Issue #42: The Five Rules of Land Development (Still True, Now Faster)

In partnership with

I’ve told a lot of stories in this newsletter—some funny, some painful, most involving cold calls or culverts. But this week, I’m not telling a story. I’m handing you something better: the five rules that built my career in real estate development.

These are the principles I’ve used to launch projects, fix disasters, and earn (or save) millions. They worked in 1998. They work in 2025. And now—with the right AI tools in your corner—they work even faster.

Also… this issue is short. Why? Because when it hits your inbox Friday morning, I’ll be waking up at a cottage in Ontario with these guys (👇), pretending it’s a very serious professional development retreat (don’t tell CRA). Spoiler: It’s mostly beer and back pain.

Advertisement

You’ve never experienced business news like this.

Morning Brew delivers business news the way busy professionals want it — quick, clear, and written like a human.

No jargon. No endless paragraphs. Just the day’s most important stories, with a dash of personality that makes them surprisingly fun to read.

No matter your industry, Morning Brew’s daily email keeps you up to speed on the news shaping your career and life—in a way you’ll actually enjoy.

Best part? It’s 100% free. Sign up in 15 seconds, and if you end up missing the long, drawn-out articles of traditional business media, you can always go back.

Check it out

The Top 5 Rules of Land Development

What 30 years of deals taught me about not going broke

Three decades in development, and if I could go back and tattoo five rules on my 25-year-old forehead, these would be them. Not because they're the smartest or sexiest rules - but because breaking any one of them can sink your entire project faster than you can say "soil contamination."

Quick disclaimer: Like most of my stories, I've changed dates, locations, and names to protect the innocent (and keep myself out of court). All the stories are real and happened - I just don't want to go to jail over a newsletter. Development's a rough business, but orange jumpsuits clash with my complexion.

Special Note

This week, I've shaken things up a bit. Instead of the usual Land Tech section at the end, I've embedded those tech and AI-powered insights directly under each rule. Why? Because in 2025, technology isn't some separate thing we tack on after the "real work" is done – it's baked into every step of the development process.

Each rule still stands on its decades-old wisdom (learned the hard way, as usual). But now you'll see exactly how today's tools transform these timeless principles from "experience you'll eventually gain" to "advantage you can have by Friday." No separate section, no artificial division – just like your actual workflow should be.

Rule #1: Never Pay Too Much for Land

The mistake that ruins everything else

Fall 2008. Three months after Lehman Brothers collapsed, and I'm staring at a purchase agreement for $18M on a residential site in northwest Calgary. The seller's desperate - bought it for $23M eighteen months earlier when the market was screaming.

ME: "Market's pretty ugly right now. I'm thinking $14M."

SELLER: "I can't take that kind of loss. I've got $19M into this thing."

ONTARIO BOSS: "Greg, what does your pro forma say?"

ME: "At $18M, we're barely making 15% return. At $14M, we're at 22%."

SELLER: "Look, I'll come down to $17M. That's my final number."

My Ontario boss pulls me aside: "Greg, his cost basis isn't our problem. Pay what the deal supports, not what makes him feel better."

I wanted to help the guy out. He seemed decent. But my boss was right.

We walked away from the deal with the Vendor stuck at $16M.

Two years later, that same site sold for $12M.

The lesson: Land price sets the floor for everything else - if you're wrong here, nothing else matters.

Land Tech: In 2008, we consulted with brokers and utilized land title information for land transaction comparables. Lots of meetings, lots of emotion, lots of opinions, lots of time. I thought our team were technology masters using SPIN2 like wizards!

Now, after dozens of late-night AI rabbit holes on Twitter, I have found that AI can analyze thousands of comparable land sales across multiple markets simultaneously, factoring in historical appreciation rates, zoning changes, and development trends.

It can generate custom pro forma scenarios in seconds, stress-testing your assumptions across different market conditions. It appears that most of this technology remains proprietary, so stay tuned for a local solution coming soon.

However, something financial-focused that can help right now, (and unlike my gut-feel approach in 2008), ChatGPT can run Monte Carlo simulations showing the statistical probability of hitting your return targets at different purchase prices. You'll see precisely how each $100K in land cost affects your entire profit structure throughout the project lifecycle.

Just one spreadsheet upload away!

Rule #2: Hire the Right Team From Day One

Because fixing bad hires costs more than anything else

Spring 2017. I'm sitting in a Tim Hortons with three general contractor bids spread across the table for a 200-unit, 13-storey condo building. Market's starting to recover from the Oil Crisis, but GCs, trades and suppliers are still hungry.

