Skip to sources
Time Machine

Real Estate Sales Agents

Scrub through 128years of this role's history, from when it first emerged, through every wave of technology that reshaped it, to the cited projections for where it's heading next.

2026drag to travel through time
1925195019752000now
Country
2026
Known today as Real Estate Sales Agent (BLS SOC 41-9022)
US Employment
193K
OEWS is a point-in-time survey snapshot, not a continuous time series; BLS advises against using it for year-over-year trend comparison.
Median Annual Wage
$52,830
≈ $51,476 in 2024 dollars
Each dot is a cited figure over time; the dotted line only links them (values between aren't measured). Hollow dots are estimates.
Tools of the era

The tools that defined the work

Select an era to see how it reshaped the work.

  • Word of mouth, courthouse records, and newspaper classified listings

    Before the NAR, before the MLS, before state licensing, real estate transactions were brokered through personal knowledge, courthouse deed records, and classified newspaper advertisements. A broker who knew what properties were available and who owned them had a genuine information advantage — one maintained through physical visits to county recorders, relationships with banks and estate administrators, and word-of-mouth networks in neighborhoods. The newspaper classified section (developed by urban papers through the 1880s-1890s) was the first mass-market listing medium: sellers could announce properties to a broader audience, but the broker was still needed to negotiate access and complete documentation. The lack of standardization meant every transaction was a one-off negotiation, with no standard contract forms and no established disclosure practices.

    Effect on the work

    The pre-professional broker's core value was holding exclusive information about available properties. The entire subsequent century of real estate technology — from the MLS to Zillow — was an attack on that information monopoly.

    Work toolChanging equipment
  • NAR Code of Ethics (1913) + first local MLS systems (1910s-1950s)

    The National Association of Real Estate Exchanges was founded May 12, 1908 in Chicago, with 120 founding members across 19 local boards. Its first Code of Ethics was adopted in 1913, establishing the principle that an agent represents a client's interests rather than simply facilitating a transaction for a fee. The "REALTOR®" trademark was created in 1916 by NAR member Charles N. Chadbourn to distinguish code-subscribing members from unaffiliated brokers. State licensing laws began with California in 1917 and spread through the 1920s — by 1930, most states required some form of examination and registration to act as a broker. The Multiple Listing Service concept emerged from informal broker cooperation in the late 1800s ("brokers regularly gathered at the offices of their local associations to share information about properties they were trying to sell," as NAR describes it) and formalized into structured systems through the 1910s-1950s. The first formal MLS, operated by a local board, enabled cooperating brokers to share listings and split commissions — the structural foundation of the modern two-agent transaction (listing agent plus buyer's agent) that would define the profession for a century.

    Effect on the work

    Licensing requirements raised the barrier to entry, reduced fly-by-night operators, and formalized commission structures. The Code of Ethics and REALTOR® brand created a two-tier market: licensed, professional agents versus anyone who could technically sell a property. The MLS made cooperation between agents in the same market the default, which expanded each individual agent's ability to represent clients on both sides of the transaction.

    Work toolChanging equipment
  • FHA / VA loans (1934 / 1944) + first national MLS (1953) + RESPA (1974)

    Three federal interventions reshaped real estate agency between 1934 and 1974. The Federal Housing Administration (1934) standardized mortgage underwriting and made 30-year amortizing mortgages the American default — creating a paperwork and appraisal process that agents were uniquely positioned to navigate on behalf of buyers. The GI Bill (1944) put homeownership in reach for millions of veterans, generating the suburban expansion of the 1950s-1960s and an enormous sustained demand for agents who could show, negotiate, and close residential transactions. NAR launched a nationwide MLS system in 1953 — formalizing the cooperative listing structure that had existed locally since the early 1900s into a national framework. The Real Estate Settlement Procedures Act (RESPA, 1974) prohibited undisclosed kickbacks and required disclosure of settlement costs, establishing that agents could only receive compensation for services actually rendered. RESPA also standardized the Good Faith Estimate, giving buyers a clearer picture of transaction costs. These federal interventions, taken together, transformed real estate agency from a local information brokerage into a regulated, federally-touched professional service.

    Effect on the work

    The FHA/VA mortgage paperwork burden increased the practical value of an agent — few buyers in 1950 understood the underwriting standards, disclosure requirements, and appraisal processes of federally-backed loans. Agents who could navigate this became indispensable to the transaction rather than optional facilitators.

