Ryan Serhant uses AI agents to lift the 80% of admin load that salespersons are burdened with, so his people can focus on relationship-building and the human elements that matter the most and connect directly to revenue generationPhoto courtesy of SERHANTI see you asking ChatGPT how to make money with AI. And you’re likely seeing the same recycled answers:Create a course, build a digital product, use AI for dropshipping, write an e-book, vibe code an AI tool, etc. But you’re overlooking the biggest game-changer as far as leveraging your expertise to make money with AI is concerned. This year at SXSW London, I sat down with Ryan Serhant, star of Owning Manhattan and Million-Dollar Listing. As CEO of SERHANT., a multi-billion dollar real estate firm and internationally recognized real estate mogul, I wanted to understand how he approaches making money with AI as an entrepreneur.When I interviewed Serhant post-event, his explanation of how it works in practice provided a much more useful answer to the making-money-with-AI question.How Ryan Serhant Uses AI To Make MoneyIn our chat, Serhant told me about his proprietary AI tool that he calls S.MPLE. It’s essentially a collection of AI agents that are assigned roles, like employees or teammates, which cut out the excess administrative workload that real estate agents are often weighed down with.“Salespeople lose time to data entry, tracking, marketing tasks, and operational drag,” Serhant, who leads a team of about 1,000, explained (edited for clarity).“They could sell more if they had more time. They could increase their price point if they had more time. And when we looked at it,” he continued, “about 80% of a salesperson’s day is doing administrative tasks in one way, shape or form, while 20% of their day is doing what only a human can do--meeting with clients, pitching, building trust, networking, all that stuff,” he said.This gave Serhant the idea: distinguish between work that supports revenue, and work that directly produces revenue. Unfortunately, he admitted, entrepreneurs and even the highest-performing salespersons spend so much time on the first that they hardly have mental bandwidth or the time capacity for the latter.This problem isn’t restricted to real estate or the sales industry alone.As a matter of fact, this is something I’ve seen frequently appear as an issue with leaders and managers, especially middle managers. Managers spend 40% of their time on busy work, and only 15% and 13% respectively are doing what they were hired to do in the first place: strategy and team leadership and people development, per Deloitte.This means that the value of managers is being sucked away into endless time-consuming admin that takes away from real people development, coaching, and enabling growth and scalability. With agentic AI, however, professionals whose roles, whose success in their roles hinge on their relationship building with people, such as sales and management, are able to delegate this time-consuming work and focus on what really matters.As an example, Serhant divides his agentic AI capabilities into teammates that serve different core functions within a business, from operations to sales support to finance to marketing. Whether you’re a freelancer and don’t have the time or the financial resources to outsource to an agency or expensive service, or you’re managing a workflow and project team in your organization, here’s how you can use AI to drive growth, revenue, and productivity like Serhant:Treat AI As A Teammate Rather Than An Individual ChatbotIdentify the repeated scenarios in your workflow or that of your team, that take away more time from the day than it should.Assign business functions for each scenario.Define what good looks like, what an unacceptable or “meh” output would like look like, and what steps your AI agent would need to take, as well as what plugins, APIs, or connected resources and apps it needs access to so it can complete the task effectively.Then build an AI agent for each part of your workflow, to amplify the human work you and your team does.Ensure there are assigned checkpoints that keep humans in the loop for quality control and reduction of errors.An AI teammate can be created through a Claude or ChatGPT skill, using an Asana teammate, or even vibe-coding it yourselfgettyExamples of AI TeammatesSerhant’s AI agents are built specifically for the real estate industry, which makes them more effective than a generic Gen AI solution that’s taken “off-the-shelf”.Therefore the specific constraints, requirements, training data, and use cases that you’ll use in your industry and role will vary. Nonetheless, here are some suggestions based on Serhant’s S.MPLE agentic AI approach:Operations AI agents: for SOP creation and documentation, meeting actions and follow-ups, customer engagement and data analysis, etc.Finance AI agents: for invoice tracking, financial modeling for pricing and project cost scenarios, cash flow and expense reporting, etc.Sales AI agents: for proposal creation and defining requirements, researching and identifying leads, customizing cold and warm outreach, identifying business partnership opportunities, etc.Creating and deploying these agents, personalized to your workflow, allows you to protect your headspace and your ability to do what you do best: being human.It gives you greater capacity to spend more time thinking creatively and strategically, engaging with leads, providing original analysis and advice to clients, and providing added value.Here are some ways that you can create these agents without touching a single line of code:Develop agentic workflows in Claude or ChatGPT using Claude Skills or ChatGPT Skills.Use tools like Perplexity Computer or ChatGPT Work’s capabilities to connect with plugins and APIs (pre-agreed and approved by your IT department of course).Vibe-code a small application that solves a specific problem you’re facing at work, that you can’t find a reasonable solution to anywhere else.These are the indirect, longer-term benefits of using AI at work. While they don’t promise overnight wealth, they afford greater opportunities for leverage so you can make more money and customize services to your clients.And that’s where the real income potential lies.Think about it:Why do more low-value work faster, when you can use AI to create real leverage from your expertise that allows you to deliver faster, close deals smarter, raise the bar and improve service quality, and create capacity without immediately needing to hire more people?So instead of asking Google or ChatGPT, “How can I make money from AI?” and hoping it invents a side hustle or dynamite business idea for you, reframe your question to:“If AI took everything boring and administrative off my plate, what revenue-producing and relationship-building work could I do more of?”