Bids: $55M, $65M, and $75M.

The low bidder slides his business card across: "Greg, we're the special projects division of Allied Construction. Small lean and mean team, but we get to use all the parent company's technology and bonding capacity. We do projects exactly like this."

ME: "Your timeline?"

LOW BIDDER: "Twenty-four months, guaranteed. We're motivated."

MIDDLE BIDDER: "I'd say twenty-eight months. Weather can be unpredictable."

HIGH BIDDER: "Thirty months. But we'll be done right the first time."

My construction lead, Patricia, leans over: "The low guy seems hungry. Might work harder for us."

ME: "Or might cut corners to hit that price."

PATRICIA: "It's $10M difference, Greg. That's real money."

I almost went with hunger over experience. Almost.

Then I called three of the low bidder's references. Two didn't call back. The third said: "Nice guy, but he left us hanging when a bigger job came along."

We went with the middle bid. Project finished in twenty-six months, a couple of change orders, and Tom became our go-to GC for the next five years.

The lesson: The wrong team will cost you more than the right team charges. Do your homework.

Land Tech:

Imagine dumping those three GC bids into ChatGPT with a simple prompt: "Compare these proposals and flag any concerning terms."

In minutes, you'd get an analysis that would take hours manually:

"Bid #1 excludes winter conditions yet commits to a January start date."

"Bid #2's payment schedule requires 40% upfront versus industry-standard 20%."

"Bid #3 has no defined change order procedures or rate card for extras."

The AI wouldn't just compare bottom-line numbers—it would catch the exclusions buried on page 37 that balloon costs later. You'd walk into that Tim Hortons meeting armed with specific questions about those sneaky clauses, not just gut feelings.

The lowest bidder might still look hungry, but you'd know exactly which "standard exclusions" to negotiate before signing. You'd catch the material escalation clauses with no caps, the undefined allowances, and the warranty limitations that didn't match your specs.

No, AI won't tell you if the contractor ghosts clients when bigger jobs come along (that's still what reference calls are for…[AI agent for that too]). But it ensures you're comparing apples to apples before committing to that seven-figure contract.

That $10M difference between bids? Might not be such a difference after all.

Rule #3: Entitle the Right Product

Because changing your mind later is brutal

Summer 2014. I'm in a stuffy planning office, sweating through a pre-submission meeting about a 40-unit townhouse project.

PLANNER: "Townhomes are taking 8-10 months for approval right now."

ME: "What about single-family? The site could work for that."

PLANNER: "Singles are moving in 4-5 months. Much cleaner process."

MY ARCHITECT: "We'd lose density, but the approval timeline might make up for it."

ME: "What's the revenue difference?"

NORM (PROJECT MANAGER): "Singles net about $1.3M each. Townhomes about $650K each, but we get 40 units instead of 20."

Quick math: $26M versus $26M gross revenue. Similar margin.

ME: "I don't want to wait 10 months. Let's go with singles."

PLANNER: "Smart choice in this market."

Three months into approvals (and no turning back), the townhome market exploded. Our competition was selling presales for $780K per unit while we were stuck with our single-family approvals.

That decision to chase a faster approval cost us about $5.2M in potential revenue.

The lesson: Build what the market will want when you deliver, not what's easier to approve today - but honestly, nobody has a crystal ball, and market timing is part luck, part homework. It’s hard to predict an energy crisis, followed by a global pandemic, followed by a supply and affordability crisis. More of a tragedy than a comedy.

Land Tech: Instead of relying on planner timelines alone, here's a simple way to use Claude or ChatGPT to make better product decisions:

"Analyze current housing market trends in [your city]. Compare single-family versus townhome demand projections for the next 24 months. What leading indicators suggest changes in buyer preferences?"

The AI response might reveal:

"Recent data shows townhome sales increasing 12% year-over-year while single-family plateaus. Three key leading indicators suggest continued townhome momentum: 1) Mortgage rate increases affecting affordability, 2) Remote work policies stabilizing at 2-3 days/week, and 3) First-time buyers delaying purchases but maintaining density preferences."

Follow up with:

"What's the historical approval-to-market delivery timeline gap for both product types in this region? How have previous market cycles affected product preferences during similar approval timeframes?"

This simple query might have revealed what I missed in 2014 – that townhome demand accelerates in exactly the market conditions we were entering, and that the additional approval time would have been worth the wait.

No complex systems needed – just targeted questions that force you to look beyond the immediate approval timeline to what buyers will actually want when you deliver. Two prompts, five minutes, potentially millions in better decisions.

