Operations By Wonderful Adebagbo, Operations Manager, NSBC ·Published May 21, 2026 ·14 min read ·Last updated 2026-05-21
Quick Answer

Hiring your first employee is not a celebration moment, it is a system change. Done well, it gives you back 15 to 25 hours a week and unlocks the next revenue tier. Done badly, it drains cash, breaks your delivery, and pushes you closer to the burnout you were trying to escape. The 2026 playbook is simple: hire only when revenue and overload both signal go, hire a role that subtracts from your calendar instead of cloning you, replace the job description with a scorecard, run a four-stage interview process that ends in a paid trial, pay at market median, and onboard for measurable productivity in 30 days. Skip any of these steps and your odds drop fast.

Most founders make their first hire too late, then make the wrong one. They wait until they are drowning, panic-hire the first warm body who can spell their own name, then spend the next four months training, retraining, and apologising on the new hire's behalf. Twelve weeks later the founder is doing the work themselves again, plus paying a salary, plus carrying the guilt of letting someone go.

This guide is for the service-business founder about to make that hire for the first time. We have onboarded enough first hires across NSBC and our client base to know the pattern. The good news: hiring well is not a personality trait, it is a process. The bad news: most of the process is unglamorous and most founders skip it.

What follows is the exact playbook we use, sequenced for someone going from solo operator to operator-plus-one without taking on debt, breaking the delivery promise to current clients, or burning the runway that took you years to build.

One note before we start. This is not a talent acquisition guide written for a Fortune 500 HR team. It is a practical, founder-grade playbook for a service business doing somewhere between 10,000 USD and 80,000 USD per month who is about to hire person number one or two. The principles scale upward, but the language, the timelines, and the costs are calibrated for owners writing the cheque themselves and feeling every dollar of it.

When You Actually Need to Hire (Not When You Think You Do)

You probably do not need to hire because you "feel busy." Every founder feels busy. Busy is not a hiring signal. The real signal is a specific combination of two things showing up together for at least 90 days.

Signal 1: stable, repeatable revenue. You have three consecutive months of revenue that covers your current cost base, the new salary you are about to add, and at least a three-month cash buffer for that salary. Not a single big project. Not a hopeful pipeline. Actual collected cash, three months running.

Signal 2: you are spending over 30 percent of your working week on tasks that fall well below your unique skill ceiling. Inbox triage. Calendar wrangling. Invoice chasing. Status updates. Booking travel. Repetitive admin. If you charge 150 USD an hour for strategy and you are spending 15 hours a week on tasks a 20 USD an hour assistant could do, you are losing 1,950 USD a week. That is not a budget problem, that is a math problem.

Both signals together mean hire. Only one signal means wait. If revenue is unstable, hiring will accelerate the cash crisis. If you have cash but you are not actually overloaded, you will hire someone you cannot keep busy, then resent paying them.

The Founder Self-Audit

Before you write a single line of a job description, do this 30-minute audit. Block out a quiet hour. Open last week's calendar and inbox. Categorise every task into one of four buckets:

Count the hours in each bucket. If your routine plus wasted buckets together exceed 15 hours a week, you have a delegation case. If your important bucket is also large, you may have a case for a more senior hire later. Either way, this audit is the foundation document for the rest of the process.

Two more readiness checks worth running before you spend a single hour writing a job post. First, the delegation reflex test: have you ever successfully handed off a real task to a contractor, freelancer, or family member and not redone it yourself? If the honest answer is no, you have a delegation muscle to build before the hire arrives. Practise on a few one-off contractor projects first; you will save yourself the more expensive version of the same lesson. Second, the documentation test: could a sharp stranger reproduce 30 percent of your week from written instructions? If your business lives only in your head, the new hire will spend their first 60 days extracting information from you and will resent you for it. Even three rough SOPs written before day one will pay back a hundred times over.

If the audit, the delegation reflex, and the documentation test all line up, you are not just ready to hire, you are ready to onboard. That distinction matters. Most founders are ready to hire long before they are ready to onboard, and the gap between the two is where bad hires live.

What Role to Hire First (Hint: Not Another You)

The single most common first-hire mistake is hiring a mini-you. You are a marketing consultant, so you hire a junior marketing consultant. You are a designer, so you hire a junior designer. It feels intuitive. It is almost always wrong.

