How to Start a Recruitment Agency: The Complete 2026 Blueprint.

Your buyer is almost never a CEO. In a 20 to 300 person company it is the HR manager or director of talent acquisition; in a manufacturing or distribution setting it is the plant manager or operations supervisor who needs 14 pickers on Monday; in professional services it is the hiring manager of the specific practice group with an open requisition and a bonus tied to filling it. In firms under 50 employees, the buyer is often the owner or the office manager who also runs payroll. These four buyer types behave completely differently: the plant manager will call you at 6 a.m. and pay net-30 without negotiating a rate card, while the corporate TA director will run you through a vendor management system, a master service agreement, insurance certificate review, and a 5 to 12 week procurement cycle before your first placement.

The economics split hard along one line: temporary/contract staffing versus direct-hire placement. Temp means you are the employer of record, you run W-2 payroll weekly, you carry workers' compensation, and you bill the client a marked-up hourly rate. Direct-hire means you invoice a one-time fee of 20 to 30% of first-year salary and carry no payroll at all. Temp produces recurring revenue and gross margins of roughly 18 to 28% in commercial and light industrial, 30 to 40% in professional staffing. Direct-hire produces lumpy revenue with effective gross margins of 60 to 80% once recruiter pay is separated out, but a single fallen-through placement can erase a month.

Realistic revenue targets for an independent firm: $500,000 to $2 million in annual revenue within the first few years for a general small agency, and $300,000 to $1 million in fee revenue for a lean executive search boutique running two or three people. Net margins run 10 to 20% before owner distributions in steady state, against 5 to 12% for the publicly traded staffing companies that carry corporate overhead. Expect breakeven or negative margins for the first 12 to 24 months. The barrier to entry is close to zero, which is exactly why the barrier to survival is high: the winners are the ones with a defensible niche, not the ones with the nicest website.

For direct-hire and professional search, physical location is close to irrelevant and a home office or a $200 to $500 per month coworking desk is correct. Candidates and hiring managers interview over Zoom and Teams; nobody visits your office. Spending $10,000 to $50,000 on a traditional office for a search practice is capital you should have put into LinkedIn Recruiter seats at $10,000 to $12,000 per year each and a sourcing tool like SeekOut or ZoomInfo. The exception is if your niche is genuinely local relationship-driven, such as regional construction or public sector, where being seen at the local chamber, the trade association chapter, and the community college advisory board actually converts.

Light industrial and commercial temp staffing is the opposite. You want a ground-floor storefront on a bus line, within 3 to 8 miles of the industrial park or distribution cluster you serve, with parking, a lobby that seats 8 to 12 people, and a small room with two computers for I-9 verification, safety orientation, and E-Verify. Walk-in traffic is a real recruiting channel in this segment, and warehouse clients will judge you on whether you can put 15 people through onboarding on a Sunday afternoon. Expect 600 to 1,200 square feet at $12 to $30 per square foot per year depending on the metro, plus $1,000 to $6,000 for deposit and first month.

Then there is the channel decision that matters more than the address: direct client relationships versus vendor management systems and managed service providers. Going direct means you talk to the hiring manager, set your own rate, and keep 100% of the spread. Going through a VMS like Fieldglass, Beeline or Coupa gets you volume and requisitions without cold calling, but the MSP takes 2 to 5% off the top, rates are capped, you compete with 10 to 40 other suppliers on the same req, and payment terms stretch to net-45 or net-60. New agencies should build direct first. VMS work is a supplement once your float can absorb the longer terms, not a launch strategy.

Take a light industrial temp at $18 per hour pay rate. You bill the client at a 1.55 markup, so $27.90 per hour. Your cost of that hour is not $18: add employer FICA at 7.65%, federal and state unemployment at roughly 1 to 4% of taxable wages early on, and workers' compensation at anywhere from 2% for clerical codes to 8 or 12% for warehouse and manufacturing codes. Call the loaded cost $21.60 to $22.50. Your gross profit is roughly $5.40 to $6.30 per hour, or 19 to 23%, which sits inside the 18 to 28% band the industry reports for commercial and light industrial. One temp working 40 hours a week generates about $216 to $252 per week in gross profit. You need 30 to 45 active temps just to cover a modest office and two recruiters.

