There is no official Zillow lead conversion rate. Not for Premier Agent. Not for Preferred. Forums will sell you a percentage. Vendors will mash 2007 contact odds into a close-rate. Zillow does not publish a national “X percent of connections close.”
Compute three columns on your file. If you cannot, you are arguing about invoices with feelings.
This page is the arithmetic, the splits that keep you honest, and the numbers you are not allowed to borrow.
Why a national rate would be a lie anyway
Premier Agent connections are shared. Preferred is a different contract - a share of GCI, not the same object as a ZIP ad click. What Zillow costs is two invoices. Mixing them into one “conversion rate” is how teams decide the product is broken when the queue is broken.
Convert is the live transfer and the first five minutes. Worth it is whether that first conversation happens at all. A close-rate on a file that was never called is a cost-per-ignored-name.
NAR’s 2025 Profile tells you 88 percent of recent buyers used an agent and median search was 10 weeks. It does not tell you what share of your Zillow connections became clients. That is your CRM.
The only three columns that matter
Pull one month of Zillow-sourced contacts. Split Preferred, Premier Agent / connection, and any other Zillow object you buy. Then:
- Contacted - a live talk in the first 5 minutes (or your published SLA). Not a text. Not “an agent will call you.” Voicemail is an attempt, not contacted.
- Held - appointments that showed. Not set. Not “penciled Saturday.”
- Cash / hold - closed GCI (or contribution you actually banked) divided by held appointments in that cohort. If you prefer cash / contacted, say so and do not switch mid-meeting.
Optional fourth: median minutes to first live attempt. That is the diagnostic. It is not a conversion rate.
Do not divide closed deals by “leads received” if half were never called. You will get a horror percentage and buy a coaching course.
Oldroyd 2007 (100× contact, 21× qualify at 5 vs 30 minutes) is a reason to fix the first column. It is not column three. HBR 2011: 23 percent never replied; 42-hour average among those who did within 30 days. If your median first call is Monday on a Sunday form, you are measuring that, not “Zillow conversion.”
Splits that keep the owner from lying
| Split | Why |
|---|---|
| Nurture / saved search vs connection vs direct contact | Different products. Different share of voice. |
| Weekday vs weekend creates | Weekend coverage is a different shift. |
| Reached in 5 minutes vs later | Otherwise a slow team and a fast team share a fake rate. |
| Preferred vs Premier Agent | Different invoice. Do not average them. |
Qualify belongs in the note on every contacted row: budget, timeline, intent, two times. A contacted lead with no hold is a finished first job and an unfinished second one. Count it that way.
How to run the worksheet this month
- Export Zillow-sourced contacts for 30 days.
- Tag the product (Preferred / connection / other).
- Mark contacted / not with the 5-minute rule.
- Mark held / set / none.
- Attach closed cash when it closes - even if that is next quarter. A trailing close is fine. Inventing a same-month close-rate on 10-week searches is not.
- Put cash / hold next to the invoice.
If Follow Up Boss cannot filter Zillow source plus first-call time, you do not have a conversion problem yet. You have a reporting problem.
Ten-week median search means cash from this month’s connections often lands later. That is fine. Keep the cohort tagged. When a deal closes, attach it to the month the contact was created - or publish a trailing-90-day cash / hold and stop arguing about calendar-month close-rate on a product that does not close in a calendar month. What you cannot do is invent a same-week “Zillow conversion” from three under-contract files and a Facebook anecdote.
mergro writes the first conversation back so those columns are not folklore. Get started if the only number in the meeting is the Zillow bill.
What should you measure this week?
Do this afternoon. One month. Three columns.
- Contacted in 5 minutes / all Zillow creates.
- Held / contacted.
- Invoice for that month / held - a cost per showed appointment, not a fantasy close-rate.
If column 1 is ugly, do not discuss column 3. Fix the ring. Worth it is that conversation.
Common mistakes
Quoting a Facebook group “we close 8 percent.” You do not have their ISA hours, their ZIP mix, or their definition of a lead.
Putting 100× on a slide titled conversion. Contact odds. Different unit.
Averaging Preferred and Premier Agent. You will kill the wrong product.
Counting set appointments as conversion. No-shows are how teams stay optimistic.
Declaring Zillow dead after one slow month with no first-call report. You measured the floor, not the portal.
Frequently asked questions
What conversion rate should I expect?
The honest answer is: whatever your contacted and held columns say after a month of complete notes. Anyone who gives you a national percentage is guessing or selling.
Should I include seller leads?
Seller Preferred is a different contract (Zillow has published seller Preferred around a 40 percent GCI share - read the current Preferred page). Do not drop seller and buyer into one rate.
Can I use Zillow’s Premier Agent dashboard as the rate?
Use it for spend and connection counts. Your CRM owns contacted, held, and cash. Dashboards that stop at “leads” will flatter the invoice.
Is a higher rate always “Zillow is working”?
Not if you stopped buying bad ZIPs or stopped counting uncalled names. Show the three columns before and after. Then argue.
Sources
- Zillow Help Center, Premier Agent FAQ - routing and product, not a close-rate.
- National Association of REALTORS®, 2025 Profile of Home Buyers and Sellers.
- James Oldroyd, Kristina McElheran, and David Elkington, The Short Life of Online Sales Leads, Harvard Business Review, March 2011.
Want the first conversation on the record so the rate is real? Get started.