Assessment Matters by Valuebase.ai

Hosted by Will Jarvis

TechnologyInterviews guests

Website RSS feed

Episodes

Latest episode

Aug 2026

Language

About the show

Assessment Matters presented by Valuebase is a weekly show that delves into the world of real estate valuation and property assessment. Our expert guests share their insights on the latest trends and techniques in the field, and offer practical advice for professionals in the industry. Whether you're a seasoned appraiser or new to the field, you'll find valuable information and inspiration. From understanding mass appraisal methodologies to staying up-to-date on the latest technology, our podcast is a must-listen for anyone interested in the world of real estate valuation.

Listen to episodes

60 recent

October 22, 202645 min

85: Will Jarvis | Large Language Models, Prompting, and Free AI Tools for Assessors

Join host Will Jarvis for a solo presentation on the rapidly evolving world of large language models and what they mean for property tax assessors. Will covers the connection between AI and property tax through Sam Altman's Moore's Law for Everything thesis, explains how transformer-based models work and why prompting skill matters, walks through practical use cases like drafting correspondence, summarizing appeals, and enriching parcel data, and demos two free ValueBase tools—ValPal and Ratio Pal—that assessors can start using immediately. <ul><li>00:00 - Introduction and background on ValueBase, Sam Altman's investment, and the macro connection between AI and property tax</li><li>03:00 - The astonishing pace of AI improvement: frontier math breakthroughs and what rapid model advancement means for society</li><li>06:30 - Why humans still matter: agency, real-world interaction, and the enduring role of knowledge workers alongside AI</li><li>08:00 - The wolves-to-dogs analogy and the debate over pausing AI development</li><li>10:00 - Economic context: slowing productivity growth, rising costs of essential goods, and why we need more technological innovation</li><li>14:30 - How large language models work: the transformer architecture, Attention Is All You Need, and text prediction fundamentals</li><li>16:30 - The context problem: why LLMs are like the smartest intern who knows nothing about your office</li><li>18:30 - Practical use cases for assessors: drafting correspondence, summarization, organizing tasks, and data enrichment</li><li>25:00 - The critical importance of verifying model outputs and avoiding "AI slop grenades"</li><li>29:00 - Why good grammar, encouragement, and clear prompting lead to better model performance</li><li>33:30 - Anatomy of a useful prompt: task, context, output format, checks, and asking for clarification</li><li>37:00 - Demo of Ratio Pal: uploading sales data, running ratio study statistics, and visualizing vertical equity</li><li>41:00 - Demo of ValPal: assessor-specific LLM with state property tax codes, IAAO standards, and code execution</li><li>44:00 - Q&A on learning to use AI tools, recommended resources like Andrej Karpathy's guides, and closing remarks</li></ul>

August 26, 202614 min

84: Ellen Brideau | Annual Revals, Equitable Exemptions, and Defending Against AI-Generated Appeals

Join host Will Jarvis as he sits down with Ellen Brideau, a veteran Massachusetts assessor with 20 years in the field and roots in assessment going back to the early 1980s. Ellen shares her perspective on why annual reassessments lead to fairer outcomes, makes a case for broadening the tax base by reconsidering exemptions for nonprofits and institutions, and discusses the importance of good property data and letting assessors inside homes. She also touches on career advice for new professionals entering the field and previews her upcoming conference presentation on identifying and defending against AI-generated property tax appeals. 00:00 - Introduction and Ellen Brideau's background in Massachusetts assessing 01:48 - What the public most misunderstands about property taxes 03:30 - How Massachusetts, Connecticut, and New Hampshire differ on reassessment cycles 05:55 - Policy ideas: broadening the tax base and reconsidering exemptions for nonprofits and institutions 07:48 - Assessment caps, deed-restricted affordable housing, and full market value in Massachusetts 09:22 - Personal lessons from a career in assessment and advice for new professionals 12:00 - AI-generated appeals: how Ellen was caught off guard and what she's learning

August 14, 202630 min

83: Brad Elridge | Splitting the Pizza: Community, Exemptions, and the Case for Property Tax

