The hardest number in this business sits in NAR's own member research, and new agents almost never see it before they start.
The gap, stated plainly
Per NAR's 2025 member reporting, median gross income for members with two years or less of experience was $8,100. For members with sixteen or more years, it was $78,900. Across all members, the median was $58,100, and the typical member had 12 years of experience.
Those are gross figures before expenses, and medians rather than predictions for any individual. But the shape is unambiguous, and it is worth understanding correctly, because the usual explanations are wrong.
The gap is not mainly about skill. Contracts are learnable in months. It is that this business runs on people knowing you, and in year one nobody does.
What year one is actually for
The rest of the data explains the mechanism. Per NAR's 2025 buyer research reported by Virginia REALTORS, 43% of buyers found their agent through a referral and 18% used an agent they had worked with before. A new agent has neither: no past clients, and almost nobody who has ever referred them.
So year one is not primarily about learning the job, though that happens. It is about manufacturing the smallest working version of the two things experienced agents already have. A group of people who know what you do, and something findable that confirms you are real.
Everything below follows from that, and so does everything worth skipping.
Tell the people who already know you, individually
This is where most first deals come from, and it is free, and it is the step new agents most often replace with something less uncomfortable.
Not an announcement post. Individual messages to everyone who would recognise your name, including the large group who have no idea you got licensed: former colleagues, university friends, people from unrelated parts of your life, your parents' friends. Most of them do not know, and telling them is information rather than a pitch.
Say what you have started doing, where you work, and that you would appreciate being kept in mind. Then keep a list and contact them again a few months later about something that is not a request. That is the whole system, and it is the same one described in building a referral system, started from zero.
Become verifiable before you become visible
The second job is making sure that anyone who hears your name and looks you up finds something. Right now, most new agents fail this quietly.
Two things cover it. Claim and complete a Google Business Profile under Google's practitioner rules, which permit an individual agent their own profile separate from the brokerage. And put one page online, under your own name, that plainly states who you are, your brokerage, the areas you work, and how to reach you.
That is a low bar and it is above where a surprising number of licensed agents sit. It matters because every other effort in year one ends with somebody checking, and an empty result reads as inexperience more loudly than a small number of transactions does.
Substitute knowledge for track record
The obvious objection: why would anyone hire an agent with no closings?
Because clients are not really asking for closings. They are asking whether you know their situation. Track record is the usual proxy, and it is not the only available evidence.
Pick one small area, smaller than feels sensible, and learn it properly. Attend the open houses. Read the strata documents. Learn which buildings have levy histories, which blocks take water, what sells in a week and what sits. Within a few months you can know one place better than most agents who have worked the whole city for a decade, because they were never looking that closely at that particular place.
Then write it down, as described in blog ideas that earn AI citations. Demonstrated knowledge is credibility you can build in months, while a track record takes years. It is the one shortcut in this post, and it is a legitimate one.
What to skip in year one
New agents spend their first year on things that feel like progress and produce no clients. The main offenders.
A designed logo and a full brand identity. Nobody hires an agent for a logo, and the version you would choose now is not the one you will want in three years.
Paid leads, in most cases. These are the hardest contacts to convert, and your conversion skill is at its lowest right now. Paying premium prices to practise on low-intent leads is an expensive way to learn, and the economics are covered in buying leads vs owning your own source.
Most software. A spreadsheet answers the only question that matters early, which is who have I not spoken to recently. Add tools when a specific bottleneck demands one.
Judging whether it is working before the income arrives
The cruelty of the income curve is that it lags the relationship curve by a year or more, so judging year one by earnings will always say quit.
Watch earlier signals instead. How many individual conversations you had this month. Whether anyone contacted you first. Whether someone described you to a third party, which is the first genuine referral even if it produces nothing. Whether people in your chosen area recognise your name before you introduce yourself.
Those rising while income stays flat is the normal shape of a business that is working and has not paid yet. Those flat while you stay busy is the signal to change what you are doing.
The brokerage decision, judged by year one's actual needs
New agents usually choose a brokerage on commission split, which is close to irrelevant in a year where the split is applied to very little.
Judge it on what year one actually requires instead. Whether an experienced agent will genuinely answer your questions, repeatedly, including the ones you should already know. Whether you can attend other people's showings, listing presentations, and inspections, since watching the job done is how the learning curve actually shortens. Whether the fees are survivable at low volume, because desk fees and monthly charges hurt far more than a percentage when you close two deals.
A better split at a brokerage where nobody has time for you is a worse deal than it looks, and the difference shows up as a slower first year rather than as a smaller cheque, which makes it hard to notice.
One caution about lead-supplying brokerages. Leads that come with a higher split can be a reasonable trade early on, when your alternative is no leads at all, provided you understand that they are not portable. When you leave, they stay, along with any pipeline built from them. That is the same rented-versus-owned distinction that runs through everything else here, and it is worth entering deliberately rather than by accident.
The takeaway
NAR's numbers say the first two years are hard for almost everyone, and the reason is structural rather than personal: the business runs on being known, and you are not yet. So spend year one manufacturing the two assets that produce the referral majority, a circle of people who know what you do and a findable record that confirms you exist, and add depth about one small place as a substitute for the track record you cannot yet have. Skip the logo, skip the software, and be suspicious of anything that feels like progress but involves no conversations.
