Recruitment agencies, Job Boards and Headhunters

Choose a hiring channel before choosing a person

Build 3674 has three employer-side recruitment channels: the two agencies, in-store Job Boards, and headquarters headhunters. They solve different stages of the same pipeline.

Anderson Recruitment is the early operational channel for roles supported by ordinary retail and warehouse businesses, while City Workforce supplies educated office and headquarters roles. The selected business filters the available occupations: choosing a law firm will not produce a cashier, and choosing a shop will not expose an HR manager. Multiple developer replies identify City Workforce as the correct source when a headquarters player sees only cleaners at the other agency. The apparent missing-role bug is therefore usually a wrong agency or wrong destination business, although a restart and F2 report remain appropriate after both are confirmed.

The Job Board, added officially in EA 0.8, is a purchased object placed in a store. It attracts walk-in candidates supported by that store, most commonly its primary role. The official description calls them easier to recruit but not necessarily highly skilled. It is a low-attention pipeline rather than a precise search. The candidates still appear in MyEmployees under Recruits, so the board does not bypass screening, expiry, assignment or schedule work. Use boards where steady turnover is acceptable and a broad compatible applicant is more valuable than an exact demand profile.

A headhunter is an HQ employee and the late-game precision channel. It can search roles across the organization, set skill and demand filters, continue or stop after a selected candidate count, and link to HR managers for permanent replacement. Its value is control and scale, not magic availability. Highly restrictive filters can define an empty population. The hiring decision should therefore begin with required role, start date, mature demand profile and acceptable acquisition cost, then select the cheapest channel capable of meeting those requirements on time.

Version map and claims to reject

Hiring advice written before EA 0.3, EA 0.5 or EA 0.8 omits systems that now define candidate quality and workforce continuity.

EA 0.3 replaced agency age selection with full-time and part-time selection, moved applications into a sortable candidate tab, imposed a seven-day candidate lifetime, and introduced headhunters plus automatic replacement. EA 0.5 added salary negotiation and rival poaching, then made employees remember worked hours so a midweek schedule edit could not erase overtime. EA 0.8 added Job Boards and improved candidate and scheduling workflows. Version 1.0 retained these systems and Build 3674 fixed a quit-state fault that had interrupted training and management processes.

Reject the obsolete assertion that a fully trained HR manager supports twenty-five employees. Current evidence places the Build 3674 planning cap at fifty, with lower capacity at lower skill. Reject the idea that headhunter exclusions permanently control demands when hiring low-skill people: a developer explains that filters inspect demands visible at recruitment, while low-skill workers may reveal additional demands later. Reject exact universal candidate wages and agency fees because difficulty settings, negotiation and balance patches affect them. Record live quotes instead.

SEO pages commonly mix the old twenty-five-person cap, pre-negotiation wages and current Job Board screenshots. A page can be accurate about one button and wrong about the economics around it. Accept a numeric claim only when it is in current F1 text, the Build 3674 interface, a later official update, a dated developer reply not contradicted by a subsequent patch, or a reproducible test that names version, difficulty and controls. Reference catalogs can identify role names, but they cannot establish current caps merely by being indexed recently.

Convert an opening into a candidate specification

A good requisition describes coverage and constraints, not merely a job title. This prevents paying for a candidate who cannot fit the roster that created the vacancy.

Start with workstation-hours. If one register must be covered from 08:00 to 20:00 seven days, the requirement is eighty-four hours, not one cashier. Subtract hours already covered by compatible employees, then classify the residual by day and time. Forty weekday morning hours suggest one full-time worker; fourteen late-evening hours suggest a part-timer; scattered two-hour gaps may justify changing opening hours before recruiting anyone. Add overlap only when queues or lost-customer evidence demonstrate that simultaneous capacity produces incremental contribution.

Write a candidate specification with role, destination, employment band, target weekly hours, required days, prohibited demands, minimum useful skill, latest start date and loaded-cost ceiling. Distinguish a true deal breaker from a preference. If the store needs weekends, No Weekends is a deal breaker; a sofa request is usually a manageable preference. Excess exclusions reduce headhunter throughput, while vague agency orders shift the screening burden into a pile of applicants that expire after seven days.

Include the future state. A low-skill worker may develop additional demands and a higher wage while an expert arrives with a more mature visible profile. A headquarters role may need insurance eligibility and enough skill to unlock useful plan capacity. A factory worker should match the stage that constrains completed output, not the machine with the most empty stations. Forecasting six to twelve in-game days beyond opening prevents hiring the cheapest immediate fit and replacing the same person after the first training cycle.