Why? Because the work clogging your calendar is rarely the work you are uniquely good at. It is the admin, the operations, the inbox, the scheduling, the invoicing, the project coordination, the client follow-up. Cloning your skill set does not free up the time. It just adds another person doing the same kind of work, while the admin pile keeps growing.

The Subtraction Hire

The right first hire is a subtraction hire. They subtract the most hours from your week while costing the least. For most service-business founders, that means one of three roles.

Role A: the Executive Assistant or Operations Assistant. Owns the inbox triage, calendar, scheduling, travel, expense tracking, invoice issuance and chasing, basic client communication, and meeting prep. Typically remote, 20 to 40 hours a week, costs 800 to 2,500 USD per month depending on geography. Gives back 12 to 20 founder hours a week.

Role B: the Generalist Operations Coordinator. One step up from EA. Owns the EA scope plus light project management, client onboarding sequences, internal SOPs, and vendor coordination. 30 to 40 hours a week. Costs 1,200 to 3,500 USD per month. Gives back 15 to 25 hours and starts to take operational decision-making off your plate.

Role C: the Delivery Specialist. Hired only when the delivery work itself is the bottleneck. A junior designer if you run a design studio, a junior strategist if you run a strategy firm, a writer if you run a content business. Costs 1,500 to 5,000 USD per month. Only the right move if your audit shows you are spending most of your week on delivery tasks below your skill ceiling and admin is already handled.

For 8 out of 10 first hires we have helped scope, the right answer is Role A or B. Founders consistently overestimate how much delivery help they need and underestimate how much their admin load is killing them. Build operational ground under your feet first, then layer in delivery talent.

One mental model that helps: think of your time in three layers. The top layer is rare, high-value work only you can do (closing a 25,000 USD deal, designing a new offer, recording the keystone training that defines the firm). The middle layer is judgement work that needs business context but not your specific skills (drafting a proposal, triaging a client escalation, coordinating a launch). The bottom layer is operational throughput (inbox, scheduling, invoices, basic comms). A subtraction hire takes the bottom layer off your plate and starts learning the middle layer over time. A clone hire tries to share the top layer and almost always fails because the trust, context, and craft required take years to transfer. Sequence matters more than ambition.

Writing a Job Scorecard That Filters In, Not Just Out

Throw out the standard job description template. It does not work. It attracts everyone, signals nothing, and gives you no basis for performance reviews later. The 2026 standard is the scorecard, popularised by the topgrading methodology and refined by every serious hiring team since.

The Three Sections of a Scorecard

Section 1: Mission. One sentence on why this role exists. Not a list of duties, a reason for being. Example: "The Operations Coordinator exists to give the founder back 20 hours a week of high-leverage time by owning the operational and administrative engine of the business."

Section 2: Outcomes. 3 to 5 measurable results the hire is expected to deliver in their first 90 days and their first 12 months. Outcomes are results, not activities. "Reduce founder inbox time from 12 hours per week to under 2 hours per week within 60 days" is an outcome. "Manage the inbox" is not.

Section 3: Competencies. The 5 to 7 behaviours and skills required to hit those outcomes. Pick from a defined list (efficiency, attention to detail, calm under pressure, written communication, follow-through, judgement, learning agility) and tie each to the outcomes that need it.

That is the whole document. One page. Maximum 600 words. Better candidates read the whole thing and self-select in. Weaker candidates skim, panic at the specificity, and do not apply. This is exactly what you want.

The Anti-Scorecard Red Flags

Avoid these in your scorecard, no matter how often you see them in other job posts:

If you want to go deeper on the operating documents that pair with a hire, our guide on how to build SOPs that scale covers the playbooks your new hire will execute against. The scorecard tells them what success looks like, the SOPs tell them how to get there.

Where to Find Great Talent in 2026

The talent market in 2026 is more global, more remote, and more saturated than it has ever been. That sounds good until you post a role on a generic free board and wake up to 400 unqualified applications. Source the right pools, not the biggest pool.

