Lead Attribution That Doesn't Lie: A Practical Model for Real Estate Marketing
How to track where deals actually come from when your buyers take months to decide.
Published on
About a 4 min read.
Lead Attribution That Doesn't Lie: A Practical Model for Real Estate Marketing
Your marketing dashboard says Google Ads brought in 47 leads last month. Your sales team says none of them closed. Your CRM has no idea where half your deals actually started.
This gap isn't a reporting problem. It's a model problem.
Real estate deals don't happen in single moments. A buyer sees a listing on Facebook. Weeks later they click an email. They call after reading a blog post. They tour a property. They ask their agent for comps. They finally make an offer three months after that first touchpoint.
If you give all credit to Facebook, your email program looks worthless. If you give all credit to the final call, you're ignoring everything that warmed them up. If you split credit evenly, you learn nothing actionable.
You need a model that's simple enough to defend to leadership and flexible enough to handle real buyer behavior.
The Three-Touch Model
Start here: track the first, middle, and last touchpoint before a lead converts.
First touch: This is awareness. How did they find you? Search, social, referral, direct. This source matters because it tells you where your market is paying attention. But it almost never closes deals alone.
Middle touch: This is engagement. After they've heard of you, how do they interact? Email open, content download, property view, phone call. This touch is where qualification happens. It separates serious buyers from window shoppers.
Last touch: This is decision. What convinced them to act in the final 48 hours? Probably a showing, a new listing alert, a price reduction, or a direct call from your agent.
Assign weight like this: 20% to first touch, 30% to middle touch, 50% to last touch. Adjust the percentages later if you get better data. The point is defensibility, not perfection.
Why This Works
This model survives the real estate timeline. You capture awareness (Google matters). You capture engagement (your email and content actually count). You capture the close (your agents' work shows up).
It also tells you something actionable: if Google has high first-touch credit but low last-touch credit, your ads reach people but don't warm them up. If email has high middle-touch credit, your nurture sequence works. If phone calls have high last-touch credit, your team's follow-up closes deals.
You can measure each source's strength at each stage. That's the opposite of guessing.
How to Implement It
You don't need fancy software to start. You need three fields in your CRM:
First touch source (required on all leads)
Most recent touch before conversion (required at qualification)
Conversion touch source (required at deal stage)
When someone books a showing, you have two of those three. When they make an offer, you have all three. Work backward from closed deals and audit the data once a month.
This catches problems: leads with no recorded first touch, contacts jumping straight to conversion with no middle touch, deals where the source doesn't match the timeline.
The Offline Problem
Real estate lives offline. A buyer calls after seeing a yard sign. Your agent doors knocks and gets a referral. Someone overhears a conversation at the gym.
This model assumes you're capturing those touchpoints. You're not, yet. Start now:
Add a phone field to your web forms and ask "How did you hear about us?"
Train your agents to log source on every new contact
Tag inbound calls with source based on what they say
Mark referrals and repeat clients as "previous buyer/seller" at first touch
You'll never capture 100% of offline activity. But 70% beats zero.
Multi-Agent Reality
Agents will resist. They'll say "I brought in that lead, why does Google get credit?" They're not wrong—they did close it. But they also didn't reach the buyer first.
The conversation changes when you separate roles: marketing reaches people (first touch), your content engages them (middle touch), your agent closes them (last touch). Marketing and sales aren't competing. They're stages.
Show agents their own last-touch numbers. That's their close rate. Show them the full journey. That's context for how to improve the pitch.
What Breaks This Model
It fails if you don't have contact records. You can't attribute anything if leads arrive unnamed. It fails if your CRM doesn't capture source data. Garbage in, garbage out.
It also fails if you treat it as final. Markets change. Your channels shift. Your buyer behavior evolves. Audit this model every quarter. If a source drops in first-touch volume, investigate. If middle touches from one channel stop converting, test different messaging.
The point isn't to build a perfect machine. It's to build one that tells the truth often enough that you can make better decisions.