Online leads go cold in their first ten minutes because of how you route and notify them, not because the leads were bad or your agents are lazy. Fix the routing and the exact same leads, from the exact same sources, close at a different rate.
I’ve spent 22 years running the systems side of a Reno brokerage, and this is the complaint I hear most often from owners: “the leads aren’t converting.” Almost every time, the leads are fine. What’s broken is what happens between the form submission and the first human response — and that gap is usually measured in hours, not minutes, even when nobody involved thinks of themselves as slow.
The five-minute window
The pattern here is well established and shows up again and again in lead-response research across industries: contact a lead within about five minutes and it converts meaningfully better than a lead you get to after thirty minutes, and the gap keeps widening the longer you wait after that. I’m not going to hand you a precise multiplier and pin it to a named study — the specific numbers that circulate online get mangled and misattributed so often that I no longer trust any of them, including the ones I used to repeat myself. What I trust is arithmetic you can check.
So here’s a worked example, built from assumptions I’m stating out loud so you can swap in your own. Say your brokerage buys 25 online leads a month — 300 a year, which is a realistic volume for a small team running paid ads on top of MLS syndication. Say leads contacted within five minutes close at 8%, a reasonable number for cold web leads. Say leads that don’t get a real response until after thirty minutes — which, as I’ll get to, is most of them — close at 3%. That’s a modest gap, well short of the dramatic multiples you’ll see quoted elsewhere, and it still costs real money.
- Fast response, 8% close rate: 300 leads × 0.08 = 24 closed deals a year
- Slow response, 3% close rate: 300 leads × 0.03 = 9 closed deals a year
- Difference: 15 deals a year lost to slow routing, not bad leads
Now put a dollar figure on it. Take a sale price of $400,000 — a round placeholder, not a claim about current local comps, so swap in your own — and a 2.5% side commission: $10,000 in gross commission on that side of the deal. That $10,000 doesn’t land in the brokerage’s account whole, though; most of it goes to the agent who closed it. Say a typical 70/30 split in the agent’s favor, so the brokerage keeps 30%, or $3,000 a deal. Fifteen lost deals a year is $45,000 that the brokerage itself never sees, on leads it already paid to acquire. Change the split, the price, or the close rates to match your own numbers — the arithmetic is what matters, not these specific figures.
Why does the gap exist at all? Because an online lead rarely fills out one form. They fill out three or four, on your site and a couple of competitors’, in the same sitting. Whoever calls back first isn’t just first — they’re often the only one who reaches a person still sitting there, still interested, still comparing. Wait thirty minutes and you’re not competing against the market anymore; you’re competing against whichever agent got there before you, and that’s a race decided entirely by routing, not by skill.
Round-robin versus first-claim
Most brokerage lead routing lands on one of two models, and both have a real failure mode. Neither is correct in every case, and I’d be suspicious of anyone who tells you otherwise.
Round-robin assigns leads in order, one agent at a time, regardless of who’s actually available. It’s fair — everyone gets an equal share of the pipeline — and that’s also its problem. The system doesn’t know or care whether the agent next in line is asleep, at a showing, on a plane, or on vacation. The lead sits in their queue until they happen to look at their phone, and the clock that matters most — the one measured in minutes, not turns — keeps running the whole time.
First-claim fires the lead to everyone at once and hands it to whoever responds first. It solves the availability problem — someone awake and holding their phone gets it fast — but it optimizes for the wrong thing. Fastest to tap a notification is not the same skill as best fit for that particular client, and over time it quietly demoralizes the agents who are good at their jobs but slower to their phones. Lose the race often enough and most people stop racing; they let the notification go and wait for the next round-robin turn instead, which defeats the point of building a fast system in the first place.
A small, hungry team where everyone is roughly equally matched can make first-claim work. A larger team, or one with a wide skill spread, tends to do better with round-robin plus the escalation layer I’ll describe further down — fair distribution, with a backstop for when the assigned agent doesn’t answer. Some teams land on a hybrid: round-robin assignment as the default, with a short window where the assigned agent can pass the lead back to the pool if they’re genuinely unavailable, rather than letting it sit unanswered until the timer forces an escalation. It costs a little more setup work than either pure model, and for a mixed-skill team it’s often worth it.
After hours is most of the day
Here’s a sum most brokerages have never actually run. A team working 9 to 5, Monday through Friday, covers 8 hours a day for 5 days — 40 hours. A week has 168 hours. Forty divided by 168 is about 23.8%.
