Article
Claude AI Basics for Commercial Real Estate Professionals
Getting Started With Claude
If you've heard colleagues mention Claude and want to actually start using it — not just hear about it — this is the guide to bookmark.
Setting up Claude AI takes about five minutes: create an account at claude.ai, verify your email, choose a plan (Free, Pro, Max, or Team), and pick where you'll use it — browser, desktop app, or mobile. For most commercial real estate professionals, Pro is the practical minimum — the Free plan's limits don't comfortably support real deal work. If you're running Claude constantly through a busy day — reviewing multiple documents back to back, generating comp sets, drafting a full round of correspondence — you'll want to know when it's time to move up to Max, covered below.
Why CRE professionals are adopting Claude specifically
Commercial real estate runs on documents: leases, LOIs, comp sets, offering memoranda, correspondence, reports. Claude was built with strong document reasoning and writing ability, which is why it's become a common starting point for people in CRE rather than other AI tools built primarily for coding or image generation. It reads a 40-page lease, drafts a follow-up email, and summarizes a rent roll in the same conversation — the kind of task-switching a CRE workday actually looks like.
That said, Claude is a general-purpose assistant. It doesn't come pre-loaded with CRE knowledge or your firm's playbook. This series will build toward that. But it starts with getting the account set up correctly.
Step 1: Create your account
Go to claude.ai in any browser.
Sign up with a work email if your firm allows it — this keeps deal-related conversations separate from personal use and makes it easier to add a Team plan later.
Verify your email and set a password, or continue with Google sign-in.
Tip: If your firm has any policy on AI tools and client data, check with your compliance or IT contact before uploading lease documents, financials, or anything under an NDA. This applies to any AI tool, not just Claude.
Step 2: Choose a plan
Plan | Best For | Why |
|---|---|---|
Free | Testing the waters | Enough to try a few quick email draft or test a summary, not enough for a real workday |
Pro | Most Individual CRE users | Enough daily capacity for the core of the job: drafting outreach and correspondence, building comp sets or financial summaries, and turning notes into report or presentation content |
Max | Power users | 5x or 20x more usage than Pro, for anyone running Claude continuously all day — juggling multiple deals or properties at once, producing several reports or presentations back to back, or pairing heavy outreach with ongoing deal or asset work |
Team/Enterprise | Broker Teams/Full Firm | Shared billing, admin controls, a larger context window on some models, and the ability to standardize report and presentation formats across an office or team |
For someone doing large volume of deal or asset work — reviewing documents, drafting comps, prepping for meetings — Pro is the practical minimum. The Free plan's message limits get restrictive fast once you're uploading full documents for review.
Pro is the floor, not the ceiling. If you find yourself regularly hitting usage limits mid-afternoon, or you're the person on the team who lives inside Claude all day, Max is built for exactly that — see the token limits section below before you decide which tier of Max you need.
(Anthropic's plans and limits change periodically — check claude.ai's current pricing page for the latest details before you commit.)
Step 3: Set up Instructions for Claude (your standing profile)
Before you touch privacy or memory settings, set this up — it's the single highest-leverage five minutes in this whole guide, and it's easy to miss because it's tucked into a plain settings menu with no onboarding prompt.
What it is: Go to Settings > Profile, and look for the field asking what preferences Claude should consider in responses (sometimes labeled "Instructions for Claude"). Whatever you write there loads automatically into every conversation you start — you never re-type it, and Claude never asks for it.
What to put in it: Two or three sentences is enough to start.
"I work in commercial real estate, focused on [leasing / acquisitions / asset management / property management]. I write outreach emails, deal summaries, and presentation content regularly. Keep responses direct and work-ready — skip generic disclaimers, use real estate terminology correctly, and default to bullet points for comps or deal summaries."
Why this matters more than most people realize: without it, Claude treats every new chat like a conversation with a stranger — generic tone, no assumed context, no shorthand. With it set, a request like "draft a follow-up to the tenant rep" already knows you're in CRE, without you explaining that every time.
How it's different from memory (next step) and Projects (Part 4 of this series): Instructions are something you write once, deliberately, and they apply everywhere. Memory is something Claude builds passively by observing your conversations over time. They work together, not instead of each other.