The Channels That Actually Work

Ranked by quality per hour invested for first-time hires globally:

  1. Warm referrals from your network: single highest quality source. Send the scorecard to 20 trusted operators in your space and ask "do you know one person who matches this?" Expect 3 to 5 strong intros. Hire rate is often above 30 percent.
  2. LinkedIn searches with direct outreach: use the scorecard to define search filters (role keywords, location, seniority), then DM 50 to 100 fitting profiles with a personal message. Reply rate runs 15 to 25 percent.
  3. Niche job boards: for operations and EA roles, boards like Athena, Remote.co, We Work Remotely, and OnlineJobs.ph (for Philippines-based talent) outperform generic boards. Expect 20 to 60 qualified applicants per post.
  4. Community channels: niche Slack and Discord communities for your industry, alumni networks, and operator groups. Lower volume but very high signal.
  5. Former clients or collaborators: someone who has already worked with you knows the standard. Always a strong candidate if the role fits.

Where Not to Source First Hires

Aim for 30 to 60 qualified applicants for a single first hire. Fewer than that and you risk settling. More than that and your interview funnel cannot do its job.

Founder interviewing the first employee candidate

Screening: The 15-Minute Filter Call

Once applications are in, do not jump to a 60-minute interview. Run a 15-minute filter call first. Its only job is to eliminate the 60 to 70 percent of applicants who should not advance.

The Filter Call Structure

Five questions, in this exact order:

  1. "In one sentence, why did you apply for this role specifically?" Listens for whether they read the scorecard or mass-applied.
  2. "What is your current situation and what is your timeline to start?" Listens for availability and urgency.
  3. "What is your compensation expectation?" Surfaces budget mismatch before you waste hours on someone you cannot afford or who cannot afford to take the role.
  4. "Walk me through your last role in 90 seconds: what you were hired to do and what you delivered." Listens for clarity, ownership, and outcomes-language.
  5. "What questions do you have for me?" Listens for whether they have thought about the role at all.

15 minutes. Hard stop. Score immediately while it is fresh: advance, hold, or reject. Do not let politeness keep weak candidates in the funnel. Every hour you give a wrong fit is an hour stolen from a right fit.

Interviewing for Output, Not Charm

Most founders are bad interviewers. Not because they are dumb but because they default to instinct. Instinct rewards likeability, confidence, and people who look like them. None of those predict performance.

The Chronological Interview

Stage two of the funnel is a 60-minute chronological interview. Walk the candidate through their last three roles, in order, asking the same five questions for each:

  1. What were you hired to do?
  2. What did you actually accomplish?
  3. What were the biggest challenges and how did you handle them?
  4. What would your boss at that company say were your strengths and weaknesses?
  5. Why did you leave?

Listen for patterns across roles, not one anecdote. High performers describe outcomes ("I cut onboarding time from 14 days to 3"). Weak performers describe activities ("I was responsible for the onboarding process"). High performers tell honest weakness stories. Weak performers either deflect or give scripted answers ("I work too hard").

By the second role you usually see the pattern clearly. By the third you have your decision.

The Competency Interview

Stage three is a 60-minute behavioural interview against the scorecard competencies. For each competency, ask one "tell me about a time when..." prompt and probe with follow-ups. Examples:

Do not ask hypothetical questions ("what would you do if..."). Hypotheticals reward storytellers. Past behaviour predicts future behaviour. Stick to "tell me about a time when."

The Paid Trial Project

This is the stage most founders skip and then regret. The paid trial is the single highest predictor of hire success in our experience. It costs 200 to 800 USD and saves you 6 months of pain when it spares you a bad hire.

How to Run a Trial

Pick the top one or two candidates after stage three. Offer each a paid trial: 5 to 10 hours of real, representative work over 5 to 7 days, paid at their target hourly rate or a flat 200 to 500 USD project fee.

The trial brief should:

For an operations hire, that might be: "Audit our client onboarding process from this Loom video. Write a one-page summary of what is broken and a suggested 7-day fix sequence." For a delivery hire, give them a representative project file and ask for a single output.

What to Look For

Score the trial on three axes: work quality (is the output good?), communication (did they ask the right questions, send updates, hit the deadline?), and judgement (did they make sensible calls in the ambiguous area?).

One axis weak is recoverable. Two axes weak is a no-hire. The trial reveals the truth that interviews dress up.

Compensation: What to Pay Your First Hire

Underpaying your first hire is the most expensive way to save money. Attrition costs 1.5 to 3 times the salary of the role you are refilling, and that ignores the lost productivity, the client damage, and your own time. Pay market median, not the floor.

2026 Benchmarks for First Operations Hires

Approximate global ranges for full-time operations and assistant roles:

On top of the base, add 15 to 25 percent for taxes, statutory benefits, equipment, and software licences. A 2,000 USD per month hire is really a 2,300 to 2,500 USD per month true cost. Budget for the true number.