Join host Will Jarvis as he sits down with Brad Elridge, county appraiser in Lawrence, Kansas, to discuss the case for property tax and how to reframe it as shared community investment. The conversation covers the challenges of vertical equity and valuing properties at both ends of the spectrum, the administrative burden of exemptions and assessment caps, the impact of declining community bonds on public attitudes toward taxation, and lessons from states like California and Florida on what happens when tax policy goes sideways. 00:00 - Introduction and Brad's background spanning private and public sector appraisal 04:22 - Rebranding property tax as shared community expenses 06:42 - Why property taxes feel more painful than income taxes and the case for monthly payments 09:47 - Progressive property taxation, vertical equity, and the challenge of valuing high-end and low-end properties 13:43 - The problem with exemptions: administrative burden, unfunded mandates, and policy creep 15:42 - Assessment caps, inequity, and lessons from California and Florida 20:42 - The need for a property tax policy playbook for policymakers 23:36 - Declining community bonds, cultural shifts, and their connection to anti-tax sentiment 26:30 - Economic signals, retail sales trends, and navigating uncertain times 29:20 - Brad's one-sentence pitch: splitting the pizza as a community

August 13, 202622 min

82: James Cusick | Land Value Tax, Henry George, and the Case Against Speculation

Join host Will Jarvis as he sits down with James Cusick, a former AT&T Bell Laboratories software engineer and adjunct faculty member at Columbia University, who found his way to property tax policy through the Henry George School in Manhattan. The conversation covers the principles of land value taxation, how a single tax on land could replace all other taxes at the local, city, and even federal level, the role of land value capture in transit-oriented development like New York's High Line and subway system, and the problem of speculative land-holding in cities like New York and Detroit. James also shares insights from Japan's public-private partnership model for rail and real estate development, and introduces the interactive tools he's built to help policymakers and the public explore Georgist economics. 00:00 - Introduction to Assessment Matters and guest James Cusick 00:37 - James's background in software engineering at Bell Labs and Columbia University 01:33 - Discovering the Henry George School and the concept of land value tax 02:42 - Building a land value tax calculator covering all U.S. counties 04:38 - How a single land value tax could replace all other taxes, including New York City's $125 billion budget 05:53 - The pitch to policymakers: simplification, administrative efficiency, and reducing speculation 07:42 - Land value capture examples: NYC subway system, the High Line, and Battery Park City 09:33 - The problem of vacant lots and speculative land-holding in cities 11:52 - Property tax as a "ticket to ride" for public services and the investment vs. affordability tradeoff 15:33 - Henry George's anti-speculation philosophy and its modern relevance 17:48 - Japan's public-private partnership model for rail and real estate development 20:19 - Where to find the Henry George School's tools: LVT calculator, Global Georgism Viewer, subway land value tool, and AI Henry George avatar

July 31, 202643 min

81: Will Jarvis | The Big Picture: Why Property Taxes Matter for Economic Growth and How AI Is Changing the Office

Join host Will Jarvis as he delivers a wide-ranging conference presentation covering the macro case for property taxes, the mechanics of AI and large language models, and what these technologies mean for assessment offices. Will draws on his background as a mass appraisal researcher and CEO of ValueBase to explain why property taxes are the most efficient form of taxation for economic growth, how the historical Georgist movement shaped American assessment practices, and why AI tools—from machine learning valuation models to LLMs—are poised to transform day-to-day workflows for assessors while requiring careful validation and human oversight. 00:00 - Introduction to the presentation and Will Jarvis's background in mass appraisal research and founding ValueBase 03:24 - The big picture: total factor productivity, slowing economic growth since the 1970s, and why it matters 06:35 - Why property taxes are more efficient than income and sales taxes for driving economic growth 10:09 - Henry George and the historical roots of American property tax policy 12:33 - Tax revolts, vertical and horizontal equity challenges, and the capacity crisis in assessment offices 15:25 - Three revolutions in valuation: from the Sommer System ledger method to CAMA to the AI era 18:00 - Why AI is happening now: cheaper compute, the Attention Is All You Need breakthrough, and how LLMs actually work 24:24 - Sam Altman's Moore's Law for Everything thesis and its connection to property tax 27:00 - The emerging core skill: managing AI models with real-world context and validating outputs 30:14 - What LLMs are and aren't: not databases, not search engines, and not magic 33:09 - Machine learning valuation models: gradient boosted trees, overfitting risks, and the big five property characteristics 37:22 - Practical LLM use cases for assessors: drafting correspondence, summarizing, research, and brainstorming 39:38 - Key pitfalls: hallucinations, math errors, out-of-date knowledge, and data privacy considerations

