Skip to content

Can You Outsource Estimating to AI?

The Real Estate Espresso Podcast

6m
Transcript

Original podcast

Where to listen

Summary

Written by a model from the transcript; not yet reviewed.

This episode of the Real Estate Espresso Podcast examines the limits and practical applications of artificial intelligence in pre-construction estimating. Victor Manash argues that treating AI like a magic vending machine that instantly generates accurate bids from architectural drawings is a dangerous fantasy. Instead, he explains a four-stage framework where AI handles discrete, auditable microtasks while experienced human judgment remains in control. The discussion covers everything from setting up business and project contexts to automated quality assurance checks, breaking down quantity takeoffs, and final contract reviews. Listeners learn how to leverage AI as a force multiplier for data and paperwork without risking massive cost overruns or liquidity crunches.

  • Never treat artificial intelligence as a magic vending machine for construction pricing. 0:36
  • Connect both project context drawings and historical cost libraries before running any AI estimating. 1:39
  • Avoid asking AI to summarize messy client scopes and instead rely on human review. 2:40
  • Break quantity takeoffs into microtasks by deploying subagents to count specific assemblies separately. 3:45

Highlights

  • 0:25

    The AI Estimating Fantasyinsight

    Uploading drawings into an AI tool and expecting an instant, ready-to-bid construction estimate is a dangerous fantasy that can bankrupt a developer.

  • 1:34

    Setting Up AI's Brainstrategy

    An effective AI estimating system requires two core buckets of information: project context documents and business context historical cost libraries.

  • 2:45

    Bid Planning and Managementinsight

    Because bid packages are messy and contradictory, a human must personally read core scope documents rather than relying on AI summaries.

  • 3:45

    Breaking Down Quantity Takeoffsstrategy

    To prevent AI from hallucinating or stopping short, quantity takeoffs are broken into microtasks where subagents count individual standard assemblies.

  • 4:11

    Drafting Subcontractor Scope Packagesstrategy

    The AI drafts scope packages, levels subcontractor quotes, factors in material waste and labor adjustments, and flags top project risks.

  • 4:42

    Final Contract and Cash Flow Reviewinsight

    Before any number goes out, AI reviews contracts for unfavorable terms and models cash flow forecasts to prevent liquidity crunches.

  • 5:49

    Treating AI Like a Junior Employeeinsight

    AI should be delegated tasks at the task level rather than holding responsibility for the estimate in its entirety.

Quotes

  • “So how do you actually use AI in pre-construction without introducing massive risk?”
    Unnamed speaker · 1:02
  • “Every bid costs time and money to put together, so the AI takes the incoming bid package, compares the broad scope against historical cost libraries, and generates an order-of-magnitude conceptual estimate.”
    Unnamed speaker · 2:11
  • “For instance, you can convert 2D structural drawings into a quick 3D model to visualize complex builds, or running an automated quality assurance check to ensure the drawings are scaled and rotated properly.”
    Unnamed speaker · 2:59

About this episode

Today, we’re talking about construction—specifically, one of the biggest bottlenecks in any development or value-add project: pre-construction and estimating. If you’ve spent any time on social media lately, you’ve probably seen the hype around artificial intelligence in construction. The pitch usually goes something like this: upload a set of architectural drawings into ChatGPT, hit enter, and out pops a finished, ready-to-bid construction estimate. Well, anyone who has ever actually built anything knows that’s a fantasy. Treating an AI like a magic vending machine for construction pricing is the exact same thing as hiring a freelancer online, handing them a set of blueprints with zero instructions, and expecting an accurate quote. Under-quoting a job is the fastest way to put a contractor out of business—and when your contractor goes broke mid-project, that becomes your problem as the real estate investor. So, how do you actually use AI in pre-construction without introducing massive risk? Recently, we reviewed a comprehensive breakdown of a real-world, four-stage AI pre-construction system. Instead of asking AI to do everything in one giant leap, this approach breaks the estimating process down into small, auditable steps—giving the AI the exact context it needs while keeping experienced human judgment firmly in control. ------------ **Real Estate Espresso Podcast:** Spotify: [The Real Estate Espresso Podcast](https://open.spotify.com/show/3GvtwRmTq4r3es8cbw8jW0?si=c75ea506a6694ef1) iTunes: [The Real Estate Espresso Podcast](https://podcasts.apple.com/ca/podcast/the-real-estate-espresso-podcast/id1340482613) Website: [www.victorjm.com http://www.victorjm.com/](http://www.victorjm.com http://www.victorjm.com/) LinkedIn: [Victor Menasce](http://www.linkedin.com/in/vmenasce) YouTube: [The Real Estate Espresso Podcast](http://www.youtube.com/@victorjmenasce6734) Facebook: [www.facebook.com/realestateespresso](http://www.facebook.com/realestateespresso) http://www.facebook.com/realestateespresso%5D(http://www.facebook.com/realestateespresso) Email: [[email protected] mailto:[email protected]](mailto:[email protected] mailto:[email protected]) **Y Street Capital:** Website: [www.ystreetcapital.com http://www.ystreetcapital.com/](http://www.ystreetcapital.com http://www.ystreetcapital.com/) Facebook: [www.facebook.com/YStreetCapital](https://www.facebook.com/YStreetCapital) http://www.facebook.com/YStreetCapital%5D(https://www.facebook.com/YStreetCapital) Instagram: [@ystreetcapital](http://www.instagram.com/ystreetcapital)

Transcript

0:00
6:16

1,088 words · 1 speaker

Topics

  • AI
  • Fiverr.com
  • ChatGPT
  • Claude
  • HTML

Page updated · transcript and summary are machine-generated