monday.com just rebuilt its entire product around AI agents on May 6, 2026, and it is the clearest signal yet that the per seat SaaS model is breaking. After designing 42 enterprise products across 8 plus years, I am watching this shift up close, and I think the next 18 months will redraw the lines of what enterprise software even looks like. Deloitte predicts up to 75 percent of organizations will pour the majority of their digital budgets into agentic AI in 2026. Roughly $285 billion in SaaS market cap has already evaporated. This article unpacks what monday.com actually shipped, why this matters for product designers, and how I would think about redesigning a SaaS product around agents instead of dashboards.
I read monday.com's announcement twice on Wednesday morning, then I closed my laptop and went for a walk. It is not a feature release. It is a repositioning of the entire company from a work management platform to an AI Work Platform. The press release calls it the biggest change in monday.com's history, and that is not marketing puff. They are betting the brand on the idea that humans and agents do work together inside the same canvas.
If you have been designing SaaS dashboards for the last decade, this should make your stomach drop a little. In a good way.
"monday.com agents can draft campaigns, qualify leads, close support tickets, onboard new hires, process purchase requests, and more, 24/7, all under human supervision."
monday.com Investor Relations, May 6, 2026
What monday.com Actually Shipped
Read past the press release and the architecture is interesting. monday.com built native agents that any team member can configure and direct without writing code. The agents pull from live data across departments and run inside the same permissions model the business already trusts. They added one click connectors to Anthropic Claude, Microsoft 365 Copilot, and OpenAI ChatGPT through something called the AI Platform Gateway. They also rebuilt the mobile app so Sidekick and the agents sit in one place.
Three things stand out to me as a designer. First, agents are not a separate feature tab, they are embedded inside the existing workflows. Second, the multi LLM gateway tells me they expect customers to mix and match models per task, which is exactly how I have been advising teams to architect AI features for the last year. Third, they redesigned the mobile experience around the agent, not the kanban board.
That last one is the tell. The center of gravity moved.
The SaaSpocalypse Is Not a Meme Anymore
While monday.com pivots, the broader market is repricing every per seat SaaS company on the planet. Wall Street analysts now openly call it the SaaSpocalypse, and the numbers are brutal. Roughly $285 billion has been wiped from software stock valuations in early 2026. The investor consensus is that hundreds of SaaS companies built on per seat pricing were structurally overvalued for the agent era.
The math is simple and painful. When one user with an agent can do the work of five, you do not need five seats. Gartner now predicts that by 2030, at least 40 percent of enterprise SaaS spend will move to usage, agent, or outcome based pricing. The Next Web reported that AI native enterprise spending surged 94 percent while traditional SaaS stagnated at 8 percent.
The pricing model is downstream of the product model, and the product model is downstream of the design model. If your software still treats a human as the primary executor, you are designing for a shrinking market.
Why Monday's Bet Reframes My Job
I have spent the last two years designing AI first interaction layers on top of legacy enterprise SaaS. The biggest mental shift, the one that takes teams six months to internalize, is this: the user is no longer the only doer. The user becomes a director. The agent becomes a doer. The product becomes the rehearsal stage where they coordinate.
That sounds abstract until you sit in a real workshop. I ran one last month for a logistics SaaS that processes 2.4 million shipments a year. The team had built a beautiful dashboard with twenty plus filters. We killed three quarters of it in two weeks. Why? Because once the agent could pre filter, summarize exceptions, and recommend actions, most of those filters were dead weight. The new screen had a chat surface, an action queue, and three confidence indicators. The customer support team's task completion time dropped by a measurable margin within the first sprint.
monday.com basically made the same call at platform scale. They are the canary that flew into the agent shaped mine first.
What Product Teams Should Steal From This Playbook
If you run product or design at a SaaS company right now, here are the moves I am pushing my own clients toward, in priority order:
- Audit your dashboard density. Every chart, filter, and table is a candidate for an agent that surfaces only the exceptions worth a human's attention. If a screen exists to help users hunt for problems, an agent can usually do that hunting first.
- Redesign for delegation, not configuration. The next generation interface lets a user say what outcome they want. The agent figures out the configuration. Settings pages start to feel like legacy UI.
- Build a permissions and audit trail layer first. monday.com explicitly mentioned governance. Agents inside enterprises need scoped permissions, audit logs, and review queues, otherwise legal and security will block adoption.
- Plan for multi model swap. The AI Platform Gateway pattern is going mainstream. Your architecture should let you route different tasks to different models without rewiring the UI.
- Pick a pricing experiment now. Even if you keep per seat pricing this year, run a parallel pilot with usage or outcome pricing on one product line. The companies that wait until 2027 will be playing catch up against agent native competitors.
The Designers Who Will Win This Cycle
I keep getting asked, mostly by junior designers in my Medium DMs, whether AI is going to take their jobs. Here is the honest answer based on what I see inside enterprise rollouts. AI is not replacing designers. It is replacing designers who only push pixels in Figma. The people who are getting promoted right now, and the people I am hiring on my own teams, can do three things that did not exist on a job description two years ago.
They can write a system prompt that produces the right tone for a customer facing assistant. They can sketch a confidence model: how the UI should change when the agent is 95 percent sure versus 40 percent sure. They can design what failure looks like, because agents fail differently than humans, and a good failure UI is the difference between a feature your customers love and a class action lawsuit.
If you are reading this and you have not yet shipped an AI feature, my advice is to find the smallest agent shaped problem in your product and ship it this quarter. The learning curve is steep and the only way down it is reps. I wrote about 14 specific mindset shifts teams need to make in my Medium piece last month, and I keep going back to one of them: your product is no longer a tool, it is a teammate.
What I Am Watching Next
monday.com is one move on a board that has many more pieces. ServiceNow and Accenture announced a forward deployed engineering program on May 6 to push agentic AI from pilot to production at scale. Oracle launched OCI Enterprise AI with prepackaged AI Accelerator Packs. Pit, a Stockholm startup, exited stealth on May 7 with 16 million dollars in Andreessen Horowitz led funding to replace the spreadsheets and rigid SaaS that run enterprise operations. Pit is already live with Voi, Tre, Stena Recycling, and Kry. Wyndham Hotels even shipped a native ChatGPT app for hotel discovery and booking.
The pattern is unmistakable. Every category is getting an agent native upstart, and every incumbent is either pivoting or shrinking. The companies that will still matter in 2028 are the ones doing the hard, unglamorous work right now of redesigning their product surface around delegation.
I will keep writing about what I see from inside these rollouts. If you are a founder, designer, or PM trying to make this shift, I would love to hear what is working and what is breaking for you.
Have you started redesigning your SaaS product around agents, or are you still in the planning phase? Drop a comment with what surprised you the most. I read every one and I usually reply.
Sources: monday.com Investor Relations (May 6, 2026), SiliconANGLE coverage of monday.com relaunch, Deloitte 2026 SaaS and AI Agents report, Gartner SaaS pricing predictions, The Next Web on AI native enterprise spending, Bloomberg Agentic AI 2026 Outlook, ServiceNow and Accenture press release (May 6, 2026), Oracle OCI Enterprise AI launch, EU Startups on Pit funding (May 7, 2026), PR Newswire on Wyndham ChatGPT app. Related reading on medium.com/@iahmadullahcs and reloadux.com/blog.