July 23, 202623 min

80: Will Jarvis | Community, Service, and the Case for Fair Property Taxes

Join host Will Jarvis for a special impromptu episode recorded at a conference in Charlottesville, Virginia, where Jay — a Norfolk-based assessor and host of the Conversations in the Mirror podcast — flips the script and interviews Will. Together they explore how property taxes fund local communities, the regressive nature of current assessment models, tax abatements for large commercial properties like stadiums, practical AI tools for entrepreneurs and assessors, and what drove Will to build Valuebase from rural North Carolina into an $8 million-funded company serving assessment offices nationwide. 00:00 - Introduction and how Jay's impromptu interview came together at a Charlottesville conference 02:00 - Will's background: growing up in rural eastern North Carolina and attending UNC Chapel Hill 03:30 - The founder mindset: balancing arrogance with humility and outworking the competition 05:00 - What kept Will going through the hardest startup moments: mission over money 06:45 - What Valuebase does: building automated valuation models and tackling regressive property taxes 08:30 - How property owners should think about property taxes as community investment 10:30 - Property taxes as the foundation of local government revenue (65-75% of city budgets) 12:00 - Tax abatements for stadiums and large corporations: who really bears the burden 14:00 - Favorite AI tools: Claude, Grok, Gemini, Perplexity, and why specificity matters over AI hype 16:00 - Practical AI advice: setting up Claude CoWork for daily workflows 17:00 - How Will raised $8 million: podcasting, relationship-building, and asking for advice instead of money 19:00 - Rapid-fire reflections: Will's definitions of success, service, and identity as a founder 21:30 - The 10-year vision: Valuebase in 80% of U.S. assessment jurisdictions and the Jack Dangermond inspiration

July 3, 202629 min

79: Jamie Christensen | Reinventing the Counter: GIS, AI, and the Future of Local Government Software

Join host Will Jarvis as he sits down with Jamie Christensen, founder of Civic Vanguard, a local government software company focused on geographic information systems. Jamie shares his nearly 30-year journey in GIS and local government technology, discusses how AI is enabling a reimagined "digital counter" experience for citizens interacting with their local governments, explores the misaligned incentives in government software procurement and how AI is disrupting legacy vendor lock-in, and outlines his vision for agentic AI tools that orchestrate GIS workflows across localities nationwide. <ul><li>00:00 - Introduction and Jamie Christensen's background in GIS and local government technology</li><li>02:47 - How AI and LLMs are transforming spatial information and the ability to get answers</li><li>05:30 - Founding Civic Vanguard: Reinventing the local government "counter" experience with AI</li><li>10:05 - What's gone wrong with state and local government software: misaligned incentives and vendor lock-in</li><li>12:50 - How AI disrupts legacy software economics: lower costs, easier migration, faster innovation</li><li>16:50 - Free tools and lead magnets: Civic Vanguard's Compass product and no-cost implementation model</li><li>20:30 - Building trust and empathy with local government clients through content and partnership</li><li>23:30 - Why founder-led companies are best positioned to navigate the AI transition</li><li>25:45 - Civic Vanguard's growth plans: expanding beyond Virginia and building agentic GIS tools</li><li>28:15 - Where to find Jamie Christensen online and closing thoughts</li></ul>

June 28, 202634 min

78: Robbe Demey | Land Value Tax, Vertical Equity, and Fixing Property Tax in Europe

