strategic Failure: Early pilots should have a high failure rate. Failed projects often reveal more about organizational friction than successful ones. Use those insights to fix fundamentals. - Build a Tiered Portfolio: Balance one or two ambitious strategic projects with three to five procedural improvements and a handful of quick wins. This maintains both momentum and ambition. Maldon is blunt with leaders who hedge. When people come to her wanting to test whether an idea is viable, she pushes back: commit to it properly or don't bother. That doesn't mean betting everything on day oneâthe samurai model came after the business case, not before. But she has no patience for perfectionism slowing things down. She told one new samurai after seven weeks without a failed project: "I want to see a failure next week. Otherwise I will throw you into one". The organizational alignment matters too. Information technology needs to be ready because nothing starts without an approved tool. And the cultural side is harder. "If you want to do a culture transformation you need to have your HR officer on board being excited and potentially being the first person to build," Maldon explains. The Caveat: Context Matters for Your Organization Make sells automation. Their workforce is technically inclined and culturally predisposed to experimentation. A traditional mid-market firm with lower digital fluency faces a harder version of this challenge. The principles transfer, but the timeline and starting point will differ. Organizations without Make's automation-first culture may need to invest more heavily in foundational training before embedding samurai coaches. The lesson isn't that every company should copy Make's exact playbook. It's that enterprise AI adoption succeeds when organizations prioritize culture and coaching over tool selection and login tracking. Make achieved 96% adoption not because they picked the best software, but because they built the conditions where people wanted to learn, had support to experiment, and saw their work connected to business outcomes that mattered.