Step 4: Set up privacy and memory before you upload anything
This is the step people skip — and the one that matters most once client documents enter the picture. Do this before you upload a single lease.
What "memory" does: Claude can remember details across conversations — the properties you're tracking, your typical deal structure, your writing style — so you don't have to re-explain context every time you open a new chat.
There are two separate memory spaces, and people mix them up:
General memory — built from conversations you have outside any Project. This is the broad, everyday memory: your role, your recurring clients, your general preferences.
Project memory — each Claude Project has its own, completely separate memory space, built only from conversations inside that Project. If you set up a Project for a specific listing, deal, or client relationship, its memory stays scoped to that Project and doesn't bleed into your general chats (or other Projects). We'll cover setting up Projects in Part 4 of this series — worth knowing about now if you're planning to keep one deal's context walled off from another's.
Memory isn't instant. It updates periodically in the background — roughly daily — rather than the moment a conversation ends. That means a chat from earlier today may not show up in what Claude "knows" about you yet. Don't be alarmed if a very recent conversation isn't reflected right away; it catches up.
Recommended setup for anyone handling client/NDA material:
Go to Settings > Privacy and review "Help improve Claude" — turn this off if you don't want conversations used for model training. (Opting out means conversations are retained for 30 days for safety/technical purposes, then deleted, rather than up to 5 years under training opt-in.)
Go to Settings > Capabilities > Preferences and decide on "Generate memory from chat history." A reasonable default: leave it on for general workflow chats, but pair it with rule 3 below for anything sensitive.
Use Incognito mode (the ghost icon, top right of a new chat) for anything client-confidential, NDA-protected, or deal-sensitive. Incognito chats aren't saved to history, aren't added to memory, and are never used for training — closest thing to a "this conversation didn't happen" mode.
Read what's actually stored — this is the step most people don't realize is available. Ask Claude directly, "What do you remember about me?" and it will tell you, in plain language, what's in memory right now. This works for both general memory and (when you're inside a Project) that Project's memory specifically. Use it to correct or delete anything that shouldn't be there.
Check your firm's compliance policy on AI tools before deciding where the memory/incognito line sits for your specific role.
Bottom line: Memory on is a productivity win for your day-to-day workflow (comps, templates, your own writing voice). Incognito is the right tool the moment a specific client or deal name enters the chat. And Instructions (Step 3) plus memory (this step) are doing different jobs — one is what you tell Claude on purpose, the other is what Claude picks up on its own.
Step 5: Pick where you'll work from
Claude isn't limited to a browser tab. Where you set it up depends on how you work:
Browser (claude.ai): No install needed. Good starting point.
Desktop app: Lets Claude work alongside your other open files and apps — useful when you're moving between a lease PDF, an email draft, and a comp sheet.
Mobile app (iOS/Android): For reviewing a lease clause or drafting a quick reply between property visits or meetings.
We'll cover the tradeoffs between these surfaces — and when to use each — in Part 2 of this series.
Token limits explained: why "Pro is the minimum, Max is for power users"
This is the part most setup guides skip, and it's the part that determines whether Claude feels effortless or frustrating on a busy day.
Two different limits are at play, and people often confuse them:
Context window (length limit): how much a single conversation can hold at once — think of it as Claude's working memory for this chat. On Pro and Max, that's 200K tokens per conversation (Enterprise plans get a larger window on some models). A token is roughly ¾ of a word, so 200K tokens is very roughly 150,000 words — call it a 400–500 page document, though formatting, tables, and images eat into that faster than plain text.
Usage limits (quantity over time): how much you can use Claude across all your conversations in a given window. Anthropic uses a rolling 5-hour session limit plus a weekly limit, and these reset on a schedule you can watch in real time.
What "burn" actually looks like in practice:
A CRE workday rarely stays on one task, and that variety is exactly what eats into a session limit. Say a Tuesday looks like this:
Morning: draft a batch of outreach emails or owner/tenant correspondence (each one is small, but they add up fast in a single session)
Midday: pull comps and build out a valuation or underwriting narrative for a deal — more back-and-forth as you refine the story and get feedback on tone
Afternoon: turn notes into presentation or report content — outlining slides, tightening copy, drafting speaker notes
Late afternoon: a quick lease question comes in on an existing deal, handled in a fresh conversation
None of these individually feels heavy, but strung together across a full day, a Pro plan's session limit is where power users start to feel the wall — usually right when the afternoon's presentation or report work needs the most back-and-forth. That's the exact scenario Max is built for: the 5x and 20x tiers multiply your available session and weekly capacity, so a full day of outreach, analysis, presentation-building, and correspondence doesn't get interrupted mid-task.
