ChatGPT, Claude, and Gemini are three versions of the same kind of tool: a general-purpose AI assistant you instruct in plain English, built to draft, summarize, and rewrite the language that fills a commercial real estate day. They come from three different companies — ChatGPT from OpenAI, Claude from Anthropic, Gemini from Google — and the tech press loves to rank them on benchmarks a broker will never run. For a small firm the honest news is more useful and less dramatic: for the work on your desk, the three are more alike than different, and the right choice is decided less by which is “smartest” than by which suite you already pay for and how you plan to handle confidential deal data. This piece explains what each one is, where the real differences show up, and how a 4-to-20-person firm should actually pick.
Before the comparison, a word on why this question comes up at all. Once a firm understands that these tools are all large language models — software that produces fluent text by predicting what comes next, explained in full in our piece on what generative AI is for commercial real estate — the next question is always “so which one do we buy?” That is a purchasing decision, and like most decisions at a lean firm it rewards clarity over analysis. The CRE AI training playbook and the broader case in the small-firm manifesto both make the same point: the tool matters less than the habit. This article gets you to a defensible choice quickly so you can spend your energy on the part that pays — using it.
The One-Sentence Version of Each
Strip away the marketing and each product is a chat box you type instructions into. Here is the plain description of all three.
ChatGPT is OpenAI’s assistant — the tool that made this whole category mainstream. It is the most widely used, has the largest ecosystem of add-ons, and is the one your clients and competitors are most likely to already know by name. For a firm with no existing Microsoft or Google preference, it is the safe default.
Claude is Anthropic’s assistant, known for careful, natural writing and for holding long documents in view without losing the thread. Many people who write for a living prefer its tone: it tends to produce prose that needs less editing to sound like a person rather than a template.
Gemini is Google’s assistant, built into Google Workspace — Gmail, Docs, Sheets, Drive. If your firm runs on Google, Gemini is the one already sitting inside the tools your team opens every morning, which changes the math entirely.
One tool sits slightly outside this trio but belongs in the same decision: Microsoft Copilot, Microsoft’s assistant embedded in Word, Outlook, and Excel. If your firm runs on Microsoft 365 — as most CRE shops do — Copilot is to your stack what Gemini is to a Google shop: the assistant that lives where you already work. We include it here because ignoring it would give you a worse answer.
Why They Feel So Similar
Sit a broker down in front of all three with the same task — “summarize the rent and term details in this lease,” “draft a cold email to an industrial tenant whose lease expires next year,” “rewrite this listing to lead with the clear-height” — and the outputs are close enough that most people could not reliably tell which tool wrote which. That is not a knock on any of them. It is the reality of the category.
The reason is simple: they are trained on overlapping bodies of text, built on the same underlying approach, and updated often enough to stay roughly in step. A gap that looks decisive in a benchmark chart one quarter closes the next. For the first-draft work that fills a CRE week — letters of intent, lease abstracts, market write-ups, listing copy, prospecting emails — all three are strong, and all three carry the same two cautions: they will occasionally state a wrong figure with total confidence, and they are unreliable for exact math on money. Those cautions are properties of the technology, not a mark against any one product.
The takeaway is freeing. You are not choosing between a right answer and three wrong ones. You are choosing among three good tools, which means you can decide on the sensible criteria — cost, fit with your existing software, and data handling — rather than agonizing over a capability race that shifts every few months.
Where the Differences Actually Show Up
The tools are similar, not identical. The differences that a small firm will actually notice fall into three buckets.
Writing feel. On pure drafting — a market narrative, a broker’s opinion of value, a carefully worded email to a nervous seller — Claude and ChatGPT tend to produce the most polished prose out of the box, with Claude often edging ahead on a natural, un-templated voice. This matters most for client-facing copy where “sounds like a robot wrote it” is a real cost.
Living inside your apps. This is the difference that changes a workday. Gemini in Google Workspace and Microsoft Copilot in Microsoft 365 can act on the documents already in your Drive or on your Outlook thread without copy-pasting. Standalone ChatGPT and Claude are a separate window you move text in and out of. For a team that lives in email and shared docs all day, the embedded option removes friction that the standalone tools cannot.