Join host Will Jarvis as he sits down with Robbe Demey, a statistician at Valuebase visiting from Belgium, to explore why property tax matters from both philosophical and economic efficiency perspectives. The conversation covers the Georgist case for land value taxation, the dire state of property tax in Belgium where valuations haven't been updated since 1975, and the persistent challenge of vertical equity in mass appraisal models. Robbe Demey also dives into how cross-validation and ensemble modeling techniques are pushing the frontier of property tax assessment, and what it would take to fix property tax across Europe. 00:00 – Introduction and welcome with Robbe Demey, a Valuebase statistician visiting from Belgium 01:30 – Why property tax matters: the philosophical fairness argument and economic efficiency of land value taxation 03:50 – Natural resources, property rights, and the philosophical case for taxing land versus labor and capital 06:00 – Land, violence, and the Georgist proposal for peace through fair resource distribution 08:45 – Land reform lessons from Zimbabwe, South Africa, and Asia — and why land value tax is a better mechanism 11:20 – The state of property tax in Belgium: outdated 1975 valuations, second-toilet quirks, and government secrecy 14:30 – What it would take to fix Belgian property tax through courts, better valuation, and public support 16:30 – EU tax policy, VAT coordination, and why property tax works best at the local level 17:30 – Vertical equity in property tax: why mass appraisal models systematically overtax low-value and undertax high-value properties 22:45 – Cross-validation and holdout sets: why rigorous model testing is essential and underused in the industry 27:15 – The frontier of mass appraisal: better locational data, ensemble models, and the wisdom of crowds approach 31:30 – How to fix property tax in Europe: reducing income taxes, rallying support, and convincing landlords everyone benefits

June 11, 202648 min

77: Will Jarvis | Using Free AI Tools to Strengthen Your Cost Approach

Join host Will Jarvis as he walks through practical, hands-on ways assessors can use free AI tools to improve their cost approach valuations. In this solo episode, Will demonstrates how tools like Claude, ChatGPT, Gemini, and ValPal can be used to extract market depreciation from sales data, build effective age rubrics, standardize condition ratings, generate taxpayer-friendly explanations, and run ratio study checks — all with paste-ready prompts attendees can take back to their offices immediately. 00:00 - Introduction and overview of the session on using free AI tools for the cost approach 02:22 - Why the cost approach breaks down: depreciation tables, effective ages, and condition ratings 04:44 - How time constraints — not lack of knowledge — keep these inputs from being maintained 07:00 - The free AI toolkit in 2026: Claude, ChatGPT, Gemini, and ValPal compared 10:15 - Ground rules for using AI safely: stripping identifiers, verifying outputs, and checking IT policy 13:20 - The prompt pattern every example follows: role, data, task, and proof 16:30 - Extracting market depreciation from your own sales in one prompt, with a live Florida county example 20:00 - Effective age: replacing gut feel with a written rubric and using AI vision on field photos 23:30 - Condition ratings: running outlier scans and drafting a standardized field guide with AI 26:30 - Explainability: generating plain-English cost approach explanations for property owners 28:00 - Closing the loop: running ratio study checks with IAAO formulas via AI 30:00 - The 30-minute weekly routine, common failure modes to avoid, and concrete next steps

May 14, 202625 min

76: NYSAA | AI Tools for Assessors: LLMs, Prompt Best Practices, and Free Resources

Join host Will Jarvis as he presents a live session at a NYSAA meeting on how modern AI tools can help property tax assessors work more efficiently. Will breaks down how large language models actually work, their strengths and weaknesses including hallucinations and privacy concerns, and shares best practices for prompting these models to get the best results. The session covers real-world use cases like drafting taxpayer correspondence, summarizing legislation, and processing refunds, and highlights free tools including ValPal, RatioPal, and other resources built specifically for assessors. 00:00 - Introduction to Will Jarvis and Valuebase's mission to build AI tools for assessors 02:45 - What is a large language model and how the underlying technology actually works 06:30 - Why LLMs mimic reasoning rather than truly reasoning, and what that means for users 07:30 - Real-world use cases: refund processing, email management, and extending office capacity 08:45 - Strengths of LLMs: writing, summarizing, researching, and brainstorming 09:30 - Weaknesses: hallucinations, math errors, out-of-date knowledge, and privacy concerns 12:45 - Best practices for prompting: clarity, context, examples, roles, and even flattery 15:45 - Free tools recommended: Claude, ChatGPT, Gemini, and Perplexity 18:15 - Practical use cases: taxpayer correspondence, legal document summaries, and reusable prompt libraries 19:30 - Introduction to ValPal and its New York State property tax model 21:00 - Audience Q&A: using AI to translate legislation and summarize new exemption laws 23:00 - Overview of additional free tools: RatioPal, AgePal, AskDocs, and the depreciation table builder