Agency hiring workflow

Recruitment agencies are most efficient when the business selection, employment status and campaign size are chosen from a prepared requisition.

Create or select the destination business before calling. The agency uses that business to limit occupations to roles the location can employ. If City Workforce offers only a cleaner for an HR requisition, verify that an HQ business—not a generic office or another company—is selected. Developer troubleshooting also requires the HR employee to be assigned, scheduled and attached to a BizMan plan for later systems, but the initial candidate menu begins with the correct business type and agency.

Order more candidates than final seats when screening demands matters, but do not create an unreviewable backlog. A practical early-game ratio is two or three candidates per seat, adjusted upward for unusual weekend or evening requirements. This is an operating heuristic, not a game probability. Set a review appointment before the seven-day expiry. Compare every candidate on the same scorecard instead of hiring the first acceptable name and discovering a cheaper, cleaner fit after paying training fees.

Agency hiring is fast but relatively blunt. It works well for the first stores, for a role with inexpensive workplace demands, or when vacancy loss exceeds the expected savings from a longer search. It performs poorly when dozens of identical specialists must satisfy a precise schedule. At that scale, establish HQ headhunting and HR capacity first. Do not delay a profitable opening merely to unlock perfect automation, but include manual screening time and later replacement risk in the agency path's true cost.

Job Board workflow and limits

The Job Board is a passive candidate generator for the store where it is placed. Its strength is low friction; its weakness is limited targeting and generally modest skill.

Buy and place the board in an operating store, then monitor Employees and Recruits. A developer states that applicants should be for any position supported by that store, usually the primary position. A fast-food Job Board producing security applicants was considered unexpected, while a gift shop can legitimately surface security. Record the source store whenever a walk-in arrives; without that field, a multi-store owner cannot tell whether the role is correct or reproduce a mismatch.

Boards are useful for rolling bench strength. Place them in businesses whose primary role recurs throughout the empire, accept promising workers before a planned expansion, and train them through HR while existing stores remain covered. They are less suitable when the opening has a hard deadline, requires an HQ specialist, or demands a fully known mature preference set. Passive does not mean free: the board has an acquisition cost, candidates consume review time, and premature hires consume wages or training resources.

Test board yield with version-labelled observation. For thirty in-game days, record board location, business type, each applicant's role, skill, status, demands and arrival interval. Do not change the business type during the test. Report median skill and role distribution, not only the best applicant. A board that produces one useful hire among twenty may still be worthwhile for a high-turnover cashier pool, but the same yield is unacceptable for a sole purchasing agent needed before Monday's import cycle.

Headhunter search design

Headhunter controls are exclusions and thresholds. Misreading an exclusion as a desired trait can make a valid employee population mathematically impossible.

A red exclusion means the headhunter must reject candidates with that condition. Excluding both full-time and part-time excludes everyone because every candidate belongs to one employment band. Stacking No Weekends, No Evenings, fixed-day, equipment, insurance and high-skill conditions can leave no feasible match. When a search stalls, remove one restriction at a time, beginning with preferences that the existing workplace can cheaply satisfy, and observe whether candidates resume.

Skill threshold and demand certainty trade against one another. Developer guidance says a low-level recruit can reveal future demands that were not present for the headhunter to filter. A candidate hired near 100 percent has the mature demands visible, so exclusions are much more reliable. For business-critical specialists, search high enough to expose demands and avoid downtime. For high-volume roles, accepting lower skill and passive HR development can be economical if the organization can absorb later preferences.

Set a candidate limit rather than leaving every plan continuous. A developer confirms the headhunter can stop after a selected number, and the candidate tab still has finite review capacity. Define open seats plus a small screening surplus, pause when filled, and reopen before an expansion or retirement wave. Continuous recruitment is justified only when measured turnover or rapid buildout consumes candidates faster than they expire. It is not a substitute for linking headhunters to HR managers for automatic permanent replacement.

Candidate scorecard and loaded cost

Hourly wage is only one component of hiring cost. A scorecard should convert visible differences into comparable operating consequences.

