U.S. Real Estate Sales Agent Roles: Evidence from 145 Current Vacancies (October 2026)
MTF Institute Research Report MTF-CF-RR-2026-10-02-REA-01
Research geography: United States. Vacancy observation date: 2 October 2026.
Author: MTF Institute Research Team.
Abstract
This report examines what 145 distinct, publicly current U.S. real estate sales-agent requisitions stated on 2 October 2026. The sample spans 11 employers and three identifiable practice segments: 60 residential consumer, 44 investor-focused and 38 commercial brokerage openings, plus three mixed or unclear roles. Each posting was reopened through an approved public web or browser route and coded for duties, outputs, skills, tools, qualifications, level, cadence, interfaces, authority and escalation where its text supported the field. The resulting Role Requirements Matrix is a structured, reproducible description of this purposive sample, not a representative estimate of U.S. hiring or skill prevalence.
The strongest cross-segment evidence points to a workflow connecting client or investor acquisition, property and market information, listings or opportunities, and transaction progression. The details differ by segment. Residential consumer postings more often name supplied appointments, seller conversations, follow-up and brokerage support; investor postings emphasize building investor relationships and matching opportunities; commercial postings emphasize prospecting, market work and transaction activity. These are observations about the collected advertisements, not a claim that every U.S. agent performs each task. Repeated employer templates, unstated fields and the legal boundary of state licensing materially limit interpretation.
Research question and occupational scope
What responsibilities, work products, skills, tools and authority boundaries do current U.S. employers state when recruiting real estate sales agents? The study covers sales agents in residential consumer, investor-focused and commercial brokerage work. It excludes property managers, mortgage and title specialists, appraisers, broker-manager positions and unrelated occupations. A U.S. sales agent works within a licensed occupation and under a broker. BLS describes both residential and commercial sales work, says every state licenses brokers and agents, and distinguishes the sales agent from a broker licensed to manage a business. O*NET's Real Estate Sales Agents profile provides an independent occupational task taxonomy, but its data are not counted as vacancies in this study.
BLS reported about 427,000 U.S. real estate sales-agent jobs in 2025 within a combined broker-and-agent total of 530,600. Its combined-occupation projection estimated around 40,400 annual openings on average over 2025–35. These official occupational figures provide context; they are not the denominator for this web-collected sample, a forecast of course purchases or a promise of employment.
Method
Researchers identified public employer or employer-hosted applicant-tracking pages through focused web discovery, then reopened every accepted requisition through native web or an approved browser on 2 October 2026. Earlier machine-collected candidate lists were used only to discover URLs; no shell-HTTP-derived coding was admitted as evidence. Each accepted record has a distinct requisition key and canonical posting URL, a U.S. location, an employer, role title, observed active-page status, short necessary source excerpt, segment and template-family flag. Its coding preserves unknowns rather than treating a silent posting as a negative statement.
The final accepted source ledger contains 145 unique URLs and 145 unique requisition keys. It draws on 50 approved-route ATS records, 50 approved-route brokerage and employer records, and 45 additional residential consumer postings. Location-specific requisitions were separate units when their IDs differed, yet text reused across locations was assigned an employer-template family. A national or multi-location opening with one requisition was counted once. This distinction matters: 129 of 145 records are marked as reusing an employer template. Thirty-six Marcus & Millichap openings, for example, condense to two exact job-description variants in the checked Workday pages. Repetition shows current openings across locations, not 36 independent employers agreeing on every requirement.
The Role Requirements Matrix applies two explicit coding passes. First, it counts stated duties only from fields coded as responsibilities, using a per-phrase taxonomy; it does not turn employer benefits, product access or candidate traits into assigned work. Second, it counts stated skills separately from the hard-skill and observable-soft-skill fields. A code matches one source-derived phrase at a time, so a pattern cannot bridge separate fields to manufacture a task. Raw requisition counts are reported alongside the number of employers and employer-template clusters supporting each theme. They are not national percentages. The underlying accepted ledger links each coded record to its public posting, and the matrix preserves the supporting requisition keys for each table entry. The archival supplements provide the complete 145-row derived requirements matrix, a source-URL roster and a data dictionary, without reproducing employer job-description text.
