You can buy AI safely without an IT department if you follow rules that put the burden of proof on the seller instead of on a technical team you do not have. A 4–20 person commercial real estate firm has no engineer to read the code, no security officer to vet the data terms, and no analyst to run a bake-off. The buyer, the user, and the admin are usually the same person — often the owner. These ten rules are written for that person. Each one replaces a capability a bigger firm would delegate with a decision you can make yourself, in plain language, before any money changes hands.
Most software-buying advice assumes a procurement team, a security review, and an IT function that can pressure-test a vendor’s claims. A ten-person brokerage has none of that, so the standard playbook — run an RFP, have IT audit the integration, route the contract through legal — is either impossible or wildly out of scale. What follows is rewritten for the firm that makes the call in a partner meeting on a Thursday, where the person signing is also the person who will use the tool every morning.
Why the Usual Buying Advice Fails a Small Firm
The advice fails because it delegates. “Have your security team review the data terms” is useless when you are the security team. “Run a proof-of-concept with your analysts” assumes analysts to spare. The generic guide solves for an organization with specialists; you are solving for one busy principal who cannot personally read a model’s source code and should not have to.
That constraint is not a weakness to apologize for — it is the design brief. A small firm moves faster than an enterprise precisely because one person can decide. The rules below convert that speed into safety, so you can buy in a week what an enterprise takes a quarter to approve, without inheriting the mistakes that fast decisions usually carry. The same discipline of buying capability deliberately, rather than accumulating logins, is the through-line of the approach that lets small firms out-operate far larger ones.
The 10 Rules of Buying AI Without an IT Department
Read these as a posture, not a checklist for a single meeting. They span the whole purchase — what to do before you shop, how to decide in the room, and how to roll out afterward without creating a mess. When you are in the room with a specific vendor, pair them with the ten-question vendor-evaluation checklist that turns this posture into a pass-or-fail scorecard.
1. Start From the Workflow That Bleeds Hours, Not the Tool in the Demo
Name the problem before you look at a single product. The most common buying mistake is falling for an impressive demo and then hunting for a place to use it. Write down the one workflow where your team’s hours go to die — abstracting leases, triaging broker blasts, reconciling CAM, drafting market write-ups, chasing rent-roll updates — and buy only against that. A tool that ranks inbound deals brilliantly is worthless if your bottleneck is lease abstraction. Start from your pain, and the shortlist writes itself.
2. Get Fluent Before You Buy
The cheapest AI upgrade at a small firm is not a tool — it is a team that knows how to use the ones already on their desk. A few hours of hands-on training on ChatGPT, Claude, or Gemini, applied to your real work, does two things at once. It solves a surprising share of tasks with software you already pay for, and it makes you a competent buyer, because you can finally tell a genuine product from a repackaged prompt. Fluency training for a small team typically runs in the low thousands, far below the price of the wrong subscription. Buy the skill first; it lowers the price of every decision that follows.
3. Never Trust a Demo — Make It Run on Your Worst Data
The accuracy that matters is on your ugliest files, not the vendor’s polished sample. AI tools read tidy, well-formatted documents well and stumble on the scanned, amended, typo-ridden ones that fill a real CRE inbox. Hand any vendor ten of your own documents during evaluation, including the two worst you can find — a scanned T-12, a lease with three amendments — and check the output line by line. The failure to fear is not an obvious error, which anyone catches, but a clean-looking number pulled from the wrong column. A vendor who will only demo on their own clean deal is hiding the accuracy you actually care about.
4. Treat “AI” as a Claim to Verify, Not a Badge to Trust
Many proptech tools now advertise “AI” that turns out to be a thin layer over a general model, sold at a steep markup. That is not automatically bad — packaging, workflow, and support can be worth paying for — but you should know what you are buying. Ask one question directly: what does your tool do that we could not do ourselves with a general AI tool and a good prompt? A real product answers with specifics — connections to your systems, a pipeline trained on CRE documents, a maintained workflow. A vendor who deflects is charging you a premium for something you could rent for a fraction of the price.
5. Make the Contract Your IT Department
A contract is the only security department a small firm has, so it has to do that job. Your deal packages are confidential and often under NDA, which means “your data is secure” — a marketing line — is not good enough. Before you upload anything, get three answers in writing: where is our data stored and who can access it; do you train shared models on our uploads (the answer you want is no); and how do we export and delete our full history, and how fast. A data-processing addendum that commits the vendor on training and deletion is enforceable in a way a sales assurance never is. Treat vagueness here as a reason to walk.
