A Zillow or Realtor.com lead is a timestamp, not a relationship. The buyer often sent the same form to two other teams. Whoever starts a real conversation first usually books the viewing.
Speed to lead is the time from inquiry created to first live attempt: a ring, not a CRM notification. For paid US portal leads, the useful target is under 60 seconds. Five minutes still competes. Thirty minutes is the losing bucket in the study everyone quotes. Next morning is farming for the desks that called last night.
That is the whole claim. The rest of this page is the evidence, the brokerage-specific reasons it is worse than a normal web form, and the three numbers to pull from last month’s queue before you change tools.
When the first call happens
US portal leads. Contact and qualify odds fall as you wait.
What is speed to lead in real estate?
In a brokerage, speed to lead is not “we try to call within the hour.” It is the clock that starts when the portal or site writes a new contact into Follow Up Boss, HubSpot, or whatever the desk actually lives in, and stops at the first outbound dial or inbound pickup.
It is not:
- time to an email auto-reply
- time to a text that says “thanks, an agent will be in touch”
- time an ISA saw the lead
- time to a closed sale
Those are different jobs. Mixing them is how teams quote a five-minute SLA and still have a 40-minute median first call.
Portal leads make the clock harsher than a brand-site form. The same person can request a tour on three listing pages in one sitting. After-hours and Sunday traffic is a large share of that volume. If phones stop at 6pm, you are donating the lead to whoever is still dialing.
What do the studies actually say?
Two papers get mashed into one LinkedIn graphic. They are not the same study, they do not measure closed sales, and they do not say “call in five minutes or the lead is dead.” They still matter. Read them as written.
The 2007 Lead Response Management study
James Oldroyd, then a faculty fellow at MIT Sloan, ran the Lead Response Management work with InsideSales.com. The dataset was more than 15,000 web leads and about 100,000 call attempts.
The comparison that survived into every sales deck: calling at 5 minutes versus 30 minutes was associated with about 100× higher odds of making contact and 21× higher odds of qualifying the lead.
Hold the caveats:
- The losing side is 30 minutes, not an hour.
- The outcomes are contact and qualify, not a signed listing or a funded deal.
- These were web leads across companies, not a Zillow Premier Agent panel.
For a team buying shared portal leads, that is still the useful frame. You are not trying to be “responsive.” You are trying to be the first human voice while the buyer is still on the listing.
The 2011 Harvard Business Review audit
The HBR piece is later and different. The Short Life of Online Sales Leads (Oldroyd, McElheran, Elkington, March 2011) audited 2,241 US companies.
What they found on response behavior:
- Average first response: 42 hours
- About 23% never replied at all
In a separate dataset of 1.25 million leads, firms that responded within an hour were nearly 7× more likely to qualify the lead than those who waited one more hour, and more than 60× more likely than those who waited a day.
That paper is why “we get back within a business day” is not a process. It is a leak with a polite name.
What the studies do not say
They do not give you a Zillow conversion rate. They do not prove that a voice agent closes more sides than a licensed ISA. They do not say email is worthless. They say the first live attempt is a decaying asset, and the decay is measured in minutes, then hours, then “never.”
If someone on your team quotes “100×” as a close-rate, correct them. Contact odds and close rates are different columns.
Why portal leads die faster than a website form
A brand-site inquiry is often exclusive. A portal inquiry is paid, shared, and stamped.
- The same buyer hits three listing pages before the first callback.
- Nights and weekends carry a large share of portal volume. An ISA floor that runs 9–6 is closed for the hours the form is busiest.
- Your CRM can create the contact in two seconds and still sit uncalled for an hour if a person has to notice a toast.
- “New lead” text messages to a group chat are not a queue. They are a hope.
Speed to lead is whether the first ring happens before the next agent, not whether your software logged the person.
That is also why after-hours answering services built for dentists do not solve this. The job is not to take a message. The job is to start the same qualification call you would run at 2pm on a Tuesday.