The Pay-Structure Choices

For a first hire, keep it simple. A clean monthly salary, paid on time, with one optional element: a small quarterly bonus (5 to 10 percent of monthly salary) tied to one or two specific outcomes from the scorecard. Equity, complex bonus structures, and commission plans rarely make sense for a first non-sales hire and add admin you do not need yet.

Most importantly: pay on the date you said you would pay, every single time. Late payment is the fastest way to destroy trust with a new hire and the cheapest way to keep it intact is to set a calendar reminder and never miss.

Small team onboarding a new hire on day one

Contracts, NDAs, and the Legal Basics

Boring but non-negotiable. Get these in place before day one.

The Minimum Legal Stack

Misclassification (treating a full-time employee as a contractor) is the single most common legal mistake at this stage. If the person works set hours, on your tools, under your direction, with no other clients, they are an employee in most jurisdictions regardless of what the paperwork says. Get this right or get a tax bill later.

The First 30 Days: Onboarding That Sticks

Most hires fail in the first 30 days because the founder treats day one as the finish line of hiring, not the starting line of onboarding. Day one is the most expensive day of the relationship. Treat it accordingly.

The Pre-Day-One Checklist

Before they log in for the first time, you should have:

If any of these are missing on day one, you are signalling that they are not important. The signal compounds.

Week One: Context Over Tasks

The first week is for context, not output. They need to understand the business model, the client base, the product or service, the tone of voice, the team norms, and the why behind decisions. Schedule:

Daily 15-minute check-ins for the first 10 working days. Then move to weekly 30-minute syncs. The frequency seems excessive until you see how much faster the hire becomes productive.

The 30-Day Outcome Bar

By day 30, the hire should have delivered at least two outcomes from the scorecard, even at small scale. If they have not, that is a data point. Address it now while you still have time to coach or part ways cleanly.

Building the Performance Rhythm

Onboarding ends, but the performance rhythm does not. Build a simple recurring cadence so feedback is constant and small instead of stored up and explosive.

The Three-Tier Cadence

This cadence sounds heavy on paper. In practice it takes about 4 hours a month, which is the cheapest insurance you can buy on a hire you are paying 12 to 30 times that monthly amount.

The Feedback Standard

Feedback should be specific, timely, and actionable. "Good job on the deck" is not feedback. "The structure of the second section made the recommendation 10 times clearer; do that every time" is feedback. So is "the late response to the client on Wednesday created confusion; let us agree a 4-hour response standard going forward."

The goal is small frequent corrections that prevent the annual blow-up review. Founders who get this right keep first hires for years. Founders who avoid hard conversations lose hires at month 7.

A note on praise. Most first-time managers undercount how much specific positive feedback the new hire needs in the first 90 days. A new hire is reading every signal looking for "am I doing this right?" Silence reads as disappointment. Make a deliberate habit of naming one concrete thing they did well in every weekly one-on-one, with as much specificity as the corrective feedback. "Good work" is noise. "The way you reorganised the client folder structure cut my Monday prep time in half" is a signal they will replicate. Specific praise is not soft management, it is performance management running in the positive direction.

Equally important: protect their time. A common founder trap is treating the new hire as an always-on Slack responder, which destroys the deep work they were hired for. Block 2 to 4 hours of uninterrupted time on their calendar daily for the actual scoreable work. Reply windows are fine. Permanent interruption is not.

Why First Hires Fail (and How to Avoid It)

After scoping or supporting more than 100 first-hire processes, the failure modes are remarkably consistent. They are also avoidable.

Failure Mode 1: Hiring Too Early

The founder hires before the revenue can carry the cost. Three months in, cash is tight, the hire feels like a luxury, and the founder either burns out trying to cover both salaries or makes a panicked release. Avoid by holding the two-signal rule strictly: stable revenue plus genuine overload, both for 90 days.

Failure Mode 2: Hiring a Clone

Founder hires a junior version of themselves and is then frustrated when the hire cannot do what took the founder a decade to learn. Avoid by hiring a subtraction role, not a replication role, for hire one.

Failure Mode 3: Vague Scope

"They can help with whatever I need." Translation: nothing measurable, no scorecard, no way to know if they are doing well or not. Avoid by writing the scorecard before posting the role and using it as the contract throughout the relationship.