6. Price Your Own Hours — the Sticker Is Never the Cost
The monthly fee is rarely the real number. Setup, integration, training, and the hours to maintain a tool usually make up most of the first-year total, and those hours are yours. Price twelve honest months: subscription, onboarding, and the staff time the tool consumes before it saves any. For a custom build, the market range runs roughly $25,000 to $150,000 as a one-time project depending on scope, plus a modest monthly run cost — the full breakdown of what moves that figure is in the guide to what a custom AI automation project costs a small CRE firm. Compare the true cost against the labor hours the tool returns. If payback does not arrive inside a year for a weekly task, question the purchase.
7. Keep a Human on Every Number That Moves Money
Any tool that touches a dollar figure or a go/no-go decision must show its work and leave the call to a person. A screening score with no visible reasoning is a liability the first time it ranks a bad deal highly. Require that every extracted figure links back to its source page, so you can verify the ones a decision hinges on without re-reading the whole document. The right division of labor is fixed: AI does the reading and drafting, a human makes the judgment. A vendor who pitches the tool as replacing your analyst’s judgment is describing a risk, not a feature.
8. Buy One Tool at a Time and Prove It Before the Next
Login sprawl is how small firms waste money on AI. The pattern is familiar: a firm buys three tools in a quarter, adopts none of them, and pays for all of them a year later. Buy one, roll it out, and require it to prove real time savings on a real workflow before you evaluate the next. This also protects your team, who can absorb one new habit at a time but not five. The real cost of proptech subscriptions nobody uses is not the monthly fee — it is the workflow you never fixed because attention was scattered across dashboards.
9. Own Your Exit Before You Sign the Entry
You should own your outputs and be able to leave. A tool that locks your structured lease data or deal pipeline inside its own format recreates the vendor trap you were trying to escape. Confirm before signing that you can export your full data in a usable format on demand, and that leaving does not mean starting over. This matters most with newer vendors in a churny proptech market — a good young company can be an excellent choice, provided a shutdown would be an inconvenience rather than a catastrophe. The subscription you can walk away from is the one with real bargaining power.
10. Know When the Honest Answer Is Build — or a Prompt You Already Own
Buying is one of three answers, not the only one. For most standard tasks at your volume, an off-the-shelf subscription is the fastest and cheapest path. For a workflow specific enough that no product fits, or one that crosses systems no vendor connects, a custom build earns its cost. And for a task you do occasionally, the honest answer is often a good prompt over a tool you already pay for — no new purchase at all. Default to the cheapest option that actually works, and make the more expensive one earn the upgrade with real volume data. The full decision framework, with the three-year math behind each path, is worked through in the buy-versus-build playbook and its companion analysis of proptech subscriptions versus a custom automation project.
The Rules in One Page
The ten rules split cleanly across the arc of a purchase — before you shop, in the room, and after you sign.
| # | Rule | When it applies |
|---|---|---|
| 1 | Start from the workflow, not the demo | Before |
| 2 | Get fluent before you buy | Before |
| 3 | Prove it on your worst data | In the room |
| 4 | Verify “AI” is real, not a wrapper | In the room |
| 5 | Make the contract your IT department | In the room |
| 6 | Price your own hours | In the room |
| 7 | Keep a human on every number | Rollout |
| 8 | One tool at a time, prove before the next | Rollout |
| 9 | Own your exit before the entry | Contract |
| 10 | Know when to build — or use a prompt you own | Before |
None of these rules requires a technical background. Each one asks the seller to prove something, or asks you to decide something in plain business terms — which is exactly what a firm without an IT department needs its buying process to do.
Frequently Asked Questions
How do I buy AI software if I don’t have an IT department?
Put the burden of proof on the seller. Name the workflow you want fixed, then make the vendor prove the tool on your own worst documents, answer three data questions in writing, and price the full twelve months including your hours. A contract that commits the vendor on data training, deletion, and support is the small firm’s substitute for an IT and security team. Buy one tool at a time and require it to prove real time savings before you evaluate the next. Every step is a business decision you can make in plain language, not a technical audit.