How fast is fast enough?
| First touch | What it usually means |
|---|---|
| Under 60 seconds | Automated dial the moment the lead hits the CRM or form |
| Under 5 minutes | An ISA is watching the queue and can still win most shared leads |
| 30 minutes | You are in the 2007 study’s losing bucket |
| Next morning | You are farming for the teams that called last night |
If you buy Zillow, Realtor.com, or Homes.com leads, “we try to call within the hour” is not a process. It is a leak.
Sixty seconds is not a personality trait. Humans do not sit on a form at 11:40pm. Teams that hit that number connect the portal or website to something that can dial without waiting for a shift.
Five minutes is still a serious target if a licensed ISA is on the desk and the routing is clean. It is not a serious target if the lead sits in a shared inbox until someone finishes a showing.
How do teams actually get under 60 seconds?
The working path is boring:
- The portal or site creates the lead in the CRM (or posts a webhook).
- A voice agent dials immediately.
- The call qualifies budget, timeline, and intent.
- It offers two viewing slots and books one.
- Notes, tags, and the next step write back to the CRM. On Follow Up Boss that write-back is the difference between a toy and a desk tool.
If the buyer sends the call to voicemail, the thread should continue on WhatsApp, SMS, or email instead of dying in a “left VM” disposition.
You do not need a 20-seat ISA floor to do this. You need the lead to produce a call, not a notification.
mergro is built for that path: inbound and outbound voice, then WhatsApp, email, and SMS, with write-back to the CRMs the desk already uses. It is usage-based, not per-seat. The point of this page is the clock, not a feature list. If you want the clock on your own lead flow, get started.
What should you measure this week?
Do this afternoon. Export last month’s portal leads. Do not open a dashboard.
- Median time to first call (lead created → first dial). Mean will lie if a few next-day callbacks sit in the tail.
- Contact rate in the first 5 minutes versus after 30.
- Viewings booked from leads called under 5 minutes versus the rest.
Optional fourth, only if your CRM is clean: shared-lead loss — inquiries that also appeared on another agent’s sheet, or that booked with someone else after you waited.
If median first call is over 15 minutes, this article is not your problem. The queue is. Fix routing before you buy more Zillow spend.
A useful weekly ritual: pick 20 leads at random. For each one, write the created timestamp, the first-dial timestamp, and whether a human ever spoke to them. You will see the leak in an afternoon. You do not need a BI tool.
Common mistakes
Calling the SLA the process. “We respond in five minutes” on a hiring page is not a median. Pull the timestamp.
Counting email as first touch. An auto-reply does not compete with a ring from the other team.
Stopping the clock at 6pm. Portal volume does not.
Leaving the lead in a group chat. If three people “own” the new lead, nobody owns the first 60 seconds.
Optimizing talk tracks before the queue. A better script on a 40-minute delay still loses to an average script at 45 seconds.
Treating the 100× figure as a close-rate. It is a contact-odds comparison between 5 and 30 minutes.
Frequently asked questions
What is a good speed to lead for real estate?
Under 60 seconds for paid portal leads. Under 5 minutes if a person is watching the queue. Anything built around “within the hour” or “next business day” is the HBR leak.
Is speed to lead the same as lead response time?
People use them interchangeably. Be precise: speed to lead should mean first live attempt. Lead response time in reports often includes email and SMS, which inflates how fast you look.
Does calling in five minutes guarantee a viewing?
No. It raises the odds you reach them and can qualify them. The viewing still depends on budget, timeline, inventory, and whether you offer a time on that call.
Do the Harvard and MIT numbers apply to Zillow?
They are not a Zillow study. They are web-lead studies. Portal leads are usually worse than a brand form because they are shared and after-hours heavy. Use the studies for the decay curve, not as your conversion rate.
Can an ISA hit under 60 seconds without voice AI?
In theory, if someone is staring at the queue 24/7. In practice, nights, weekends, lunch, and showings break that. Automation is how teams keep the first ring when the floor is empty.
Sources
- James Oldroyd with InsideSales.com, Lead Response Management study (2007). Contact and qualify odds at 5 minutes versus 30 minutes. Not a closed-sale stat.
- James Oldroyd, Kristina McElheran, and David Elkington, The Short Life of Online Sales Leads, Harvard Business Review, March 2011.
Want this clock on your own lead flow? Get started.