Paul Storm

Paul StormSep 14, 2026

>> Most AI assistants still stop exactly where work begins. They answer. Then you do the job. That's why MCP matters more than another chatbot inside another SaaS product. With Omnisend connected to ChatGPT or Claude, the assistant can work with the account itself: pull reports, analyze performance, recommend segments, check deliverability and help build campaigns and automations. Now the conversation becomes operational. -"Which automations made the most revenue last month" is useful. -"Show me which message inside the top three flows is underperforming and what I should fix first" is much more useful. That distinction is where enterprise AI is heading. The value isn't generating more words. It's reducing the distance between a business question and a business action. For years, software companies competed on who could build the best dashboard. The next competition may be who can make the dashboard least necessary. The winners won't just add AI to the product. They'll let AI actually use the product. Thinking of switching? Omnisend can migrate your paid account for free in 5 days Check it out: https://lnkd.in/e3pWH_Rm Which SaaS tool in your stack would become dramatically more valuable if you could operate it entirely through conversation? #AI #MCP #SaaS #Automation #FutureOfWork

Paul StormSep 17, 2026

>> Your best engineers shouldn't be building PDF plumbing. That's not innovation. It's infrastructure. And AI makes that distinction brutal. One lesson from technology leadership: engineering time is one of the most expensive resources a company has. So if I were building an AI agent for onboarding, insurance claims or HR. I wouldn't spend that time recreating PDF filling, document generation and e-signature flows. I'd buy that layer and keep the team focused on what differentiates the product. That's why Anvil It's not another generic "AI for documents" tool. It's purpose built for developers building document flows, and its new MCP gives agents a direct path into that infrastructure. An agent can map a schema to document fields, fill or generate PDFs and move the workflow toward completion instead of handing the last step back to a human. Its API also received an A+ rating from Supergood. This is the build vs buy decision AI teams will face more often: own the intelligence that creates your advantage, buy the infrastructure that doesn't. Your moat should be the product, not the plumbing. What are your engineers still building today that should probably be infrastructure? #AnvilPartner #AIAgents #MCP #ProductManagement #Engineering

Paul StormSep 16, 2026

>> Microsoft has 30M paid Copilot seats. That should look like AI dominance. But the real story is employee choice. Microsoft has the distribution every AI company wants. Copilot is inside Microsoft 365. IT can approve it centrally. Procurement already knows the vendor. Security teams know the stack. And Microsoft says Copilot has passed 30 million paid users. Yet 66% of office professionals in a recent PagerDuty survey said they had used AI tools at work even when they believed company policy did not allow it. That tension is the story. A paid seat proves procurement. It does not prove preference. Employees will still open ChatGPT, Claude or whatever tool gets the job done fastest. Microsoft can bundle Copilot into the workflow, but it cannot bundle enthusiasm. This is why enterprise AI will not be won by distribution alone. The winning product will be the one people reach for before IT tells them to, then ask IT to approve afterward. Microsoft still has enormous AI scale. But the harder battle is becoming habit, trust and product pull. Which AI do your employees actually choose when they have a choice? --- ♻️ Repost to share with your network ➕ Follow me Paul Storm for cutting-edge AI insights Join 600,000+ professionals using AI to stay ahead: 🔗 https://lnkd.in/eFNvmcYa #AI #Microsoft #Copilot #ChatGPT #FutureOfWork