Better to spend 10 minutes with Claude than 10 months undoing an approval.

Rule #4: Read Every Single Contract

The boring stuff that saves your ass

Two weeks ago, 9:30 PM in my home office. I'm reviewing a grading contract that's supposed to be signed tomorrow morning. Thirty-two pages of technical specifications and I can barely keep my eyes open.

Page 18, buried in subsection 4.3: "Excess soil removal charged at $35 per cubic meter beyond 500 cubic meters."

I almost skipped right over it.

Next morning at the construction trailer:

ME: "This excess soil clause - what are we looking at volume-wise?"

GRADING CONTRACTOR: "Hard to say until we dig. Could be 200 extra cubic meters, could be 800."

ME: "At $35 per cubic meter, that's potentially $28K we didn't budget."

CONTRACTOR: "Yeah, but it's in the contract."

ME: "It wasn't in our discussions."

CONTRACTOR: "Contract supersedes discussions."

Thirty minutes of back-and-forth later, we'd capped the excess at 300 cubic meters at $25 per meter.

Final bill came in at 720 cubic meters of excess. That boring contract review saved us $17,700.

The lesson: The stuff that seems too boring to read is exactly the stuff that will cost you.

Land Tech: Contract review is where AI truly shines beyond human capability. While I nearly missed that critical soil removal clause on page 18, AI can scan a 100-page contract in seconds, flagging every potential cost exposure, unusual term, and deviation from standard industry language. It can quantify the financial risk of each clause, estimate probability of triggering contingencies, and compare terms against thousands of similar contracts to identify what's missing or unusual.

That $17,700 I saved through careful reading? AI would have flagged not just that clause, but likely dozens of other potential cost exposures I completely missed, even after staying up until 9:30 PM reviewing.

If you think you are starting to see what you think are hallucinations, like this, double-check your work with another AI chat like:

Gemini, Claude, ChatGPT, Grok, Qwen, or Deepseek.

Rule #5: Everything Is Negotiable (When You Do Your Research)

Knowledge is your best negotiating tool

Development permit meeting, fall 2013. The city's development officer slides a list of requirements across the table:

DEVELOPMENT OFFICER: "You'll need a traffic impact assessment for these twelve single-family homes."

NORM (PROJECT MANAGER): "That's going to cost us $25K minimum."

ME: "What's the calculation that triggers this requirement?"

OFFICER: "Alberta Ministry of Transportation standard. Ten trips per dwelling. So you're at 120 daily trips."

ME: "But that's based on suburban locations. This is inner-city with transit access. The Ministry has urban adjustment factors."

OFFICER: "I'm not familiar with those."

I'd spent two days studying traffic engineering manuals after getting burned on our last project. And back then, there wasn't a lot of AI to help – just me, coffee, and hundreds of pages of technical guidelines.

ME: "Section 4.2 allows for a 30% reduction in urban, transit-served locations. That puts us at 84 daily trips, below your 100-trip threshold."

Fifteen minutes later, we'd eliminated the $25K traffic study requirement.

But here's the thing - six months earlier, I wouldn't have known to question it. I would have just paid the $25K and complained about government bureaucracy.

Always be prepared.

The lesson: You can negotiate anything, but only if you understand it better than the person across the table. And research has never been easier in the history of humankind than it is right now with AI – what took me two agonizing afternoons of manual reading could probably be done in 30 minutes today or less.

Land Tech: Two days studying Alberta Ministry of Transportation manuals? In 2025? That's like flood plain mapping with an abacus. (I was a floodplain mapper out of university; they called it a stormwater management engineer.)

Using Perplexity:

Conclusion of Perplexity response:

See the full response here.

No shade to my 2013 self – that research hustle was legit. But in 2025, you're leaving money on the table if you're not using AI to prep for these negotiations.

The playing field has changed, and the folks across the table are counting on you not knowing that.

Start learning AI in 2025

Keeping up with AI is hard – we get it!

That’s why over 1M professionals read Superhuman AI to stay ahead.

• Get daily AI news, tools, and tutorials

• Learn new AI skills you can use at work in 3 mins a day

• Become 10X more productive

Sign up and start mastering AI

Do you need a development co-pilot? After 30 years in the trenches, I help developers avoid expensive mistakes through strategy sessions, deal analysis, and road mapping.

Save time and money. Why navigate the development maze alone when carrying costs are eating your returns?

Hit reply with "DEVELOPMENT 911" for a no-obligation call about your project challenges, or share your biggest development headache for a free quick take.

See you next Friday.

- Greg