Failure Mode 4: Skipping the Trial

Founder loved the interview vibe and skipped the paid trial. Six weeks in the work is below standard. Avoid by treating the paid trial as mandatory, not optional. Two hundred USD on a trial saves 20,000 USD on a failed hire.

Failure Mode 5: No Onboarding

Day one is a laptop and a welcome message. By week three the hire is lost, the founder is annoyed, and the relationship is already strained. Avoid by treating the first 30 days as a structured programme with daily check-ins and a clear outcome bar.

Failure Mode 6: Avoiding Hard Conversations

The hire is underperforming. The founder tells themselves "they just need more time." Six months later the gap is wider, the cost is higher, and the parting is messier. Avoid by setting clear 30, 60, 90-day checkpoints and acting on the data they produce.

Manager celebrating a successful first hire and team growth

A Composite Case: From Solo to Two in 60 Days

A consulting founder we worked with last year had built a service business to roughly 18,000 USD per month in stable revenue, working 65-hour weeks. She was convinced she needed a senior strategist to "share the strategy load." We ran the audit instead. Her week was 22 hours of strategy work, 19 hours of admin and inbox, 14 hours of client coordination, and 10 hours of "wasted" meetings.

The right hire was not a strategist. It was an operations coordinator. We wrote a scorecard targeting three outcomes: cut founder inbox time by 80 percent within 60 days, own client onboarding end-to-end within 30 days, and stand up two new SOPs per month. We sourced through LinkedIn and one warm referral, ran 4 filter calls, advanced 2 candidates through chronological and competency interviews, ran a paid trial with both, and hired the stronger candidate at 1,800 USD per month plus benefits. By day 35 the founder's working week was down to 47 hours. By day 90 revenue had grown to 23,000 USD per month because the founder was finally selling again. Annualised, the hire paid for herself within 4 months.

A second composite, in a different direction: a designer founder hired a junior designer too early, before admin help. Within 90 days the founder was doing all the same admin she had been doing before, plus reviewing and reworking the junior's design output, plus paying a salary. Revenue stayed flat. She let the junior go at month 5 and re-hired an EA instead. The EA hire freed 18 hours a week. Three months later the founder hired the junior designer back, this time successfully, because the admin foundation made the delivery hire viable.

Same founder, same role, completely different outcome based on sequence. The order matters more than the people.

The pattern repeats across industries. A boutique law firm hiring a junior associate before a paralegal. A wedding photographer hiring a second shooter before an editor. A growth agency hiring a strategist before an account coordinator. In every case, the founder added cost at the top of the stack while the operational floor stayed unfinished, and the math never worked. The fix is always the same: stop, audit, hire the missing operational layer, then revisit the more glamorous hire 60 days later when the foundations can actually carry the weight.

The 45-Day Hiring Timeline

Here is the full sequence compressed.

Days 1 to 5: Confirm readiness and define role. Run the audit, confirm both hiring signals, decide the subtraction role. Write the one-page scorecard.

Days 6 to 15: Source and screen. Open warm-referral channel, post to 2 to 3 niche boards, run targeted LinkedIn outreach. Receive applications, run 15-minute filter calls, score and shortlist.

Days 16 to 30: Deep interview and trial. Run 60-minute chronological interviews with top 4 to 6. Advance top 2 to competency interviews. Advance top 1 or 2 to paid trial. Score against scorecard, decide.

Days 31 to 38: Offer, contract, pre-start. Send offer with 30-60-90 plan attached. Sign contract, NDA, IP assignment. Order equipment, set up accounts, prepare SOPs.

Days 39 to 45: Onboarding week one. Run structured first week with daily 15-minute check-ins, context-heavy schedule, first low-stakes task ownership by Friday.

45 days from "I think I need to hire" to "they are productive." Faster than most founders believe possible if you respect the sequence and skip none of the steps.

The Mindset

Hiring your first employee is not a status milestone. It is an operating change. You stop being a soloist and start being a manager. That role has its own learning curve, its own emotional cost, and its own multiplier when done right.

The founders we see scale past their first hire are the ones who treat it as a process, not a personality test. They write the scorecard. They run the funnel. They pay the trial. They onboard for 30-day productivity. They build the cadence. They have the hard conversations early. None of it is glamorous. All of it compounds.

