Pricing AI Employees: How Homeward Abandoned Per-Seat SaaS Models

Pricing AI Employees: How Homeward Abandoned Per-Seat SaaS Models

An AI agent doesn’t clock out at 5pm. It doesn’t take lunch breaks. It can handle a thousand tasks simultaneously while your human employees sleep. So why would you price it like software that charges per login?

In a recent episode of Category Visionaries, Amar Kendale, President and Co-Founder of Homeward, explained why his company abandoned the sacred cow of SaaS pricing—the per-seat model—and built something entirely new. The decision reveals a deeper truth about AI products: they don’t fit traditional software economics, and forcing them into conventional pricing models creates the wrong incentives for both vendor and customer.

Why Per-Seat Pricing Breaks for AI

The logic of per-seat pricing made perfect sense in the software era. Each user gets a login. Each login represents a person using the tool. More people using it means more value delivered, so charge per person. Simple, predictable, fair.

AI employees break this model immediately. “We actually don’t like to charge by the seat because Homeward is an AI employee. The way to think about Homeward is you’re essentially hiring a bunch of employees,” Amar explains.

Consider what happens when an AI agent handles customer support tickets. It doesn’t need a seat—it needs access to the ticketing system, knowledge base, CRM, and communication tools. It’s not logging in and out like a human. It’s continuously monitoring for new tickets, processing them in parallel, and resolving issues without any concept of concurrency limits.

If you charged per seat, what would that even mean? Is each AI agent a seat? If so, customers would be incentivized to minimize agent deployment even when more agents would deliver more value. Is the person who configured the agent a seat? That creates an arbitrary constraint where the value delivered has nothing to do with the price paid.

The deeper problem is that per-seat pricing was designed for tools that augment human productivity. AI employees don’t augment productivity—they replace entire workflows. The economic value isn’t “this person is 20% more efficient.” It’s “this work no longer requires human attention at all.”

Aligning Price With Value Delivered

Homeward’s pricing model starts with a different question: what actually determines the value an AI employee delivers? The answer isn’t how many people have access to it—it’s what work gets done.

The model considers factors like workflow complexity, task volume, and integration depth. A simple AI agent that scores leads once per day delivers different value than one that handles hundreds of customer support tickets continuously. An agent that integrates with three systems creates different value than one that orchestrates across a dozen.

This approach aligns pricing with actual economic impact. When an AI agent automates a workflow that previously required two full-time employees, the pricing reflects that labor replacement value. When it handles peak load that would otherwise require temporary staff, the pricing scales with that capacity value.

The shift also changes customer behavior in productive ways. With per-seat pricing, customers worry about adding users and often restrict access to control costs. With value-based pricing, customers are incentivized to deploy AI wherever it creates value. There’s no artificial constraint that prevents a customer from automating a valuable workflow because they’ve hit their seat limit.

The Deployment Economics

One of the most compelling aspects of Homeward’s pricing approach is how it enables rapid deployment. “We probably on average can deploy an end to end workflow within a week,” Amar notes. This speed is possible partly because customers aren’t worrying about seat economics during deployment.

Traditional enterprise software deployment often involves lengthy negotiations about user counts, department allocations, and seat assignments. These conversations slow everything down and create organizational friction. Finance wants to minimize seats. Business units want unlimited access. IT needs to manage licenses. The software’s value gets lost in procurement logistics.

Homeward eliminates this friction by divorcing price from user count. Deploying an AI agent doesn’t require counting how many people might interact with it or predicting future user growth. The conversation focuses on what work needs to be done and what value that work represents—a much more productive discussion.

This also enables the product-led expansion that drives Homeward’s growth. “A lot of times customers will start with one workflow, but then they will expand,” Amar shares. That expansion happens organically because there’s no seat-based barrier. A successful deployment in customer support naturally leads to questions about sales operations, data analysis, and compliance workflows. Without per-seat constraints, the expansion conversation focuses purely on value.

The Free Tier Strategy

Homeward’s pricing philosophy extends to their free tier approach, which defies conventional SaaS wisdom. Most freemium products artificially limit functionality or capacity to create upgrade pressure. Homeward does the opposite.