Sample composition
| Segment of accepted posting | Distinct requisitions | What the segment mainly represents in this sample |
|---|---|---|
| Residential consumer sales | 60 | Home-buyer and home-seller agent roles, including appointment-led models |
| Investor-focused sales | 44 | Investors and off-market or investment-property opportunities |
| Commercial brokerage sales | 38 | Primarily commercial investment brokerage openings |
| Mixed or unclear | 3 | Public sales-agent roles whose segment cannot be assigned confidently |
| Total | 145 | 11 employers; structured purposive sample |
The distribution was chosen to expose different forms of agent work, not to match the U.S. occupation's actual segment proportions. Sixty residential postings come from a small set of brokerages, and many Mark Spain Real Estate and Orchard pages use common wording. Investor evidence is also concentrated in repeat location pages from employers such as New Western. Commercial evidence is dominated by Marcus & Millichap. No pooled count below should be read as a random-sample probability.
What the current postings state
The table reports coded duty statements, not skill claims or employer-provided products. A posting may match several rows. A zero or low count can reflect sparse advertising text; it cannot establish that a duty is absent from the profession.
| Duty code | Requisitions stating the duty, of 145 | Employers | Template clusters | Segment interpretation |
|---|---|---|---|---|
| Listing and property marketing | 103 | 7 | 10 | Appears across segments, with different property and client contexts |
| Transaction coordination and closing | 86 | 6 | 12 | Includes deal progression or closing language; legal authority is not inferred |
| Prospecting and business development | 82 | 5 | 7 | Especially concentrated in investor and commercial postings |
| Market and property analysis | 59 | 6 | 9 | Evidence on property comparison, market work or pricing varies by segment |
| Offers and negotiation | 45 | 4 | 10 | Most of this sample's explicit duty language comes from commercial pages |
| Appointments, showings and open houses | 37 | 3 | 8 | Primarily the residential consumer and mixed rows collected here |
| Follow-up and record control | 35 | 2 | 5 | Explicit duty wording is concentrated in appointment-led residential models |
| Collaboration and handoffs | 7 | 2 | 4 | A conservative code; many advertisements leave formal handoffs unstated |
The employer count is the most useful restraint on interpretation. Eighty-two prospecting-coded requisitions arise from only five employers; 36 commercial postings share one employer and two exact description variants. A requirement repeated across cities is important to those openings but is not independent evidence from as many organizations. Likewise, the low collaboration-code count does not mean agents rarely work with others. The matrix requires explicit work wording rather than inferring a duty from a brokerage name or an advertised support service.
The skill pass found explicit communication or relationship language in 134 requisitions from nine employers and 17 template clusters. Sales or prospecting skill language appeared in 103 requisitions from eight employers. Market and pricing skills appeared in 86, digital or CRM skills in 64, and organization or follow-up skills in 48. These are source-text code counts, not measures of proficiency or a nationally valid rank. Soft skills are interpreted as behaviors—timely response, clear explanations, rapport, organized follow-up or coachability—when a posting actually states them. Employer-supplied software, lead appointments, coaching, transaction coordination and proprietary products are not silently converted into learner competencies.
Segment differences and practical outputs
The residential consumer segment includes agent openings centered on prepared buyer or seller appointments, client communication, home pricing or positioning, and follow-through. Some postings explicitly mention brokerage-provided leads or transaction coordination; those are employer offers, not evidence that a new agent can independently generate a pipeline or manage legal transaction documents. Several Orchard advertisements mainly describe support and candidate criteria. Their 13 records remain in the current-vacancy sample, but duties and work products are marked unknown and contribute zero to duty counts. That protects the study from filling a sparse advertisement with occupational assumptions.
Investor-focused postings emphasize relationships with investors, learning the investor's goals, identifying or matching property opportunities and progressing a deal. These are not interchangeable with advising an individual home buyer about financing or a commercial seller about an investment property. Commercial brokerage postings, largely from Marcus & Millichap, emphasize prospecting and market work; their repeated templates explain why raw counts for some codes are large. The market cannot be summarized as if these were 145 descriptions of one homogeneous day-one job.
Where an advertisement states an output, the evidence includes an appointment or client interaction completed, a property or investor opportunity evaluated, a marketing or listing proposal, and a transaction advanced or closed. Many listings do not specify the form of the work product, who approves it, or a named handoff. A model workflow can therefore use these output categories as research-derived teaching targets, but it cannot assert that all U.S. brokerages use one form, platform or approval sequence.
