How to trust automation in real estate workflows

Every real estate executive is asking the same AI questions. So Chris Kelly and Adam Pase from Stackpoint finally answered them. After building a dozen AI companies and talking to nearly 100 leaders, here's what they found: Real estate owners see AI's potential. But they don't know where to start. The problem? They're asking the wrong questions. Instead of "How does AI work?" they should ask "Where can AI help my workflows?" Chris and Adam built more than a dozen AI companies from scratch. They know what works and what doesn't. Here's their framework for getting AI right: 1/ Think like building architecture • AI has four layers: compute, models, frameworks, applications • Quality depends on every layer, not just the interface • Ask vendors: "Can you upgrade to better models down the line?" Best AI tools were designed around AI's strengths from day one. 2/ Use their four-bucket framework • Retrieve: Find information from documents and systems • Predict: Forecast outcomes from data • Generate: Create content and recommendations • Act: Take actions and trigger workflows Most powerful AI combines all four buckets into one workflow. 3/ Spot AI-native vs band-aid solutions • Legacy vendors add AI features to old systems • AI-native vendors build workflows around what AI does best • Ask: "If you built this today, how would you design it?" It's retrofitting a skyscraper vs building from scratch. 4/ Know what's ready vs what needs humans • Ready for AI: document processing, lease abstraction, tenant screening • Keep humans: complex negotiations, relationship management, strategic planning • Rule: automate where you need speed, keep humans where you need judgment AI excels at consistency and scale, struggles with nuance and strategy. 5/ Start with 2-week experiments • Pick high-friction, low-stakes workflows first • Test one tool, measure results, learn fast • Don't wait for 12-month AI strategies Early momentum builds organizational muscle for bigger wins. 6/ Your Job Is Clearing Red Tape • Remove barriers to small experiments • Protect early adopters from "that's not how we do it" pushback • Make AI fluency expected for managers The biggest barriers are organizational, not technical. You don't need to understand the technology. You need to understand where friction lives in your business. While others debate whether AI will work, smart operators are already running tests and finding wins. What's the biggest friction point in your workflows right now? Check out the full letter in the comments.

Building Trust in Agentic Experiences Years ago, one of my first automation projects was in a bank. We built a system to automate a back-office workflow. It worked flawlessly, and the MVP was a success on paper. But adoption was low. The back office team didn’t trust it. They kept asking for a notification to confirm when the job was done. The system already sent alerts when it failed as silence meant success. But no matter how clearly we explained that logic, users still wanted reassurance. Eventually, we built the confirmation notification anyway. That experience taught me something I keep coming back to: trust in automation isn’t about accuracy in getting the job done. Fast forward to today, as we build agentic systems that can reason, decide, and act with less predictability. The same challenge remains, just on a new scale. When users can’t see how an agent reached its conclusion or don’t know how to validate its work, the gap isn’t technical; it’s emotional. So, while Evaluation frameworks are key in ensuring the quality of agent work but they are not sufficient in earning users trust. From experimenting with various agentic products and my personal experience in building agents, I’ve noticed a few design patterns that help close that gap: Show your work: Let users see what’s happening behind the scenes. Transparency creates confidence. Search agents have been pioneer in this pattern. Ask for confirmation wisely: autonomous agents feel more reliable when they pause at key points for user confirmation. Claude Code does it well. Allow undo: people need a way to reverse mistakes. I have not seen any app that does it well. For example all coding agents offer Undo, but sometimes they mess up the code, specially for novice users like me. Set guardrails: Let users define what the agent can and can’t do. Customer Service agents do it great by enabling users to define operational playbooks for the agent. I can see “agent playbook writing” becoming a critical operational skill. In the end, it’s the same story I lived years ago in that bank: even when the system works perfectly, people still want to see it, feel it, and trust it. That small "job completed" notification we built back then was not just another feature. It was a lesson learned in how to build trust in automation.

The real estate businesses winning with AI aren't making grand, sweeping changes overnight - they're learning by doing. They’re systematically solving problems, building momentum and capabilities that their competitors simply can't match. If you're exploring AI for your real estate business right now, here's our Crawl-Walk-Run framework we use with customers to take them from quick wins to full-scale AI-native operations. 𝟭) 🐢 𝗖𝗥𝗔𝗪𝗟: 𝗤𝘂𝗶𝗰𝗸 𝘄𝗶𝗻𝘀 𝘁𝗵𝗮𝘁 𝗴𝗲𝘁 𝗲𝘃𝗲𝗿𝘆𝗼𝗻𝗲 𝗲𝘅𝗰𝗶𝘁𝗲𝗱 (The first 30-90 days) Start with a few quick-win use cases that make things a little bit easier, less manual, more fun, and more effective. 𝗧𝗛𝗘 𝗖𝗥𝗜𝗧𝗜𝗖𝗔𝗟 𝗧𝗛𝗜𝗡𝗚: Find workflows with measurable ROI - this progress becomes the foundation for your expansion and future investment decisions. Data extraction use cases are often great candidates: think extracting data from loan docs, assembling rent rolls, or asking questions of research files and getting instant answers instead of waiting 4-6 weeks. These first wins get everyone excited and leaning forward rather than hesitating. It's about emotional buy-in. You clear that one hurdle and suddenly people are interested - leaning forward instead of back! 𝟮)🚶🏽♀️ 𝗪𝗔𝗟𝗞: 𝗧𝗮𝗰𝗸𝗹𝗲 𝘁𝗵𝗲 𝗰𝗼𝗺𝗽𝗹𝗲𝘅 𝗰𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲𝘀 (3-6 months) Once you've built confidence with those initial wins, parallelize efforts for more difficult, longer-term engagements. This is where you start asking deeper questions: How do we unite different data sources? What patterns are we missing? How can we be more proactive rather than reactive? The walking phase connects your basic automation to more impactful decision-making processes. You're going to want to have thought about these longer term, more difficult potential use cases for an AI-native platform. 💭 𝟯) 🏃♀️ 𝗥𝗨𝗡: 𝗕𝗲𝗰𝗼𝗺𝗲 𝗮𝗻 𝗔𝗜-𝗻𝗮𝘁𝗶𝘃𝗲 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀 (12+ months) Now you're ready to tackle truly transformative use cases that fundamentally change how your business operates. Think combining tenant sentiment analysis with lease renewal options to understand what's driving high vacancy rates. Is it one problematic unit causing 80% of issues? These larger scale, data-unifying approaches help you get ahead and make proactive decisions that transform your business. It's about thinking through how to get ahead and how to be proactive in terms of impactful decision-making in your business. ___ You’ll get the most value out from working with us if you’re already marinating with excitement and with potential ideas that we can help you develop before we start working together. 🧠💭 I’ve found this to be absolutely critical for success. It sets us up nicely to be strategic partners where we’re tackling problems one by one and building it up from there. Want to discuss where your business is on this journey? Book a chat: https://lnkd.in/eiuUnQjk