For each candidate, record primary skill, hourly wage, requested employment status, schedule restrictions, equipment demands, insurance tier, age or retirement horizon, recruitment fee and negotiation terms. Add intended weekly hours and the HR group that could manage the hire. Mark each demand as already satisfied, shared-cost accommodation, individual-cost accommodation or incompatible. This immediately distinguishes a slightly higher wage candidate who fits the current system from a cheap candidate who requires a new insurance plan and custom workstation.

Estimate loaded weekly cost as wage times hours, plus insurance allocation, amortized recruitment and training, and any incremental amenity or schedule cost. For insurance, use the actual HR plan quote and remember the ten-person minimum: adding the first insured worker to an otherwise empty plan can effectively trigger charges for ten, while adding the eleventh worker is normally marginal. Avoid assigning a universal dollar amount because plan tier, negotiation, HR skill and difficulty can change it.

Score expected contribution separately. Skill matters only where the role is a bottleneck. A 100-percent cashier in an empty store is not worth more than a 50-percent cashier merely because the number is green; marketing, stock or demand may cap sales. A high-skill purchasing agent serving large imports can save more than their wage premium. Rank candidates by expected contribution minus loaded cost, then use demand fit and succession risk as tie breakers. Keep the calculation visible so later results can correct the assumptions.

Worked hiring decision for a seven-day shop

This example compares full-time and part-time candidates for one 10:00–20:00 register open every day, requiring seventy weekly workstation-hours.

Option A hires two full-time workers for thirty-five hours each. Candidate A1 works five seven-hour weekday shifts; A2 works five shifts covering the weekend and three weekdays. Option B hires one full-time worker for forty-two hours and one part-time worker for twenty-eight. Option C hires three part-timers around twenty-four, twenty-four and twenty-two hours. All three cover seventy hours arithmetically, but only demand-compatible days and times make them valid.

Suppose each worker has the same wage, but insurance is charged per covered person. Option B generally has lower benefit headcount than Option C and more flexibility than two full-timers balanced close to the minimum. If the full-timer refuses weekends, B fails unless the part-timer can legally and operationally cover all weekend hours. If peak periods require two registers, the apparent residual hours change: recruit from the peak coverage matrix, not the simple single-register total.

Select after calculating loaded cost and resilience. Assign a notional vacancy-loss value to every uncovered hour, then simulate one sick day. HR temporary staffing may cover a managed employee, but a new business without HQ must carry its own redundancy. If losing a single worker closes the shop, the cheapest roster may be too brittle. The decision record should state why the chosen design wins: lower cost, fewer incompatible demands, insurance efficiency, peak flexibility or better succession—not merely fewer names.

Skill strategy: hire expert or develop novice

The expert-versus-trainee choice is an investment comparison with uncertain future demands, not a simple rule that one route is always cheaper.

The expert path pays a higher search or wage cost but gains immediate output, visible mature demands and no direct training downtime. The novice path may lower acquisition cost, can work while HR trains them, and spreads development expense, but higher skill can raise wages and reveal insurance or equipment demands. Manual training can also remove the employee from service. Price both paths through the date when the novice reaches the expert's skill, not only on acceptance day.

Use total comparison cost: recruitment and negotiation payments, wages through the comparison date, training charges, contribution lost while unassigned, HR overhead, insurance added after demands appear, and the terminal weekly wage. A March 2026 player test on a named contemporary build path found trained lawyers and HR managers costing more per hour than newly hired 100-percent equivalents, conflicting with an older patch note that training should remain cheaper. This is unresolved enough that every difficulty should be tested rather than generalized.

Run a paired Build 3674 test. Recruit two same-role candidates within a narrow starting-wage band, one near the intended operating skill and one low. Record every cost and wage at each ten-point skill interval, keep both under equivalent HR and insurance conditions, and compare contribution after maturity. Do not claim causation from two different roles or difficulties. Even if the expert is financially cheaper, novice hiring may remain valuable when high-skill filters produce no candidates before a launch deadline.

Negotiation and retention economics

Candidate negotiation can reduce recurring wages, but an up-front payment or failed deal has a payback period that should be calculated before acceptance.

For a wage reduction, payback weeks equal negotiation cost divided by hourly saving times scheduled weekly hours. If a five-dollar reduction costs ten thousand and the worker serves forty hours, the weekly saving is two hundred and simple payback is fifty weeks. That can exceed the worker's useful tenure or the player's expansion horizon. A smaller up-front payment with a lower acceptance chance also has an expected cost: include failed attempts and the value of delay.