How do I know if a proptech tool is real AI or just AI-washing?
Ask the vendor exactly what the tool does that you could not do yourself with a general AI tool like ChatGPT or Claude and a good prompt. A real product answers with specifics: connections to your systems, a pipeline trained on CRE documents, a maintained workflow, and support. A thin wrapper deflects or talks only in benefits. This does not make a wrapper worthless — packaging and workflow can be worth paying for — but you should know you are paying a markup for convenience, not for technology you could not otherwise access, and price it accordingly.
How do I protect confidential deal data when buying AI without a security team?
Get three answers in writing before you upload anything: where the data is stored and who can access it, whether the vendor trains shared models on your uploads (you want a firm no), and how you export and delete your full history. A data-processing addendum that commits the vendor on training and deletion is enforceable; “your data is secure” is not. Because deal packages are confidential and often under NDA, treat any vague answer here as a reason to walk rather than a detail to sort out after signing. The contract is the only security department a small firm has.
How much should a small CRE firm budget for AI?
It depends on buy versus build. A short, hands-on fluency workshop for a small team typically runs in the low thousands, roughly $2,000 to $15,000 depending on scope. Off-the-shelf subscriptions are usually a monthly per-seat or per-firm fee. A custom automation project generally lands between $25,000 and $150,000 as a one-time build plus modest monthly run costs. The subscription sticker is rarely the full cost — setup, integration, training, and maintenance make up most of the first-year total, so price twelve months with your own hours included.
Should I buy a proptech subscription or build a custom automation?
Buy when a standard tool fits your asset classes and volume, because a subscription starts faster and cheaper. Build only when your workflow is specific enough that no product fits, or when it crosses systems no vendor connects. Most 4–20 person firms should default to buy and earn the build case with real volume data. Many land in the middle: a general tool for standard tasks plus custom prompt workflows for the parts unique to how the firm makes money. For an occasional task, the cheapest correct answer is often a prompt over a tool you already own.
Can I just use ChatGPT or Claude instead of buying a proptech tool?
Often yes, for the first pass and at low volume. A well-written prompt over ChatGPT, Claude, or Gemini plus a spreadsheet will extract and summarize many documents using tools your firm already pays for. The limits: general models can confidently misread a figure off a messy scan, and they do not connect to your inbox or keep a pipeline, so you manage by hand. Use them to get fluent and to prove whether you need a dedicated tool at all — and verify every number they pull before it moves money.
What is the biggest mistake small firms make when buying AI?
Buying the demo instead of the workflow. A polished demonstration sells a capability the firm may not need, and the tool ends up as one more unused login. The second most common mistake is buying several tools at once, adopting none, and paying for all of them a year later. Both come from shopping for products before deciding which of your own workflows is worth fixing first. Name the problem, buy one solution, prove it, and only then look at the next.
How do I test an AI tool before I trust it with real work?
Run it on ten of your own documents during the evaluation, including the two ugliest you can find, and check the output line by line against the source. The error to hunt for is not an obvious mistake — those get caught — but a clean-looking, wrong number pulled from the wrong place. Require that every figure the tool extracts links back to its source page so a person can verify the ones a decision depends on. A vendor who will only demonstrate on a clean sample deal is hiding the accuracy that matters most to you.
Do I need to hire a technical person before buying AI?
Not to buy well. The rules that protect you — proving tools on your own data, getting data terms in writing, pricing your hours, keeping a human on every number — are business judgments, not technical ones. What helps far more than a hire is a few hours of fluency training so you and your team can tell a good tool from a bad one and handle simple tasks yourselves. A technical partner becomes worthwhile later, when you have proven a workflow is worth automating and want it built to run on data you control.
Where to Start
Rules are only useful once you know which workflow to point them at, and most firms start shopping for tools before they have decided which problem is worth solving first. That inventory — which of your workflows a tool should fix, in what order, and whether the honest answer is a subscription, a build, or a prompt you already own — is the real first step. A free AI-readiness assessment produces exactly that: a working session that maps your firm’s workflows, flags the ones where a tool pays for itself, and hands you a ranked plan with real cost ranges before you sit across from any salesperson. Book a free AI-readiness assessment if you want that map first. If no purchase is the right spend for your firm right now, the assessment will tell you so — and you will still leave with the plan.
Arthur Wandzel