How to check your usage before you hit the wall:
Go to Settings > Usage (Pro, Max, Team, and seat-based Enterprise plans).
The Plan usage limits section shows two progress bars: your current 5-hour session (plus time remaining until it resets) and your weekly limit.
Note that Claude.ai, Claude Desktop, and Claude Code all draw from the same usage pool — if you're also using Claude Code or the desktop app for other tasks, that usage counts against the same limit as your browser sessions.
If you're consistently running close to the edge, you have three options: wait for the reset, enable pay-as-you-go usage credits to continue past your included limit, or upgrade from Pro to Max (or Max 5x to Max 20x).
Practical ways to stretch your limit regardless of plan:
Use Projects for recurring deal work — documents stored in a project are cached, so referencing them repeatedly costs far less than re-uploading the same lease in every new chat.
Keep project instructions short and save task-specific detail for the chat itself.
Turn off extended thinking for straightforward tasks where you don't need deeper reasoning — it's not needed for something like drafting a routine follow-up email.
Batch related questions into one message rather than sending them one at a time.
Common setup mistakes to avoid
Skipping the work email — makes it harder to upgrade to a Team plan later if your office adopts Claude firm-wide.
Starting on Free and getting frustrated by limits — if you're doing deal work, Pro removes a real friction point in week one.
Uploading sensitive client data before setting up privacy and memory — set memory, training opt-out, and incognito habits before your first client document, not after.
Assuming Pro is enough once you're a daily power user — if you're consistently hitting your 5-hour session limit mid-afternoon, that's the signal to check Settings > Usage and consider Max, not a reason to work around Claude.
Treating it like a search engine — Claude works best with context and a specific ask ("summarize this lease's renewal options"), not a bare keyword search.
FAQ
Is Claude AI free for commercial real estate professionals? There's a free tier, but its usage limits are restrictive for regular deal work like reviewing full lease documents. Most people who use Claude for daily work move to a paid Pro plan.
Can I upload lease documents and offering memoranda to Claude? Yes, Claude accepts document uploads including PDFs and Word files. Confirm with your firm's compliance policy before uploading anything containing client-confidential or NDA-protected information.
Do I need technical or coding skills to use Claude? No. Claude is designed for plain-language conversation. Anyone without a technical background can use it productively from day one.
What's the difference between using Claude in a browser versus the desktop app? The browser version requires no install and is the fastest way to get started. The desktop app allows Claude to work alongside other open files and applications on your computer. Part 2 of this series covers this comparison in depth.
When should I upgrade from Pro to Max? When you're regularly hitting your 5-hour session or weekly usage limit during real work — for example, reviewing several document packages back to back in one afternoon. Check Settings > Usage to see how close you're running to your limit before deciding.
Does Claude remember client-specific details between conversations? Only if memory is turned on and you haven't used Incognito mode. For anything client-confidential or deal-sensitive, use Incognito mode (the ghost icon on a new chat) — those conversations aren't saved to history, added to memory, or used for training.
What's the difference between Instructions, memory, and Project memory? Instructions (Settings > Profile) are rules you write once and that apply to every conversation. Memory is built automatically from your conversations and updates periodically, roughly daily, rather than instantly. General memory covers your conversations outside any Project; each Project has its own separate memory space that only draws from conversations inside that Project. You can ask Claude "what do you remember about me?" at any time to see exactly what's stored.
Does uploading documents to Claude use up my usage limit faster? Yes, document content counts toward both your context window and usage limits — but if you store documents in a Claude Project, they're cached, so referencing the same lease or OM repeatedly across a deal costs far less than re-uploading it in every new chat.
What's next
This is Part 1 of a series on Claude AI for commercial real estate professionals. Part 2 covers Chat vs. Cowork vs. Design — the three ways to work with Claude — and how to decide which one fits a given task, from a quick lease question to a full market survey deck.