Professional and technical contract staffing runs better: a $65 per hour contractor billed at $92 to $98 produces 30 to 40% gross margin, roughly $27 to $33 gross profit per hour, so 10 to 15 active contractors do what 40 light industrial temps do. This is why capital-constrained founders should look hard at professional contract before commercial temp. The float per head is higher, but the gross profit per head is two to five times higher.

Direct-hire is a different animal. A $90,000 salary at a 22% fee is a $19,800 invoice. Delivery cost is your time plus tools: assume 25 to 45 recruiter hours across sourcing, screening, coordination, and closing, plus a share of the roughly $1,500 to $3,000 per month you spend on ATS, LinkedIn Recruiter, and job boards. If the recruiter is on $60,000 base plus commission, fully loaded cost per placement is commonly $4,000 to $8,000, leaving $12,000 to $16,000 in contribution. But apply the fall-off rate: 10 to 20% of placements do not survive the 30 to 90 day guarantee, and you refund or replace. Model 8 to 14 placements per recruiter per year in professional search, not the 24 the optimistic spreadsheet says.

At opening, most independent agencies are one person doing a full desk: business development in the morning, sourcing and screening in the afternoon, and payroll or invoicing on Friday. That is the honest picture, and it works for 6 to 18 months. The first hire depends on which half of the job you are worse at. If you are a strong salesperson, hire a recruiter or sourcer at $45,000 to $65,000 base plus a per-placement bonus of $500 to $2,000 or 5 to 10% of gross profit. If you came from delivery, hire a business development manager at $55,000 to $80,000 base plus 5 to 10% commission on gross profit, and accept a 4 to 8 month ramp before they pay for themselves.

For temp models, the role people skip and regret is the staffing coordinator or on-site supervisor: $40,000 to $55,000, handling timecard collection, I-9 and E-Verify, no-show replacement, safety orientation, and the 6 a.m. call when three people do not show at the plant. Without this person the owner spends every morning firefighting and never sells. In light industrial, expect to need one coordinator per roughly 60 to 120 active temps. Back office (payroll processing, tax filings, certificates, invoicing) is usually cheaper to outsource at 2 to 6% of gross payroll than to hire for, until you are past roughly $4 million to $6 million in annual revenue.

Two roles you do not need at opening: a marketing hire and a full-time recruiter for a niche you have not proven. Both are ways of spending money to avoid making cold calls. What you do need on contract is a CPA who has handled staffing clients (multi-state payroll tax gets ugly fast, and $150 to $400 per month is cheap insurance) and an employment attorney on call for co-employment questions, restrictive covenant disputes, and the first time a client tries to hire your temp directly without paying the conversion fee. Budget $2,000 to $6,000 per year in ad hoc legal after formation.

Build a named list of 150 to 300 companies inside your niche and geography, with the specific hiring manager name, direct dial, and email, not a general info@ address. Use LinkedIn Sales Navigator or ZoomInfo, then verify by calling the switchboard. Then work it hard: a realistic daily standard for a founder is 40 to 60 dials, 20 to 30 personalized emails, and 2 to 5 meaningful conversations. Expect roughly 1 signed MSA per 60 to 120 companies contacted in the first quarter, and 4 to 8 weeks of consistent outbound before the first agreement. Anyone telling you inbound marketing will fill your pipeline in the first 90 days has not run a new desk.

The single highest-converting play for a new agency is the marketing call with a specific candidate, not a generic capability pitch. Source three genuinely strong, available people in your niche, then call hiring managers and say you have a Class A CDL driver with a clean MVR available Monday, or a controller with NetSuite and a two-week notice period. This turns a cold sales call into a supply conversation and routinely produces a job order on the first contact. Recruit first, sell second. It also proves your delivery capability before anyone has to trust your brochure.

Layer in three cheap channels that compound. First, mine your own network: former colleagues who are now hiring managers convert at 5 to 10 times cold rates, and the first 10 to 20 clients of most successful agencies came from the founder's prior employer's ecosystem. Second, attend the two or three trade associations where your buyers actually go, such as a local SHRM chapter for HR buyers, ASSP or a manufacturers association for plant buyers; a chapter membership runs $100 to $500 per year. Third, post real jobs on Indeed and LinkedIn even when they are speculative-but-honest openings in your niche, which builds the candidate database that makes the marketing call possible. Track everything in the ATS from day one: companies contacted, conversations, MSAs signed, job orders received, submittals, interviews, placements. If your submittal-to-interview ratio is worse than 1 in 3, your problem is candidate quality, not sales volume.