    Work toolChanging equipment
  • Computerized MLS + fax machines + lockbox systems

    The 1980s brought the first major information-technology wave to real estate: computerized MLS databases replaced handwritten and mimeographed listing sheets, fax machines allowed offer documents to be transmitted across cities in minutes rather than days, and electronic lockbox systems (first deployed by the Supra brand in the 1980s) let agents schedule showings without the listing agent physically present. Computerized MLS did not displace agents — it made cooperation between agents faster and more systematic. An agent in 1990 could search listings by bedroom count, price range, and school district in seconds; the same search in 1975 required manually flipping through paper listing books. The productivity gain was real but it expanded what agents could do rather than reducing how many were needed. NAR membership grew steadily through the 1980s-1990s, exceeding 700,000 members by the mid-1990s and approaching 800,000 by the late 1990s. The internet was still emerging as a listing medium; the first online real estate listings (early Realtor.com, launched 1996 as a partnership between NAR and Move Inc.) appeared in the mid-1990s, before the wave that Zillow would eventually bring.

    Effect on the work

    Computerized MLS reduced the time-per-search dramatically but increased agent productivity without reducing agent headcount — the efficiency gains were absorbed by serving more clients and maintaining larger territories rather than by reducing total agents.

    Work toolChanging equipment
  • Internet listings — Realtor.com (1996) + Zillow (February 2006) + Trulia (2005)

    Zillow was incorporated in December 2004 and launched its public website on February 8, 2006 — the site crashed on launch day from overwhelming traffic. Co-founders Rich Barton and Lloyd Frink, both former Microsoft and Expedia executives, built Zillow explicitly to democratize real estate information: the Zestimate tool put automated property valuations on 160 million homes for anyone to see for free. Trulia launched in 2005. For the first time, a buyer could search every listed property in a city, see estimated values, view satellite photos, and read neighborhood statistics — all without talking to an agent. The incumbent theory was that disintermediation was inevitable: if the MLS information monopoly was the agent's core value, free internet search had just destroyed it. The reality was more complicated. The housing bubble of 2004-2006 (NAR membership peaked at a record 1.37 million in 2006) masked the structural change. The crash of 2007-2011 showed that agents who survived were the ones whose value was in negotiation, local knowledge, transaction management, and relationship trust — not in search. Buyers found listings on Zillow; they still called agents to get inside.

    Effect on the work

    Internet listing platforms did not reduce agent headcount during the bubble years — NAR membership was at its all-time high in 2006. What they did was eliminate the information-monopoly justification for agent commissions, setting up the commission-restructuring story that would play out in the 2024 settlement. The bubble masked the structural change; the bust revealed it.

    Work toolChanging equipment
  • iBuyer platforms — Opendoor (2014) + Zillow Offers (2018-2021) + Redfin Now

    If internet search could not disintermediate the agent, perhaps algorithmic buying could. Opendoor launched in March 2014 with a direct offer model: homeowners could request an instant cash offer, skip the showings and negotiations, and close in days. The company bought 37,000 homes in 2021 at the peak of its scale. Zillow launched Zillow Offers in 2018, extending its own brand to the iBuyer model. Redfin, which had launched in 2004 with flat-fee commission refunds and a salaried-agent model, launched Redfin Now as its iBuyer product. The theory: if an algorithm could price homes accurately enough to buy and resell at a profit — or even at break-even while generating ancillary mortgage and title revenue — the listing and buyer agent became irrelevant to the transaction. The theory failed at scale. In November 2021, Zillow CEO Rich Barton announced the shutdown of Zillow Offers, citing $420 million in losses in Q3 2021 alone (Zillow owned approximately 7,000 homes at shutdown). The algorithm had consistently over-paid for homes in a rising market and could not adjust fast enough when prices peaked. Opendoor lost $662 million in 2021 and $1.4 billion in 2022. Redfin shuttered Redfin Now and laid off 13% of staff in November 2022. The iBuyer experiment demonstrated that algorithmic home valuation at scale, under volatile market conditions, consistently mispriced assets in ways that individual local agents — with their granular neighborhood knowledge — did not.

    Effect on the work

    iBuying, at its 2021 peak, handled approximately 1% of US home transactions. Even at that volume it generated hundreds of millions in losses. The agents who worked iBuyer-adjacent transactions (helping buyers purchase Opendoor homes, helping sellers understand offers) retained employment throughout the experiment. The iBuyer moment showed that the agent's value is in local judgment under uncertainty, not in the paper transaction.