If you are getting close to your first hire and want a second pair of eyes on the scorecard, sourcing plan, or onboarding programme, our operations consulting team has scoped first hires across dozens of service businesses. The fastest way to avoid the common mistakes is to talk to someone who has watched them play out a hundred times.

For deeper context on what to do after the first hire lands, see our pieces on building your first team and growing revenue in a service business. The hire is step one. The system around the hire is what makes the revenue actually grow.

Two final framing reminders before you start. One: protect the relationship even when you have to end it. The way you let someone go in year one travels through your industry faster than the way you hired them. Pay severance you can afford, give honest references where they are warranted, and write a public-facing version of the parting that preserves their dignity. The talent market is small, your future hires will Google their predecessor, and the founders who handle exits with care attract the best next hires by reputation alone.

Two: enjoy the part that is genuinely joyful. The first time a new hire ships a piece of work you would have written yourself, you will feel the leverage that built every business larger than yours. Pause and notice it. That moment is the entire point of the playbook. Everything before it is the cost of admission, and everything after it is compounding.

Need a hiring sanity check?

Book a 30-minute First Hire Audit. We review your scorecard, sourcing plan, and onboarding programme before you spend a dollar on the wrong person.

Book Your Free Audit Browse the Operations Tools

Frequently Asked Questions

When should I hire my first employee?

Hire when you have at least 3 consecutive months of stable revenue that can cover the new salary plus a 3-month cash buffer, and when you are personally working over 55 hours a week on tasks that fall well below your unique skill ceiling. If both conditions are true, you are paying for the hire in burnout already. The right time is rarely when it feels comfortable; it is when not hiring is costing you more than hiring would.

Should I hire a virtual assistant first or a full-time employee?

For most service-business founders, the right first hire is a part-time virtual assistant or operations contractor working 15 to 25 hours a week. It costs 600 to 1,500 USD per month, removes the admin and inbox weight crushing your calendar, and lets you test whether you can actually delegate before committing to a full-time payroll. Jump straight to full-time only when you have a clearly defined role that needs 40 hours a week and revenue to match.

Contractor or full-time employee?

Start with contractors. They are faster to hire, easier to release, and shift the tax and benefits burden off you. Convert to full-time when the work is steady, the person has proven they perform, and you want to lock in their loyalty. Misclassification is a real legal risk in many jurisdictions, so if a contractor works full-time hours under your direction with your tools, convert them to employee status to stay clean.

How much should I pay my first hire?

Pay the local market median for the role, not the floor. Underpaying a first hire is a false economy because attrition costs more than the savings. For a generalist operations role globally, the range is usually 800 to 2,500 USD per month for remote talent in emerging markets and 3,500 to 6,000 USD for US or UK-based hires. Add 15 to 25 percent on top for benefits, taxes, equipment, and software seats.

How do I write a job description that attracts the right people?

Skip the generic job description format. Write a one-page scorecard with three sections: the mission (one sentence on why the role exists), the outcomes (3 to 5 measurable results expected in 90 days and in 12 months), and the competencies (the 5 to 7 behaviours and skills required). This format filters in the right candidates and filters out the resume-spammers. Most weak candidates will not finish reading it.

What questions should I ask in the interview?

Stop asking abstract questions like "what is your greatest weakness?". Ask the topgrading-style chronological interview: walk through their last 3 roles, asking what they were hired to do, what they actually accomplished, what their boss would say their strengths and weaknesses were, and why they left. Patterns of high performance or chronic underperformance show up in the second role. Add one behavioural prompt per competency, structured as "tell me about a time when...".

How do I know if my new hire is working out?

Set 30, 60, and 90-day outcome milestones in writing before day one. At each checkpoint, score the hire against the scorecard. By day 30 they should be productive on at least 2 outcomes. By 60 they should be running their core responsibilities with light oversight. By 90 they should be hitting target on the metrics that matter. If they are not at least 70 percent on track by day 60, they probably will not be at 100 percent by day 180. Act early.

What do I do if I hire the wrong person?

Act fast, act fair, act legally. Most founders wait 3 to 6 months too long, which damages the team and drains cash. The moment a hire is clearly not working after honest coaching, schedule the conversation, document the gap against the agreed scorecard, follow your local employment law for notice and severance, and part ways with dignity. Then write the post-mortem so the next hire is sharper. A bad hire is a system failure, not just a person failure.