Salesforce crowdsources AI roadmap with customers, signaling faster software cycles

Salesforce is meeting some customers as often as once a week to shape its AI roadmap, a sign that enterprise software is moving from slow planning cycles to a far tighter feedback loop. The company says it is treating its 18,000 customers as an ongoing source of product direction, not just a pool to survey at quarterly checkpoints. Jayesh Govindarajan told TechCrunch that those customers are a wellspring of information needed to build toward customer success, and that improving models will let agent systems take on more autonomous behavior over time.

For monday.com, that is more than a Salesforce story. It is a live signal about how work-management vendors are being judged by customers who want evidence that product teams are listening fast enough to keep up with AI. monday.com has already framed its own AI strategy around three pillars, AI Blocks, Product Power-ups and the Digital Workforce, in a July 10, 2025 rollout the company said was built in response to real customer needs. By the end of the first quarter of 2025, monday.com said customers had completed more than 26 million AI-driven actions across the platform.

The scale behind that push is now large enough to make iteration speed matter inside every function. monday.com says it has more than 250,000 customers worldwide, and in its fiscal 2025 results it said customers with more than $50,000 in annual recurring revenue represented 41% of total ARR. That mix puts more weight on enterprise expectations, where product, design and sales teams have to prove that AI features are not just shipping, but solving specific workflow problems.

monday.com has also been moving from AI messaging to product execution. The company said monday vibe became the fastest product in its history to surpass $1 million in ARR, and that more than 60,000 apps were built on monday vibe in about three months. On February 18, 2026, monday.com said it was welcoming AI agents to its platform with dedicated onboarding and infrastructure, and on March 23, 2026, it launched Agentalent.ai, a hiring platform for enterprise AI agents.

That sequence puts monday.com in the same competitive lane Salesforce is now signaling: the winner may not be the vendor with the loudest AI pitch, but the one that can turn customer input into shipping decisions quickly without breaking reliability. For engineers in New York and Tel Aviv, that means building an architecture that can absorb faster change. For product managers and sales teams, it means treating AI less like a finished roadmap and more like a co-developed system that has to earn trust release by release.