Use negotiation aggressively for high-hour, long-tenure specialists when the payback is short relative to expected service. Be conservative for part-timers, temporary launch staff or roles likely to be automated away. Do not use a real-world bargaining formula as a hidden game rule; the interface provides the offer and response, while exact acceptance probabilities are not officially published. Record proposed wage, payment, response and skill to build a save-specific negotiating history.

The same logic applies when a rival poaches an employee. Added retention cost is raise times remaining weekly hours plus any new demand expense. Replacement cost includes headhunter or agency cost, vacancy contribution, training, schedule repair and lower initial output. Retain when the present cost of the concession is lower than expected replacement loss. A linked headhunter shortens vacancy and changes that threshold, so hiring architecture and retention policy cannot be optimized separately.

Demand priority during screening

Screen the critical employment demand first, then schedule compatibility, then benefits and workplace items. This order follows operational damage rather than visual prominence.

Full-time or part-time is normally Critical. A mismatch continuously pressures satisfaction and cannot be repaired by a cheap chair. Day-count, weekend and time-of-day demands determine whether the candidate fits the roster. Insurance is often Important in current developer discussion; meeting employment status can still produce a positive trend when insurance is absent, unless another demand adds enough negative weight. This makes an uninsured early hire possible, but not automatically optimal.

Classify demands by marginal accommodation cost. A refrigerator or shared sofa may satisfy many workers for one purchase. An administrative-desk preference consumes floor area and can force a blueprint change. A Gold plan may be cheap at scale but disproportionately costly when it opens an otherwise unused ten-person minimum. Owner-happiness or peaceful-environment demands introduce organization-wide constraints that may be inconvenient even when their direct dollar cost is low. The cheapest wage can therefore carry the highest operational cost.

For a high-skill headhunter search, exclude genuinely incompatible demands because they should already be visible. For low-skill recruiting, reserve contingency for later demands and avoid building a roster whose economics fail if insurance appears. A robust policy might accept one manageable future demand per worker while rejecting combinations that require both a custom schedule and a new benefit tier. State that policy before reviewing names so attractive skill percentages do not override the operating design.

Bulk hiring, onboarding and launch control

Large openings fail when accepting candidates is treated as completion. Onboarding must create assignments, schedules, management links and cost visibility before doors open.

Build an intake table with candidate, role, destination, workstation, shift template, uniform, HR manager, training route and insurance tier. Accept in batches small enough to place immediately. An employee left unassigned after acceptance can generate cost or satisfaction risk while contributing nothing, and a manual training command may erase the business assignment that was planned during screening. Use Hire and Assign where available, but verify the resulting schedule rather than assuming assignment means coverage.

Stage specialists before volume staff. Purchasing, logistics, HR and headhunter plans may need several days of setup, negotiation or training. For retail, secure the workers covering mandatory base stations before peak overlap and optional cleaners. For a factory, staff the bottleneck recipe stage first. Every launch roster should pass four tests: all profitable opening hours covered, each employee within the employment band, every hard demand feasible, and management capacity available below its maximum.

Run a launch-week variance report. Compare planned versus actual scheduled hours, loaded payroll, uncovered workstation-hours, queues, sales, production and candidate wastage. Separate errors caused by demand from errors caused by hiring. If the shop has no customers, another cashier is not the answer. If customers leave a visible queue while a compatible candidate expired, the requisition or review cadence failed. Correct the pipeline stage that produced the variance instead of simply ordering more people.

Automatic replacement and succession hiring

Permanent replacement requires a headhunter linked to the HR managers who own the affected employees. HR temporary staff cover sickness only.

Official EA 0.3 notes describe automatic replacement for employees who retire or resign, and later developer guidance expands it to quits, poaching and retirement. The replacement should be broadly similar in wage, demands and skill but younger. Do not interpret similar as an exact clone; audit the new employee's status, schedule and demands. Replacement can take time, especially for poached or resigned employees, while retirement has had special timing behavior in older notes.

One developer reply summarizes scale as one headhunter per roughly one hundred employees, reflecting two fully trained HR groups of fifty. The configuration surface may present this as two HR-manager links rather than a direct one-hundred-person number. Use the displayed plan capacity as the actual limit and leave buffer. If either HR group is under-skilled, the headhunter's theoretical downstream coverage is lower. Assigning two managers is not equivalent to covering one hundred people unless both groups are full.

