An AI-first company rethinks its strategy, ways of working, and people instead of adding isolated AI features.
2
Sprout Social is balancing today's customer roadmap with experiments aimed at experiences that AI makes possible.
3
AI transformation requires broad AI fluency, T-shaped talent, purposeful processes, and continued attention to reliability, trust, and human judgment.
Summary
Rossella Blatt Vital describes Sprout Social's AI transformation as unfinished and full of trade-offs. She defines AI-first work as a shift from adding intelligence to existing features toward imagining experiences that were previously impossible. That change affects strategy, planning, product design, engineering processes, hiring, and company culture. Sprout Social is moving away from fixed roadmaps because capabilities change quickly, while still shipping work customers need today. The company uses ritualized discovery, MVPs for learning, and processes judged by whether they improve decisions and unblock teams. Blatt Vital also argues for T-shaped talent and AI fluency across the whole organization, supported by newsletters, podcasts, show-and-tells, and a shared platform for prototyping. She is clear that AI does not remove the basics: products still need to solve customer problems, perform reliably, earn trust, and preserve human judgment and care.
AI-first means rethinking the company around what AI makes possible
Blatt Vital separates an AI-first strategy from an AI-enhanced one. An AI-enhanced product adds intelligence to an existing experience. An AI-first company asks what new experiences can exist because of AI, including problems that customers may not yet know how to describe. She says AI differs from earlier shifts such as cloud, mobile, and DevOps because it disrupts product, architecture, people, process, and ethics at the same time. The change is not a switch that a company flips. It is an evolution, and companies need to understand where they are and move with purpose.
Companies must run today's roadmap and tomorrow's experiments in parallel
The company still has to ship work that meets current customer needs while investing in a future that keeps changing. Overweighting the present can leave a company behind. Focusing only on the future can disappoint customers, slow revenue, and deprive innovation of resources. Blatt Vital describes this as the innovator's dilemma, made harder by AI. At Sprout Social, a three-month roadmap can feel stale within three weeks, so planning now allows learning and discovery to change the destination. The company is trying to combine enterprise discipline with startup curiosity and nimbleness.
AI-first product design unifies workflows instead of exposing team boundaries
Blatt Vital uses the fictional puppy health app Wufill to show why separate AI features are insufficient. Pineapples, the customer, tracks meals, activity, digestive insights, and supplement recommendations, with each area owned by a different team. If Pineapples says, "I'm feeling kind of blah," narrow feature responses might mention skipped breakfast or unusual stool. A unified response can combine the new kibble, declining activity, digestive changes, and a recent supplement into one suggestion. The customer experiences one problem, not the internal feature ownership. AI-first design therefore requires teams to build broader value across workflows over time.
Sprout Social is moving away from ad hoc innovation, where someone proposes an idea, tests it in isolation, and then moves on. Blatt Vital calls for ritualized discovery, which means reserving time in planning cycles for experiments, hackathons, and learning forums that are visible and actionable across the company. The team also treats MVPs as tools for learning and direction, rather than only as ways to launch quickly. Some bets will fail or behave differently than expected. In her framing, that failure produces clarity when the path is still ambiguous.
Processes should be judged by the outcomes they create
Older processes were built for predictable roadmaps and risks that could be identified months ahead. They can become a poor fit when AI capabilities and timelines change monthly. Adding many new processes can create overhead and slow teams down, so Sprout Social has started treating process as a product. The questions are practical: does the process create clarity, unblock teams, or help people make better decisions faster? If it does not, the process should be changed or removed. Blatt Vital does not reject process. She wants process that gives teams clarity and flow.
Fast execution alone can produce chaos when teams do not know what they are trying to achieve. Direction without momentum can lead to stagnation. Blatt Vital says Sprout Social is moving from speed to "smart velocity," which means moving quickly with purpose. Prototypes, MVPs, and rapid delivery are useful when they support clarity, momentum, and adaptability. The point is not to ship indiscriminately. Teams need enough direction to learn from what they build while retaining the ability to change course as the technology and customer expectations develop.
AI-first companies need T-shaped builders and company-wide fluency
Blatt Vital says deep AI specialization remains important, but it now needs to be paired with broad building skills. Sprout Social is investing in T-shaped talent that can prototype, work across silos, and bring systems from idea to production. The shift extends beyond people who train models or build agents. Marketers, designers, product managers, and care agents should all be able to understand and use AI in their work. Sprout Social supports this with AI newsletters, podcasts, cross-functional show-and-tells, and a platform that lets product and engineering teams prototype without depending deeply on the AI team.
AI does not remove the fundamentals of good software
The company's strongest AI features still begin with customer problems rather than the novelty of AI. Blatt Vital says user experience, performance, reliability, and trust remain non-negotiable. Human creativity, judgment, and care also remain part of leadership, decision-making, and customer work. Her advice to transformation leaders is to be honest about trade-offs, invest in people as well as models and agents, change the roadmap when necessary, learn openly, and make room for unusual ideas. She does not present a finished playbook. She says companies can begin by asking better questions and accepting that the path will change.
"You basically need the discipline of an enterprise and the curiosity and nimleness of a startup, but you need both to be running in parallel."05:47
Who should watch
You are responsible for an SaaS roadmap and need to keep current customer commitments while making room for AI-driven product changes.
Your AI work is concentrated in a specialist team, and you need practical ways to build fluency across product, engineering, design, marketing, or customer care.
Your organization is adding experiments and processes faster than it can learn from them, and you want a clearer test for what should stay.