Long documents. All three handle long files, but the assistants known for holding a very large document in view — a full lease, a stack of estoppels, a long offering memorandum — without losing track of an early clause are a genuine convenience for abstraction work. This is a rare place where a capability difference maps to a specific CRE task.
| Best-in-slot strength | Where a CRE firm feels it | |
|---|---|---|
| ChatGPT | Largest ecosystem, familiar to everyone | Safe default when you have no Microsoft/Google preference |
| Claude | Natural writing voice, long documents | Client-facing prose and lease abstraction |
| Gemini | Built into Google Workspace | Firms living in Gmail, Docs, and Drive |
| Microsoft Copilot | Built into Microsoft 365 | Firms living in Outlook, Word, and Excel |
Read that table as tendencies, not verdicts. Verify the current specifics on each vendor’s own page before you commit, because the feature set genuinely changes quarter to quarter.
The Real Decision Rule: Start With Your Stack
Here is the shortcut that saves a lean firm a week of comparison-shopping. Do not start with “which AI is best.” Start with “what software does our firm already run on,” because the assistant that lives inside your existing suite wins on the one metric that determines whether people actually use it: friction.
- If your firm runs on Microsoft 365 (Outlook, Word, Excel), the natural first choice is Microsoft Copilot. It is already in the apps your team opens all day, your IT setup is a toggle rather than a new vendor, and the data stays inside an environment you already trust.
- If your firm runs on Google Workspace (Gmail, Docs, Drive), the natural first choice is Gemini, for the identical reason.
- If you have no strong suite preference, or your priority is the best possible drafting on standalone documents, choose ChatGPT or Claude — ChatGPT for the largest ecosystem and broadest familiarity, Claude for the most natural writing voice and the easiest handling of very long files.
Notice what this rule does. It replaces an unwinnable “which is smartest” debate with a question you can answer in ten seconds by looking at your own email client. For most small CRE firms that answer is Microsoft, which points to Copilot as the sensible starting line — with a standalone ChatGPT or Claude seat for whoever on the team does the heaviest writing. That is a perfectly good stack, and it costs less than one lunch meeting a month per person.
Confidentiality Decides Your Tier
There is one criterion that overrides writing feel and app integration, and a broker feels it the moment they consider pasting a rent roll or an NDA-bound offering memorandum into any of these tools: is my confidential deal data safe here?
The short answer is that it can be, but only on the right tier. The consumer free versions of these products are the wrong place for client data. The business and enterprise tiers — ChatGPT’s business plans, Claude’s team and enterprise plans, Gemini through a paid Google Workspace subscription, Copilot through Microsoft 365 — state that they do not train their models on your business content by default, and they come with the administrative controls a firm needs. That single fact makes the paid business tier non-negotiable for a firm that touches confidential material, regardless of which product you pick.
This is exactly why the tool choice and the data-handling policy are one decision, not two. Whichever assistant you standardize on, put the team on its paid business tier and set a simple rule for what deal data is allowed into it. The full version of that rule — which documents go in freely, which get redacted first, and which never leave the building — is worked out in our confidentiality framework for safe AI use with client deal data. Do not skip it. The confidentiality question is the one that turns a productivity tool into a liability if handled carelessly, and it is entirely manageable if handled on purpose.
Standardize the Team, Then Specialize
The most common mistake a small firm makes here is letting every person pick their own favorite. One broker uses ChatGPT, another swears by Claude, the ops manager is on Gemini because that is what Google put in front of her. It feels harmless. It is not.
When the whole team is on one tool, three things get easier at once: training is a single curriculum instead of three, your data rules are enforceable because everyone sits inside the same controls, and people can share prompts that actually work for the next person. A firm gets far more from one tool used fluently by everyone than from three used tentatively by individuals. The capability gap between the products is smaller than the gap between a team that has standardized and one that has not.
So make the call, write it down, and move on. Pick the one that matches your stack, put everyone on its business tier, and treat the choice as settled for the year. Specialized proptech tools — lease-abstraction platforms, deal-screening software, CRM assistants — can come later, once you know which repeated tasks justify a dedicated product. What changes a firm’s output first is not the perfect tool; it is the whole team getting fluent with a good-enough one, the same lesson running through how AI is reshaping brokerage work: the firms pulling ahead standardized early and trained hard.
What It Costs
Pricing is small relative to almost anything else a CRE firm buys. The paid business tiers land in the range of roughly $20 to $30 per user per month, with Copilot and Gemini priced as add-ons to a Microsoft 365 or Google Workspace subscription your firm likely already pays for. Verify the current number on each vendor’s pricing page, since tiers and names shift, but the order of magnitude is stable: for a 10-person firm, standardizing on one business-tier assistant is a low-hundreds-per-month line item, not a capital decision.
Set against hours spent each week drafting LOIs, abstracting leases, and writing market copy by hand, the subscription is the cheapest part of the equation. The part that decides whether the money is well spent is the training that turns the subscription into a habit — structured team training for CRE tasks typically runs in the low thousands for a small firm, and custom automation, if you later need it, is a larger project in the tens of thousands and up. The tool is cheap; the fluency is where the value is.
