Title: James Goody on What AI Actually Does Inside a Real Estate Business "Everybody underestimates how much work it takes to use AI, and then overestimates how much AI will do for you." That's James Goody, early in the first episode of Net Effective. James is a CPA and the Managing Partner of Oak 21 Real Estate, an advisory and execution partner (not a broker) working with family offices and operators across multifamily, single family and commercial. He is also CFO and COO of Coastal UP Partners, which runs two workforce housing funds. Before that he was CFO of two multibillion dollar real estate platforms, one of them backed by Goldman Sachs and Carlyle. He has been involved in the ownership and operation of more than $6 billion of real estate, including more than 40,000 apartment units. He has run IT at several companies and describes himself as a prop tech nerd. He is now using AI across accounting, utilities, renovation tracking and research. In this episode From doing to reviewing. The clearest change AI has made to James's work. He compares it to standing up an outsourced team, where your own people move from producing the work to reviewing it. Pulling a rent roll, checking renovation status unit by unit and tallying spend used to be manual. That work is automated now, and what comes back is the outlier list showing which units are lagging. Connecting the systems. James is wiring QuickBooks Online, AppFolio and monday.com together to pull property insights he could not get before. He is candid that the break points between systems are where the trouble lives, and that connecting them is sometimes simple and sometimes anything but. Utilities and AP. James sits on the advisory board of LDGR Systems, which automates utility payments and accounts payable. A property manager he works with moved an accounting person off manual reconciliation entirely. That team now spends its time asking why utility costs moved in a given month and how that tracks against occupancy. The accuracy problem. AI returns an answer in seconds. James finds it pulling the wrong column, misreading what he asked for, or producing something that does not make sense. He and Jonas land on the same conclusion: AI output demands a different and often more demanding review than human work, because the mistakes are stranger. Claude, ChatGPT and the cost of switching. James started on ChatGPT and moved to Claude. He makes the case for using it as a deep research tool, explains why the context you build up inside one model makes switching genuinely hard, and gives his read on how Anthropic and OpenAI are each perceived right now. Whether AI replaces offshore teams. Jonas asks directly. James expects a combination of replacement and augmentation, and describes an AI company where outsourced workers wore recording headsets while they worked. Why service still has to be human. James tried to get help with a software plan and reached only AI agents, with no route to a person. Escalating produced the response "you seem upset." Both of them connect this to property management, where the product is somebody's home and bad service is not something a resident shrugs off. James makes the point that tenant retention is the name of the game right now, with rents flat and above market leases worth keeping in place. Privacy, security and who controls the data. As a former bank officer and national director of real estate at First Republic, James is cautious here. He gets into what happens to private company financials once they are uploaded, why he will not connect his personal Gmail to anything, and the current split between RealPage keeping data closed and Yardi going open. He frames that as Betamax versus VHS, with customers waiting to learn which format wins. AI agents, and the gap between easy and useful. Agents are what James most wants to learn next. Starting one takes seconds. Trusting the output, and deciding how much autonomy to hand over, is the hard part. AI in his personal life. He uses it about as much personally as professionally, including planning a trip to Healdsburg for a wedding. Neither he nor Jonas will let it book anything yet. The callout: Nick Virovec & Fischer Asset Management. Every episode of Net Effective ends with a guest naming an operator who deserves more credit than they get. James picked Nick Virovec, President of Fischer Asset Management in Cleveland. Nick is tech forward, piloted Ledger Systems, and used AI to reconcile a prior period adjustment across general ledgers, work that used to mean hours of Excel formulas. Mentioned in this episode James Goody, Oak 21 Real Estate · Coastal UP Partners · Nick Virovec, Fischer Asset Management · LDGR Systems · AppFolio · QuickBooks Online · monday.com · Claude · ChatGPT · RealPage · Yardi · First Republic Bank Net Effective is a conversation with the people running residential rentals about how they are actually using AI. New episodes weekly, about 30 minutes. Subscribe at neteffective.show.
