Real estate development looks linear on paper but behaves non-linear in the real world.
IRR is the perfect example — one small change in assumptions can swing the project from “great deal” to “walk away.”
When we started building Feasibility.pro, this was the first thing we wanted to solve: make sensitivity mapping native, not an afterthought.
The Problem: IRR Isn’t a Single Number
Developers often treat IRR as a static output:
“What’s the IRR on this land deal?”
But real feasibility doesn’t care about a single output.
It cares about ranges under uncertainty:
- Different construction costs
- Different absorption rates
- Different sale price assumptions
- Delays in execution
- Changes in financing terms
With land deals especially, sensitivity becomes the only realistic lens.
Step 1: Anchor the Base Case
Every model starts with a “clean hypothetical.”
**
In our workflow:**
- Land cost
- Buildable area
- Product type (res, commercial, mix)
- Construction cost
- Approvals timeline
- Debt assumptions
- Sales/revenue assumptions
This produces the base case IRR.
It’s not the decision-maker, it’s simply the reference point for everything that follows.
Step 2: Identify High-Impact Variables
Not all variables move IRR equally.
From hundreds of model tests, we found these have disproportionate impact:
- Land acquisition cost
- Sale price per unit/ft
- Absorption velocity
- Construction cost
- Financing rates & structure
- Execution delays
Variables like design fees or marketing typically add noise, not signal.
In Feasibility.pro, these are tagged as elastic variables — meaning they are allowed to vary in sensitivity maps.
Step 3: Build the Sensitivity Engine
This is where software beats spreadsheets.
Instead of manually editing cells in Excel, we built a multi-axis model that can sweep through ranges such as:
cost = ±5%, ±10%, ±15%
price = ±5%, ±10%, ±15%
timeline = +3 months, +6 months, +9 months
This produces a matrix of IRR outcomes that shows zones of viability.
It’s much closer to how developers actually think:
- “If land becomes 10% cheaper this works”
- “If absorption slows, we need cheaper capital”
- “If costs escalate, walk away”
Step 4: Encode Deal Logic, Not Just Math
Mathematical sensitivity is useless without deal logic.
For example:
- If absorption slows → debt extension required
- If cost escalates → margins compress → equity return erodes
- If price increases → debt can be refinanced at better terms
In Feasibility.pro, scenarios run through conditional rules (deal behavior), not just recalculated IRR.
This was intentional — real estate is behavioral, not just numerical.
Step 5: Visualizing the Decision Zone
The most underrated part of sensitivity is visualization.
We output a visual decision band where the deal is:
🟢 viable
🟡 borderline
🔴 unviable
This does two things:
- Makes the model usable to non-finance stakeholders
- Forces discipline in acquisition decision-making
Before building this, we saw deals get approved on optimism, not structure.
The Hidden Insight: Sensitivity ≠ Optimization
The aha moment after modeling dozens of land deals:
Sensitivity isn’t about finding the “best” number.
It’s about discovering whether the deal survives reality.
80% of failed projects fail not because models were wrong, but because assumptions were never pressure-tested.
What We Learned About Land Deals Through Sensitivity
Some general truths:
- Land cost has the highest kill-switch impact
- Absorption saves bad deals more than sale price does
- Time hurts IRR more than cost escalation
- Debt magnifies upside and amplifies downside
- Faster approvals outperform cheaper land
- Pricing power masks execution inefficiency (but only in bull markets)
These patterns informed how we structured Feasibility.pro to evaluate land first, project next.
Closing Note (Subtle Founder POV)
When we built Feasibility.pro we didn’t want to replace Excel.
We wanted to replace guesswork.
Sensitivity modeling became the bridge between:
- developer optimism
- financial realism
- market uncertainty
Land deals demand that bridge.
Top comments (0)