“You can go to our website, sign up, use the product for free. There’s actually a generous free tier. You can connect your entire data, you can ask questions, you can deploy workflows,” Amar explains. This generosity only makes economic sense if you’re not thinking in per-seat terms.

With traditional pricing, a generous free tier means giving away seats you could charge for. With Homeward’s model, the free tier costs are directly tied to actual usage and value delivered. Light users cost little to serve and create expansion opportunities. Heavy users quickly exceed free tier limits because they’re getting substantial value—value that justifies paid plans.

The free tier also serves as natural qualification. Customers who can’t find value in the generous free tier probably aren’t good fits anyway. Those who quickly hit free tier limits are self-identifying as high-value customers who should convert to paid plans. The pricing model itself handles qualification and segmentation.

Pricing as Product Strategy

The pricing model isn’t just about revenue—it’s a core part of Homeward’s product strategy. By pricing around work completed rather than access granted, Homeward reinforces their positioning as AI employees rather than AI assistants.

“We actually think of Homeward as an AI employee. It’s not a chatbot that you’re asking questions to, but it’s actually an employee that’s getting work done,” Amar notes. The pricing model makes this positioning concrete. When customers pay for workflows automated and tasks completed, they naturally think about ROI in employment terms: how much would this work cost with human employees versus AI employees?

This framing also changes competitive dynamics. Competitors using per-seat pricing are competing on software metrics: features per dollar, seats per price tier, discount structures. Homeward competes on employment economics: cost per workflow automated, capacity per dollar spent, ROI per AI employee hired. These are fundamentally different conversations with different decision-makers and evaluation criteria.

The Scale Economics

Perhaps most importantly, Homeward’s pricing model scales with their vision. “We have about 400 AI agents deployed right now across all of our customers,” Amar reveals. This number will grow to thousands as customers discover new workflows to automate and deploy specialized agents for narrow tasks.

Per-seat pricing would completely break at that scale. Imagine a customer with 500 AI agents. What would seat-based pricing even look like? 500 seats for 500 agents? That’s absurd economics—the customer would be paying for “seats” that represent pure automation value. Price per employee who manages the agents? That creates the wrong incentive structure where customers are penalized for having lean operations teams.

Homeward’s model scales naturally because it’s based on value delivered. 500 agents handling 500 different workflows represents genuine business value—value that can be quantified in terms of labor replaced, capacity added, or efficiency gained. The pricing can scale linearly with that value without hitting arbitrary seat-based constraints.

The Market Education Challenge

Abandoning per-seat pricing isn’t without challenges. The SaaS market is deeply conditioned to think in seat-based terms. Sales cycles include standard questions: “How many seats do we need? What’s the price per seat? Can we get volume discounts on additional seats?”

Homeward’s approach requires educating customers on a different mental model. Instead of “how many seats?” the questions become “what workflows are we automating? what volume of tasks will the agents handle? what’s the economic value of this automation?”

This education happens through the product experience itself. When customers deploy their first workflow and see an AI agent autonomously handling work, the employment framing clicks. When they deploy additional workflows without worrying about seat counts, they experience the freedom of value-based pricing. The product teaches the pricing model through actual usage.

The Broader Lesson

Homeward’s pricing strategy reveals a broader principle for AI-native companies: don’t force AI products into software pricing models. AI economics are fundamentally different from software economics. Software scales through replication—the same features serve every customer. AI scales through automation—unique value for each workflow automated.

The companies that recognize this distinction and build pricing models around AI value rather than software metrics won’t just have better unit economics. They’ll enable customer behaviors that traditional pricing prevents: rapid deployment, aggressive automation, continuous expansion. They’ll compete on value delivered rather than features per dollar. And they’ll scale in ways that per-seat pricing makes impossible.

When you’re selling digital employees that work continuously, handle unlimited concurrent tasks, and automate complete workflows, charging per seat isn’t just impractical—it’s leaving money on the table. The question isn’t how many people log in. It’s how much work gets done.