    Work toolChanging equipment
  • NAR v. Sitzer/Burnett settlement (2024) + AI property tools + rate-shock market

    On October 31, 2023, a federal jury found NAR and co-defendants (Keller Williams, HomeServices of America, Anywhere Real Estate, Re/Max) liable for conspiring to inflate real estate commissions, awarding nearly $1.8 billion in damages in the Sitzer/Burnett case. In March 2024, NAR settled for $418 million, agreeing to eliminate its MLS commission rules and waive the right to appeal. The two critical practice changes, effective August 2024: (1) buyer agent compensation can no longer be advertised on MLS listings — the standard 2.5-3% buyer-side commission that sellers had effectively always paid is no longer a default MLS field; (2) buyers must sign a written compensation agreement with their agent before viewing properties together. These changes structurally restructure how buyer agents are compensated: for the first time in the MLS era, buyers and buyer agents must explicitly negotiate and agree on compensation rather than assuming it will flow automatically from the seller. Early post-settlement data suggested NAR membership declining, some buyer agents exiting the market, and commission rates compressing on both sides. Simultaneously, AI property tools — Zillow's natural-language search, AI-powered comparative market analysis, automated disclosure review — have automated the research and documentation tasks that previously justified significant agent time, raising the question of what an agent's hour is worth when software can produce a CMA in seconds.

    Effect on the work

    BLS projects +3.1% employment growth for real estate sales agents 2024-2034 (National Employment Matrix). The settlement's net employment effect is uncertain: it may reduce the number of buyer agents (whose compensation model changed most dramatically) while concentrating the remaining agents on higher-value transactions requiring more relationship management. The 53.7% self-employment rate means most agents absorb market changes directly rather than through employer layoffs.

    Work toolChanging equipment
Projection cone · present → 2034

What credible sources project

Scrub the slider past now to anchor each scenario on the scrubber. The spread is the range of futures credible sources project for this role.

Employment outlook
Projected change in the number of people doing this work.
Housing demand recovery scenario (optimistic)
2034
+10%
The optimistic scenario rests on three factors: (1) Pent-up millennial and Gen Z household formation — millennials are the largest generation in US history, currently in prime homebuying years (25-40), many locked out of the market by 2022-2024 rate increases; a return of rates toward 5-6% would release significant pent-up transaction volume. (2) Housing supply shortfall — Freddie Mac estimates a 3.8 million unit housing shortfall as of 2021, a supply constraint that supports sustained price levels and transaction activity. (3) Agent concentration among higher-productivity survivors — as marginal agents exit post-settlement, remaining agents handle more transactions each, maintaining industry aggregate commission income with a leaner workforce. This scenario implies 10% employment growth from 2024 by 2034 as transaction volumes recover.
BLS National Employment Matrix 2024-34
2034
+3%
BLS Employment Projections 2024-34 cycle (most current). Baseline: 420,900 (2024); projected: 433,700 (2034); net change: +12,800 (+3.1%). Wage and salary employment is projected to grow faster (+4.8%) than the overall occupation (+3.1%), reflecting the gradual shift from pure commission self-employment toward salaried agent roles at tech-forward brokerages like Redfin. Annual openings: 36,600 (new jobs + replacement need). BLS classifies growth as "average." The projection does not explicitly model the NAR settlement impact, which was finalized after the projection cycle baseline was set.
NAR settlement disruption scenario (pessimistic)
2030
-20%
The March 2024 NAR settlement ($418M) requires buyers to explicitly negotiate agent compensation rather than relying on the traditional seller-pays-buyer-agent model. If this restructuring causes a material fraction of buyers to opt for reduced-service or unrepresented transactions, buyer agent employment could contract significantly. The pessimistic scenario assumes: (a) 20-30% of first-time and price-sensitive buyers reduce or eliminate buyer agent representation; (b) average buyer agent income falls 15-25% due to commission compression; (c) agents operating at the margin of profitability exit. This would imply a 15-25% contraction in the buyer-agent segment specifically, partially offset by continued seller-side representation (listing agents face less pressure from the settlement). The -20% scenario represents the tail risk if the settlement significantly restructures consumer behavior.
AI task exposure
Share of the role’s tasks that researchers estimate AI can do. This is a measure of task exposure, not a forecast of jobs lost.
Frey & Osborne (2013)
2033
50%
of tasks
Frey & Osborne's Gaussian-process classifier assigned Real Estate Sales Agents a probability of computerization of approximately 0.86 — placing them in the top decile of automation risk in the 702-occupation dataset. The bottleneck factors they measured (persuasion, social perceptiveness, negotiation) scored medium; the tasks that scored highest for automation susceptibility were information-retrieval (finding listings, gathering property data) and documentation. The F&O thesis was structurally wrong about the mechanism: search, not the agent, was automated. A buyer can find a listing on Zillow without an agent. But the F&O prediction that automation would attack the job through its information-brokerage function was directionally correct — the information monopoly dissolved. What F&O underweighted was the transaction coordination and relationship management functions that survived. Employment has not fallen 50%; it fell approximately 20% from the 2006 bubble peak to 2024, a decline driven by the housing bust rather than automation.
Eloundou et al. — "GPTs are GPTs" (2023)
2030
10%
of tasks
Eloundou et al.'s GPT-4 task-exposure labeling on O*NET tasks rates real estate sales agents at high LLM exposure for documentation tasks (drafting disclosure statements, writing property descriptions, preparing contract summaries) but low exposure for the core negotiation, showing, and relationship functions. The -10% estimate represents the curator's interpolation of Eloundou's framework applied to the documentation-heavy administrative tasks within the occupation: AI tools that draft offer letters, generate comparative market analyses, and synthesize property disclosures reduce agent administrative hours without necessarily reducing the number of agents needed for relationship-dependent functions. The net effect is a productivity gain (more transactions per agent) rather than headcount reduction, but some agents at the margin — those who competed primarily on documentation throughput rather than relationship quality — may exit.
Today, in this role