That means a standard business-hours schedule covers less than a quarter of the week. Roughly 76% of the week — evenings, early mornings, and every weekend — falls outside it. Online buyers don’t browse listings on a 9-to-5 schedule; if anything, evenings and weekends are when people actually have time to look. A routing setup built only for business hours is, by construction, absent for three-quarters of the time your leads are showing up.
This is the part that trips up owners who otherwise run a tight operation. They’ll staff the office, answer the office phone, and still let every web lead default to “we’ll follow up Monday” because nobody built a path for Saturday afternoon. The fix isn’t asking anyone to work more hours — it’s making sure the routing and notification system doesn’t assume business hours in the first place, so a lead that arrives at 8pm on a Friday reaches somebody, or at least gets logged and escalated, instead of sitting untouched until the office reopens.
When responding faster makes things worse
I want to be straight about the limits of all this, because speed for its own sake is not automatically good, and I’ve watched it backfire.
If you point instant push-and-SMS alerts at every single form submission, including the ones that are obviously junk, a wrong number, or a curious neighbor pricing their own house, you don’t get a fast team — you get a burned-out one. Agents who get buzzed for garbage leads at 9pm learn, fast, to stop trusting the notification. Once they stop trusting it, they stop responding to it, including for the leads that were real. At that point the speed you paid for is gone, and it’s gone for everyone, not just for the junk. Qualify before you accelerate — score the lead or ask one quick question before it ever reaches a human’s phone, and save the urgent alert for leads worth an urgent alert.
The other trap is more subtle. A reply that arrives in under a minute, every single time, in the same canned wording, reads as a bot — because it is one — and some leads bristle at that more than they would at a slightly slower, obviously human message that actually addresses what they asked about. Fast and generic can lose to slower and specific. The goal isn’t the fastest possible reply. It’s the fastest reply that still sounds like a person read the message.
A routing setup that holds up
Getting this right takes three things most setups skip, not exotic tooling.
- Notification redundancy. Push alone fails silently — Do Not Disturb, an unopened app, an OS holding the notification back — and you’ll never know it happened. Pair push with SMS. A text gets through when an app notification doesn’t, and the two together cost you almost nothing to run.
- Escalation on a timer. If the assigned or first-in-line agent hasn’t claimed the lead within, say, 5 to 10 minutes, it should automatically escalate — to a backup agent, then to you. Nobody should be able to let a lead die quietly just by not looking at their phone.
- Logging, on every lead. Time from lead creation to first human contact, tracked per agent, is the one number that tells you whether any of this is actually working. A team that can’t measure its response time can’t improve it — it can only guess, and that measurement is only as good as the contact records feeding it, which is its own problem when your CRM and your MLS aren’t actually in sync.
The escalation and logging pieces are the kind of thing I set up under marketing automation — rules that watch the clock and act on it without anyone having to remember to check. If you want the wider picture of how I think about technology for brokerages generally, that’s covered on my real estate technology page.
Frequently asked questions
A few questions I get from brokers once they’ve seen the after-hours math and started looking at their own routing.
What tools actually do lead routing?
Three broad categories: CRM-native routing built into whatever system already holds your leads, dedicated lead-routing tools that sit between your lead sources and your team, and general-purpose automation platforms configured to watch for new leads and act on rules. Which fits depends on what you already run, not on any single product’s current feature list.
What if my agents won’t adopt it?
Start by asking why. Usually it’s alert fatigue from too many low-quality pings, not resistance to speed itself. Qualify leads before alerting, keep the notification volume reasonable, and show agents the response-time numbers — most competitive agents adopt fast once they see who’s actually winning the leads and why.
How do I measure speed to lead?
Log two timestamps for every lead: when it arrived and when an agent first made real contact, not just claimed it. Track the gap in minutes, per agent and team-wide, and review it monthly. Without that log you’re guessing at how fast your team actually responds, which is exactly the problem this whole post is about.
Does this work for referral leads too?
The routing and escalation mechanics apply the same way, but the urgency math is different. A referral already trusts you, so a same-day reply usually doesn’t lose the deal the way it does with a cold web lead. Route referrals fast and courteously, but save your five-minute alerts for the leads with no relationship to fall back on.
If you’re paying for online leads and suspect the source, not the routing, is the problem, it’s worth checking the routing first — it’s cheaper to fix and usually the actual cause. No sales pitch, just a straight conversation: visit my real estate technology page and I’ll look at what’s actually happening to your leads after they come in.