Avoid circular succession. A developer diagnosed a case where the missing headhunter was assigned to an HR manager whose replacement depended on that same headhunter; nobody remained to replace the specialist. Use cross-covered pairs or a dedicated backup chain. Review the graph after every HQ reorganization and before an aging wave. The correct unit of succession planning is not headcount but critical links: each HR group, its headhunter, the headhunter's own HR owner, and an independent route back to a staffed plan.

Hiring diagnostics and audit fields

A hiring system is diagnosable when every search, candidate decision and placement has a minimal evidence trail.

For each requisition track request date, business, role, required hours, channel, employment band, minimum skill, exclusions, target candidate count, fee, promised arrival, actual candidates, expiry and final disposition. For each accepted worker add wage, negotiation payment, mature demands known, intended schedule, business, workstation, HR owner, insurance group, headhunter coverage and first productive day. These fields distinguish search scarcity from delayed review, rejection policy, onboarding delay and schedule failure.

When an agency lacks a role, verify selected business and correct agency, then restart once before reporting. When a headhunter finds nobody, inspect mutually exclusive filters and lower one constraint at a time. When Job Board applicants have an invalid role, record the board's store and business type. When candidates disappear, check the seven-day expiry. When an accepted hire cannot be scheduled, check business assignment, role compatibility, hours already worked and current training state. This sequence preserves evidence for F2.

Run a Monday pipeline audit: expiring candidates within two days, open requisitions, unassigned hires, workers in completed training, vacancies without successor coverage, HR groups above eighty-five percent, and headhunters recruiting without a current need. Monthly or before a major expansion, calculate channel yield, time to productive start, ninety-day equivalent retention, loaded acquisition cost and rejection reasons. Retire exclusions that eliminate candidates without preventing meaningful cost, and add a condition only after a documented failure demonstrates its value.

Build 3674 reproducible hiring tests

Small controlled tests turn uncertain community advice into local facts without pretending the result applies to every difficulty or future patch.

To test agency roles, create one compatible business of each family, call each agency with that business selected, and capture the role menu before ordering. Repeat after restarting only if a role is missing. To test candidate expiry, record arrival time and observe the candidate daily without hiring. To test Job Board output, use one unchanged store for thirty days and log every applicant. These tests establish interface behavior and distributions without relying on an SEO table.

To test headhunter filters, begin with role and employment status only, request a fixed small sample, then add one exclusion per run. Record search time and candidate count. Do not compare overlapping continuous searches because they compete with expiry and review timing. To test mature demands, recruit matched low- and high-skill cohorts, keep working conditions equal, and record demands at every skill milestone. The conclusion should be probabilistic: filters can inspect current demands, while later demands remain possible for developing employees.

To test total hiring economics, follow one worker from request through a full comparable operating period. Record fee, negotiation, wage, training, insurance, idle days, schedule hours, temporary cover and role output. Pair with another worker in the same role and difficulty hired through a different channel. Exclude launch hype and rival attacks. Publish the build, settings and raw observations alongside any recommendation. Retest after an update that mentions wages, candidates, training, demands, insurance or scheduler behavior.

Preserve rejected candidates as aggregate data rather than anecdotes. Count how many failed because of employment status, weekend limits, time-of-day restrictions, insurance, furniture, insufficient skill, high wage or simple role mismatch. Divide each reason by candidates reviewed for that channel. If forty percent fail one condition, decide whether changing the roster or workplace is cheaper than continuing to buy searches that reject the same people. Repeat the analysis separately for agencies, boards and headhunters because their populations and controls differ. Also track false economy: candidates accepted for a low wage who later required training, a new plan, schedule closure or early replacement. A hiring policy improves only when time to productive start, loaded cost and avoidable turnover improve together. A lower agency invoice with more empty workstation-hours is not success. Keep one unmodified comparison period after changing policy, and avoid changing wages, opening hours and screening rules simultaneously; otherwise the cause of any improvement cannot be identified.

Use explicit stop rules for random searches. Before ordering, define the maximum acceptable vacancy days, loaded weekly wage, training horizon, benefit-floor increase and schedule accommodations. If no candidate clears the screen before the vacancy-loss ceiling, relax one soft constraint or switch channels; do not keep paying because earlier fees are sunk. Log the exact change and restart the comparison clock. This makes hiring policy reproducible and prevents an idealized candidate profile from costing more through empty workstation-hours than any realistic mismatch would have cost.