Tools, qualifications, cadence and authority
The postings name a mixture of CRM systems, brokerage marketplaces, pricing or analytics tools, listing technology and employer-specific services. A tool mentioned in an advertisement may be offered to the agent, required of a candidate, or part of a stated duty; the ledger retains that distinction. The evidence supports the transferable skill of checking data provenance, maintaining accurate client and property records, and using authorized systems. It does not establish one universal MLS interface or a national AI adoption rate.
License wording is common in the collected ads: 144 of 145 contain some license criterion in their stated qualification text, ranging from an active license to readiness to obtain one. This count is only a description of the selected postings and does not determine whether a criterion is legally sufficient in a particular state. One Mark Spain posting locates the job in Rock Hill, South Carolina while specifying North Carolina license wording; the discrepancy is flagged unresolved and excluded from state-specific license interpretation. Three other residential buyer-agent postings specify a job location but do not name the license state in the qualification wording. Among the 45 additional residential records, only 41 have unambiguous state-specific license-comparison eligibility; no state-license prevalence estimate is calculated. BLS, not this sample, supplies the general fact that every state requires agents to be licensed.
Cadence and decision rights are incompletely disclosed. Some ads say full-time, evenings or weekends, or appointment-driven work; many omit weekly, monthly and lifecycle rhythm. Some identify a mentor, manager, broker or transaction-support service, yet do not state who signs a contract, approves a price, interprets a disclosure or handles a legal escalation. The Role Requirements Matrix records those fields as unknown unless the posting says otherwise. Professional preparation should teach a learner to ask an employing broker for the applicable work instructions and authority boundaries before a real transaction, rather than infer permission from a job-ad phrase.
Geography, rights and limitations
Every accepted vacancy in the principal sample is a public U.S. opening with an employer or employer-hosted source URL. This is a point-in-time collection, not a longitudinal hiring dataset. Posting dates were often absent; an open page on 2 October does not prove a new job was created that day. The sample is skewed toward employers with public structured boards and repeated location templates. Independent small brokerages, informal recruitment and market sectors with less public ATS exposure may be underobserved. We did not stratify by national employer population or validate a sampling frame that would justify calling the results representative. Counts describe these 145 identified openings only.
The main current-change article, U.S. Real Estate Agent Work in 2026: Six Recent Changes to Track, uses a separate public-source corpus. It focuses chiefly on residential transactions. Its national market measures cannot validate a commercial or investor vacancy count, and its Colorado regulatory example remains Colorado-only. A geography-coherence record binds each material report claim to the U.S. vacancy population, the relevant practice segment, and BLS or other applicable context. No non-U.S. source fills a U.S. vacancy gap.
Public job pages are cited by link and represented through original, minimal codes and short necessary excerpts in the private ledger. This report does not reproduce full job descriptions, employer forms, NAR member content or proprietary training material. NAR's membership-mark rules reserve REALTOR as a membership mark, so it is not used as a synonym for agent or as a course brand. The report's original compilation may be licensed by MTF; linked third-party job pages retain their owners' rights.
Conclusion
The evidence supports a real role-development problem: newly entering or transitioning agents must connect client contact, property information, authorized listing and showing activity, documented follow-through and broker-supervised decisions. It also shows why a single generic playbook would be misleading. Residential consumer, investor-focused and commercial brokerage postings stress different situations, while many ads do not spell out approvals, cadence or a usable work product. A defensible professional course can teach adaptable communication, evidence and record habits and explicit escalation, provided it does not present its certificate as a state license, approved prelicensing or continuing education, or permission to perform licensed acts. The exact course title requested by the owner remains Real Estate Agent Professional Certificate; this research report is an evidence input, not a credentialing authority.
Citation: MTF Institute Research Team. U.S. Real Estate Sales Agent Roles: Evidence from 145 Current Vacancies (October 2026). MTF Institute Research Report MTF-CF-RR-2026-10-02-REA-01, 2 October 2026.
Continue learning
Apply the evidence from this report through MTF Institute's Real Estate Agent Professional Certificate. The programme turns the identified capabilities into structured theory, guided AI practice and reusable workplace artifacts.