NAR Data · November 2, 2026 · Wesley Harper
NAR released its 2026 Member Profile on June 25, 2026, covering 2025 production. One table in it explains more about this business than anything else published this year.
28%Share of a typical member's business from repeat clients in 2025, up from 20% the year before
22%Share from referrals by past clients
0%Both of those figures, for agents with two years or less in the business
Put the first two together and the typical REALTOR got about half of their business in 2025 from people they had already worked with or who were sent by someone they had already worked with. For members with sixteen or more years in, it was about 81 percent. For agents in their first two years, the median was zero on both lines.
The veteran is not outworking the new agent by four to one. The veteran is being paid by work they did years ago.
That zero is the part worth sitting with. It is not a failure, and it is not unusual. It is arithmetic. A database cannot refer you business before it exists. The first two years are the years you fund with activity, and the only question that matters in that window is whether the activity is building something that will still be there in year five.
What the rest of the report says about the person this is happening to
The same report puts the typical member at nine transaction sides in 2025, with a median of thirteen years in the business, out of 1,438,569 members. The year before, the medians were 20 percent repeat and 21 percent referral across all members, and 41 percent and 28 percent for the sixteen-plus group.
So the repeat-client share moved up meaningfully in a single year while total membership stayed enormous. Read that as a market where the people who already had relationships kept transacting and the people who did not had a harder year. That is what a tight market does. It rewards the agent who does not have to find a stranger first.
The practical read
Three things follow from this, and none of them are motivational.
- Your past-client list is a production asset, and most agents cannot describe it. If you cannot say how many people in your database have actually closed with you, you cannot tell whether you are building the 81 percent or not.
- A first-year or second-year agent should not be measured on referral percentage. The median is zero. Measuring it only teaches a new agent to feel behind. Measure the inputs instead: conversations, added contacts, and whether anyone is being contacted twice.
- The difference between 22 percent and 32 percent is a decade of staying in touch, or a system that does it deliberately. The second one is faster.
I have spent thirty years and more than 1,800 closings on the referral side of this business, so I am not neutral about it. But the number I would point at is not mine. It is the zero. Everybody starts there, and the only variable is how long you stay.
Industry data, shared for education. Nothing here is a prediction about your market or your production.
Sources
- National Association of REALTORS, 2026 Member Profile, released June 25, 2026 (2025 production data) · nar.realtor newsroom
- NAR Economists' Outlook, The Feature Sheet: 2026 Member Profile · nar.realtor
- HousingWire coverage of the 2026 Member Profile · housingwire.com