Where to Start
You do not have to resolve this alone or guess which tool fits your firm’s mix of brokerage, management, and acquisitions work. A free AI-readiness assessment is a short working session that looks at what software your firm already runs on, how you handle confidential deal data, and which document-heavy tasks would pay off first — and points you at the right tool, the right tier, and a sensible first project. Book a free AI-readiness assessment and you will leave with the decision made and a plan matched to your deals, instead of four browser tabs and a benchmark chart.
Frequently Asked Questions
What is the difference between ChatGPT, Claude, and Gemini?
They are three general-purpose AI assistants from three companies — ChatGPT from OpenAI, Claude from Anthropic, Gemini from Google — and they do fundamentally the same thing: take a plain-English instruction and produce drafted, summarized, or rewritten text. For commercial real estate work like LOIs, lease summaries, and market write-ups, the outputs are close enough that most users could not tell them apart. The differences that matter to a small firm are practical: Gemini is built into Google Workspace, Claude is often preferred for natural writing and long documents, and ChatGPT has the largest ecosystem and broadest familiarity.
Which AI tool is best for a commercial real estate firm?
The best tool is usually the one built into the software you already run on, because that is the one your team will actually use. If your firm is on Microsoft 365, start with Microsoft Copilot; if you are on Google Workspace, start with Gemini. If you have no strong suite preference or your priority is the cleanest possible drafting on standalone documents, choose ChatGPT or Claude. For most firms the marginal capability gap is smaller than the benefit of getting everyone on one tool, so pick by fit with your stack rather than by benchmark scores.
Is ChatGPT or Claude better for writing?
Both produce strong prose, and for many client-facing tasks — a market narrative, a carefully worded email, a broker’s opinion of value — they need less editing than a generic template. Claude is frequently preferred for a natural, un-templated voice and for handling very long documents such as full leases without losing track of early clauses. ChatGPT has the larger ecosystem and is the tool most people already know. For pure writing feel, either is a defensible pick; the practical tie-breakers are cost, data handling, and whether the tool needs to live inside your existing apps.
Should my firm use Microsoft Copilot or Gemini instead?
If your firm runs on Microsoft 365, Copilot is a strong first choice because it works directly inside Word, Outlook, and Excel without copy-pasting, and it keeps data in an environment you already administer. The same logic makes Gemini the natural pick for a Google Workspace firm. These embedded assistants remove the friction of moving text between a separate chat window and your documents, which is often the difference between a tool people use daily and one they forget about. Standalone ChatGPT and Claude remain useful for heavy drafting, and many firms run one alongside their embedded assistant.
Is it safe to put confidential deal data into these tools?
It can be, but only on the paid business or enterprise tier — not the free consumer version. The business tiers of ChatGPT, Claude, Gemini, and Copilot state that they do not train their models on your business content by default and include administrative controls a firm needs. Whichever tool you choose, put the team on its business tier and set a clear rule for what deal data is allowed in, redacting or excluding NDA-bound and personal information. The tool choice and the data-handling policy are really one decision.
How much do ChatGPT, Claude, and Gemini cost for a small team?
The paid business tiers generally run in the range of about $20 to $30 per user per month, sometimes discounted with annual billing. Copilot and Gemini are typically add-ons to a Microsoft 365 or Google Workspace subscription your firm may already have. For a 10-person firm, standardizing on one business-tier assistant is a low-hundreds-per-month expense — small relative to the hours it saves. Confirm the current price on each vendor’s page, since tiers and names change, but the order of magnitude is stable and modest.
Should everyone at the firm use the same AI tool?
Yes. Standardizing the whole team on one assistant makes training a single curriculum, keeps your data rules enforceable because everyone sits inside the same controls, and lets people share prompts that work. A firm gains far more from one tool used fluently by everyone than from three tools used tentatively by individuals, and the capability gap between the products is smaller than the gap between a standardized team and a fragmented one. Pick the one that matches your stack, put everyone on its business tier, and treat the choice as settled.
Which tool should a firm choose if it has no IT department?
Choose the assistant built into the software you already pay for, because it requires the least setup: Copilot for a Microsoft 365 firm, Gemini for a Google Workspace firm. Both are enabled through your existing subscription rather than a new vendor relationship, which matters when no one on staff manages technology full-time. Put the team on the paid business tier for the data protections, agree on a simple rule for confidential material, and you have a working setup without needing technical staff. The harder part is not the software; it is training people to use it well.
Arthur Wandzel