What's shifting in the work right now

The historical view above shows how this role has moved. This is the present-day detail: which AI tools are picking up which tasks, where the edge still is, and the natural directions this work can grow.

What's changing in your day

Three parts of your work where AI is already doing real lifting, and what stays yours.

AI is taking this onSchedule and manage showings, confirm appointments, and collect post-showing buyer feedback: use Zillow ShowingTime+ AI to auto-schedule showing requests, send automated confirmation and reminder texts to buyers and sellers, and gather structured feedback from buyer agents — all without agent involvement until a feedback review or offer conversation is needed.

Schedule and manage showings, confirm appointments, and collect post-showing buyer feedback: use Zillow ShowingTime+ AI to auto-schedule showing requests, send automated confirmation and reminder texts to buyers and sellers, and gather structured feedback from buyer agents — all without agent involvement until a feedback review or offer conversation is needed.[8],[7]

Tools picking this up
Where your edge is

Showing logistics is one of the most fully automated tasks in the listing agent workflow. ShowingTime+ handles 90%+ of the scheduling, confirmation, and feedback-collection cycle. Use the time you recover to focus on what the AI cannot do: call the buyer agent directly after a showing to understand what their client thought, build the relationship, and surface objections early. That real-time intel from a personal call still beats automated feedback surveys when you are deciding whether to reduce price.

AI is sitting alongside you hereWrite listing descriptions and prepare marketing packages: direct ListingAI or ChatGPT to generate SEO-optimized MLS copy from property spec inputs, review and edit for accuracy and local voice, then coordinate with Restb.ai-tagged property photos and Virtual Staging AI-rendered room visuals — reducing listing prep from a half-day to under an hour.

Write listing descriptions and prepare marketing packages: direct ListingAI or ChatGPT to generate SEO-optimized MLS copy from property spec inputs, review and edit for accuracy and local voice, then coordinate with Restb.ai-tagged property photos and Virtual Staging AI-rendered room visuals — reducing listing prep from a half-day to under an hour.[9],[7],[10]

Where your edge is

AI-generated listing copy is now table stakes — every listing agent in your market has access to the same tools. Your edge is editing for local voice and micro-neighborhood storytelling that the AI cannot source: the specific school boundaries, the walking-distance coffee shop, the HOA board dynamics. Write the first sentence yourself. Let AI draft paragraphs 2–4. Review for factual accuracy and local specificity before publishing. Pair with professional photography; virtual staging is a supplement, not a replacement for quality images in competitive markets.

AI is taking this onManage lead routing, follow-up, and pipeline nurturing: configure BoldTrail or Follow Up Boss AI to auto-assign inbound leads by geography and buyer stage, set automated SMS and email follow-up sequences triggered by listing views and form fills, and review AI-generated lead-score dashboards to prioritize which contacts warrant a personal call.

