This week, Duncan Ramsay joined over 70,000 attendees in Lisbon for Web Summit 2025, one of the world’s largest technology conferences. Unsurprisingly, the topic of the day was AI, with speakers covering quantum computing to the future of SaaS. One of the key takeaways was that across the board, it’s clear that those who are successfully becoming AI-first businesses are having to make big, strategic choices. It can’t just be small side projects or Lovable hackathons. Duncan looks at some of those big decisions leaders are having to consider:
1. Once AI is a colleague not a tool, what does your team look like?
The most striking theme from the conference was the shift from AI as a tool to AI as a co-worker. Businesses are increasingly deploying agents – autonomous AI systems that perform tasks traditionally done by humans – who are already transforming internal operations. For example, Vercel reduced its inbound SDR team from ten to one by codifying top-performer workflows into AI agents.
This isn’t just about efficiency. It’s about consistency and scale. AI can model top performers, reduce performance variability, and free up human time for higher-value work. At Vercel the remaining SDR now manages the agents, with the role becoming more analytical and strategic. That is a fundamental shift in thinking about the role of the team, types of hire, and how you want to structure your business going forward.
2. Building AI-first businesses
Shifting to an AI-first business requires cultural change. Des Traynor from Intercom shared how the company pivoted from a seat-based SaaS model to an AI-first business. Their agent, Fin, now resolves 65% of their customers’ queries and has reached a $50m run rate. The shift is significant and requires a willingness to make hard decisions and a deep understanding of the “jobs to be done” in each workflow.
The message was clear: AI transformation isn’t optional. It must be driven top-down, with a focus on measurable ROI. Companies that deploy AI in silos or treat it as a side project risk missing the real value.
3. Vertical SaaS: From software to service
Software is no longer just servicing users in end verticals; it is replicating the work those users were doing. A good example of this is Clio, which evolved from providing software to lawyers to now using AI to execute actual legal work. By automating tasks like onboarding and contract review, Clio is expanding access to legal services and unlocking new market opportunities. The same principles apply across SaaS businesses willing to take the leap, from healthcare to logistics. Can you move your business model from a tool that helps people get work done to a tool that does the work?
4. Trust, accuracy and the human in the loop
Companies are increasingly embedding trust and regulatory considerations into their AI strategies from the outset, recognising that ethical governance is essential to long-term success. Product decisions often include conducting ethical impact assessments, using frameworks like IBM’s Risk Atlas Nexus, and factoring in privacy, sovereignty, and data hosting concerns.
In regulated industries such as law and finance, firms like Luminance are mitigating risks like AI hallucinations by fine-tuning models with specialised data, using multi-model validation, and ensuring human oversight for high-stakes decisions.
5. What this means for the companies we back
For mid-market businesses, the opportunity is clear. AI can drive efficiency, improve customer experience, and unlock new revenue streams. But it requires thoughtful deployment, cultural readiness, and a clear understanding of where AI can add value.
At ECI, we’re already seeing portfolio companies get tangible value from agentic workflows (e.g. Avantia’s Holmes claims agent), AI-enhanced customer service (e.g. Moneypenny’s AI-enhanced PAs), and intelligent automation. Our role is to help them navigate this change – bringing the right talent, tools, and mindset to the table.
Web Summit 2025 reinforced that AI is a fundamental shift in how businesses operate for those willing to embrace it.
