Almost half of new hires blow up inside 18 months, and for a solo agency owner one wrong $4,000-a-month person wipes out a quarter of margin. This is the do-it-this-week kit that picks the right first role, screens out the people who interview great and deliver garbage, and gets that hire owning real work by Day 10. The decision tree, the pay table, the job post, the paid test task, the interview script, and the two-week onboarding cadence, all fill-in-the-blank.
Your download should start automatically. If it didn't, click here to grab it. We sent a copy to your inbox too.
Want us to build the system that supports the hire? Apply →Most owners hire on a gut feeling, throw the new person in with no plan, and act shocked when it falls apart three weeks later. This is the opposite. A reversible-bet system that tells you which role frees you fastest, the real dollar ranges so you don't overpay, the one screen that predicts performance better than any interview, and a day-by-day onboarding plan so the hire is producing instead of waiting on you to explain it for the tenth time.
A one-page flowchart built on the only question that matters: what is the work only you can do, and what is eating the other 16 hours? It tells you in 60 seconds whether your first hire is a VA, a delivery specialist, or an ops person, so you stop agonizing and pick right.
Exact ranges, offshore and US, for every first-hire role. General VA at $3 to $7 an hour offshore versus $25 to $45 in the US. Specialists at $10 to $17 offshore, $50 to $100 US. Plus the monthly full-time math so you see the reversible-bet number before you commit a dollar.
Copy-paste, with the one-line outcome ("You will own ___ so I stop doing it"), 5 responsibilities, 3 non-negotiables, the Loom application requirement, and the buried instruction that filters out everyone who doesn't actually read the post before applying.
The single best screen there is, and almost nobody uses it. Work samples predict on-the-job performance at .54, far above interviews alone (Schmidt and Hunter, 1998). A real $50 to $150 paid task per role, plus the 1-to-5 rubric so you score it on instructions followed, speed, communication, and quality, not vibes.
Word for word, weighted toward coachability and attitude because attitude drives 89% of hiring failures (Leadership IQ), not skill. Includes the "hard feedback in the last 48 hours" question and a 1-to-5 scoring grid so two candidates are judged the same way.
Day by day, so the hire owns a real task by Day 10 instead of stalling. Day 1 access plus a Loom of you doing the work, Days 2 to 3 they document the SOP, Days 4 to 5 they do it and you review, Week 2 a daily 5-minute async check-in (done / stuck / next).
If any of these is the line you mutter at 9pm while you're still doing the work yourself, keep reading.
Leadership IQ tracked 20,000 new hires and found 46% of them failed inside 18 months. Only 11% of those failures were about skill. The other 89% came down to attitude: not coachable, low emotional intelligence, wrong temperament. So the owner who interviews for resume bullets and "can you do paid ads" is screening for the thing that almost never causes the blowup, and ignoring the thing that almost always does. Then they skip the paid test, throw the hire in with no SOP, and call it bad luck when it fails. It wasn't luck. It was a missing process. This playbook is that process.
Leadership IQ figures are from their hiring-failure study of 20,000 new hires across 312 organizations. The .54 validity coefficient is from Schmidt and Hunter's 1998 meta-analysis of selection methods in Psychological Bulletin. These are research findings, not a promise of your hiring outcome. Ranges are ranges.
Pick the right role tonight, post the fill-in job ad tomorrow, and have someone owning real work in two weeks.
Download starting now. If not, click here.
Want us to build the system that supports the hire? Apply →"I'd burned a US hire the year before, three weeks of training and they walked, so I was terrified to try again. This time I followed the playbook: decision tree said hire a VA, I paid four candidates $75 to clean a messy CRM export and draft three client emails, and one of them was obviously better than the rest. She was owning my inbox and reporting by the end of week two. For the first time in two years I took a weekend off and nothing broke."
Interviews reward whoever is most comfortable talking, which is exactly the trait you don't need. A paid test task rewards whoever actually does the work, on time, while following instructions. That's why the work sample predicts real performance at .54 and a chat does not. Paying $50 to $150 to watch four people do the actual job before you commit a $4,000 monthly seat isn't an expense. It's the cheapest insurance you'll ever buy against the 46% failure rate. The playbook hands you the exact task and the rubric for VA and specialist roles, so you're scoring real work instead of guessing.
That's the whole reason this is built as a reversible bet. An offshore general VA runs $3 to $7 an hour, roughly $600 to $1,200 a month full-time, not the $4,000-a-month risk you're picturing. And the paid test task means you see real work for $50 to $150 before you commit to anyone. You're spending a couple hundred dollars to avoid the multi-thousand-dollar blowup, not gambling a quarter of your margin on a hunch.
The decision tree settles it on page one. The short version: hire a generalist VA first if you're drowning in inbox, scheduling, reporting, and CRM cleanup, because it frees you fastest and is the cheapest reversible bet. Hire a delivery specialist if delivery is the bottleneck and you can already sell. Hire an ops or account manager only once you're past roughly $30K to $40K a month and managing more than two contractors. Most stuck owners need the VA.
Because you've been delegating without an SOP, which is just asking someone to read your mind. The onboarding cadence fixes that: Day 1 you record a 10-minute Loom doing the task, the hire documents it into a written SOP over the next two days, then does it while you review. By Day 10 they own it, and the SOP means you never explain it again. The work lives on paper, not in your head.
You can post the job ad within the hour of opening it. Everything is fill-in-the-blank: the job post has [BRACKETS] for your outcome and responsibilities, the test task is written out per role, the interview script is 8 word-for-word questions with a scoring grid, and the onboarding plan is dated day by day. Read it, run the decision tree, post the ad tonight, and have someone in a paid test by the weekend.
The reason you became the bottleneck is that you're the only one who can sell, so the second you hand off delivery you still own client acquisition, and the feast-or-famine months never stop. A VA cleaning your inbox doesn't change that. AgencyGod builds a client-acquisition system with you that books calls, makes sure they show, and closes them, then backs it with real money: if you don't hit the result, you get $10,000. We onboard only a handful of agencies a month.
Apply to AgencyGod $10,000 guarantee. Limited monthly intake.