Manage lead routing, follow-up, and pipeline nurturing: configure BoldTrail or Follow Up Boss AI to auto-assign inbound leads by geography and buyer stage, set automated SMS and email follow-up sequences triggered by listing views and form fills, and review AI-generated lead-score dashboards to prioritize which contacts warrant a personal call.[6],[11],[12]

Where your edge is

AI CRM follow-up has made the first 5–7 automated touches essentially free. Your differentiation now starts at touch 8 — the personal call or handwritten note that signals you are a human, not a drip sequence. Agents who close 28% more transactions with AI CRMs (RISMedia, Jan 2026) are not sending fewer messages; they are spending the time saved on higher-value activities: listing consultations, showing feedback calls, and referral-partner lunches. Configure your AI sequences and then get off the dashboard.

Where this role is heading

Natural next steps for someone with your foundation: not exits, evolutions.

A direction you could grow

Property, Real Estate, and Community Association Managers

Property managers share most of the foundational skills with real estate agents — landlord-tenant law, lease documentation, vendor coordination, property marketing — but generate recurring management-fee income rather than commission-per-transaction income. The income model is meaningfully more stable: a portfolio of 50 managed properties produces predictable monthly revenue regardless of whether any individual home sells. BLS projects 5% growth for Property Managers through 2034. AI automates maintenance-request routing and rent-payment processing, but the tenant relationship, owner reporting, and physical property inspections remain human. For agents who want more predictable income while staying in real estate, property management is the clearest lateral pivot with a low licensure barrier (most states allow sales agents to manage property under an active license).

What you'd add
  • · Property management software: AppFolio, Buildium, Yardi Breeze — maintenance ticketing, rent ledgers, owner portals
  • · Landlord-tenant law for the applicable state(s): eviction procedures, habitability standards, security deposit rules
  • · Lease drafting and lease-renewal negotiation for residential and small commercial properties
  • · Vendor management: soliciting bids, approving repairs, managing contractor relationships
  • · HOA governance: CC&Rs, board meeting facilitation, reserve study interpretation (for community association management)
What it takesMost of your skills carry over
Share this year
Drops anyone you send it to straight into 2026.
Preview card
Part of Sales · see all 18roles →
Different role?

See the same long-arc view for your own profession.

Browse the directory by industry, or search by title or SOC code. New roles ship every few weeks. Every profile cites every claim.

Browse all roles

The data behind this timeline

On record since1908
Latest tracked employment193,370 (US, 2025)
Latest median pay$52,830 (2025)
Outlook+3% by 2034 (BLS National Employment Matrix 2024-34)
View all 27 cited data points
YearUS employmentMedian annual paySource
192075,000n/aESTIMATE
1950140,000n/aESTIMATE
1975290,000n/aESTIMATE
2000408,000$30,000ESTIMATE
2003123,490$32,610BLS-OEWS
2004126,470$35,670BLS-OEWS
2005150,200$39,240BLS-OEWS
2006529,000$39,760ESTIMATE, BLS-OEWS
2007172,030$40,600BLS-OEWS
2008164,080$40,150BLS-OEWS
2009151,550$40,100BLS-OEWS
2010153,740$40,030BLS-OEWS
2011158,340$39,070BLS-OEWS
2012380,000$39,140ESTIMATE, BLS-OEWS
2013158,850$39,800BLS-OEWS
2014157,660$40,990BLS-OEWS
2015151,700$43,370BLS-OEWS
2016151,840$44,090BLS-OEWS
2017147,960$45,990BLS-OEWS
2018156,760$48,690BLS-OEWS
2019470,000$48,930BLS-OEWS
2020168,740$49,040BLS-OEWS
2021175,920$48,340BLS-OEWS
2022193,010$49,980BLS-OEWS
2023445,700$54,300BLS-OEWS
2024420,900$56,320BLS-OEWS
2025193,370$52,830BLS-OEWS
Embed this timeline on your site

Free for any site. Paste this where the timeline should appear; it stays interactive, every datapoint stays cited, and it sets no cookies on your page. How embedding works

<iframe src="https://futurehistory.earth/embed/41-9022"
  width="100%" height="430" style="border:0"
  title="Real Estate Sales Agents, a Future History timeline"
  loading="lazy"></iframe>

See all roles in Sales