May 27, 2026
Every predevelopment team has felt it. You're sizing up a site, the clock is ticking on a purchase agreement, and somebody on your team is three tabs deep in a municipal code that was last reformatted in 2008. Across the table, an architect is waiting on setbacks. The lender wants a yes or no on feasibility by Friday. And the answer is buried in an overlay district referenced on page 412 of a PDF.
Manual zoning research is one of the slowest, most error-prone steps in the early development pipeline. It also happens to be one of the most important. Get the use, density, height, and overlay analysis wrong, and the rest of the underwriting falls apart.
The cost of that complexity is measurable. Regulation imposed by all levels of government accounts for an average of 40.6 percent of multifamily development costs (NAHB and NMHC, 2022). The same research found that 74.5 percent of multifamily developers encountered "Not In My Backyard" opposition on a project, which added an average of 5.6 percent to development costs and delayed completion by an average of 7.4 months (NAHB and NMHC, 2022). On the entitlement side, projects requiring an Environmental Impact Report spent an average of 504 additional days in the approval period in Los Angeles (UCLA Anderson Forecast, 2023).
This is why the conversation around zoning due diligence software has shifted. The question is no longer whether AI can help. It's how a conversational interface changes the workflow that predevelopment teams already use.
Why Manual Zoning Research Slows Teams Down
Three problems compound on every site.
First, the source material is messy. Municipal codes live in PDFs, supplements, ordinances, and live web portals that update on their own schedule. Overlays, planned unit developments, and special districts each carry their own rules. A single parcel can sit inside two or three of these layers at once, and the relevant text is often distributed across hundreds of pages.
Second, the interpretation is local. A code can read one way and be applied another way by the planning department. Practitioners build that knowledge over years, and sharing it across a growing team is hard. When interpretations conflict late in the process, redesigns get expensive, which is part of why regulation accounts for more than 40 percent of development cost (NAHB and NMHC, 2022).
Third, scale breaks the workflow entirely. Manual zoning research works on one site. It does not work on a portfolio. Brokers, appraisers, and developers screening multiple sites at once cannot read each municipal code from scratch.
The result is that zoning compliance analysis becomes a bottleneck. Teams underwrite fewer deals, push decisions later in the timeline, or rely on a single subject matter expert whose calendar becomes the gating factor for the entire pipeline.
A Conversational AI Workflow for Zoning Due Diligence
The industry is already moving on this. JLL's 2025 Global Real Estate Technology Survey, which polled more than 1,500 senior CRE investor and occupier decision-makers across 16 markets, found that 92 percent of occupiers and 88 percent of investors, owners, and landlords are now running AI pilots (JLL, 2025). Only 5 percent of those organizations report having achieved all of their AI program goals (JLL, 2025). The gap is workflow choice. Access to the technology is no longer the constraint.
For zoning research, the workflow most ChatAEC users settle into looks like this.
1. Anchor the Conversation to a Site
Drop in an address, parcel number, or APN. The system pulls the base zone, applicable overlays, and any special districts that touch the parcel. You now have a single source for the rest of the conversation.
2. Ask the Obvious Questions First
Permitted uses. Maximum height. Front, side, and rear setbacks. FAR or density caps. Parking minimums. These are the inputs every massing study and pro forma needs. A conversational interface returns them in seconds, with citations back to the code section, so you can verify the source rather than trust a black box.
3. Layer in the Overlays
This is where manual research breaks down. Historic districts, transit overlays, design review zones, and PUDs each modify the base. Ask the tool what changes for this parcel under each overlay. The answer comes back in plain language, with the code citation attached.
4. Test the Entitlement Path
Most sites have two answers: what you can build as-of-right, and what you could build with a variance, rezone, or conditional use permit. Ask both. A good zoning compliance analysis surfaces which entitlement levers exist, what the planning department has historically approved, and what the political risk looks like. Given that an EIR alone can add more than 500 days to a Los Angeles approval timeline (UCLA Anderson Forecast, 2023), understanding which path triggers which review is worth time at the front of the deal.
5. Generate a Feasibility Summary
Once the questions are answered, export a zoning report your team and your lender can read. The summary is the artifact. The conversation is the audit trail.
What Changes When You Stop Doing It Manually
Three things shift.
Time to decision compresses. When a parcel's zoning analysis is a conversation instead of a research project, sites that would have taken days of manual review can be screened in an afternoon. Given that regulation accounts for more than 40 percent of multifamily development cost (NAHB and NMHC, 2022), compressing the front of the funnel changes how many deals a team can underwrite within the same window.
Accuracy improves because the citation is always one click away. Real estate due diligence automation only works if your team trusts it. Citing the code section every time is what builds that trust. It also gives staff a defensible record when interpretations get challenged later in the process.
Scale becomes possible. The same workflow that screens one site screens twenty. Property development feasibility analysis stops being a single-site exercise and starts being a portfolio exercise. Brokers can prioritize the listings worth showing. Appraisers can run comparable site assessments quickly. Developers can hold a wider funnel without adding headcount.
Where to Start
JLL's data is a useful reminder that AI pilots are everywhere, but only 5 percent of organizations report hitting all of their AI goals (JLL, 2025). The gap is workflow choice. Picking a use case with a clear before-and-after, like zoning research, is one of the fastest ways to close it.
The best way to see whether conversational zoning tools fit your team is to run a site you already know through one. Pick a parcel where you already know the answer. See how fast the workflow gets you there, and what the citations look like.
Start with 10 free questions at aiaec.com/chataec.