I ran my first 30-60-90 check-in with our first customer success hire the way I ran every other check-in: how's it going, any blockers, great, talk next month. Ninety days in, I still couldn't tell you whether the hire was working, and by the time I could, it would have cost a full year instead of a quarter.
Why the generic check-in fails a first CS hire
A new engineering hire has a team lead to compare notes with. Your first CS hire has nobody. There's no manager to flag when their instincts are wrong, no existing playbook to measure against, and no peer to say an account should have been escalated two weeks ago. If you used the hiring checklist to scope this role correctly, the 30-60-90 check-in is where you confirm it's actually playing out that way. Generic onboarding questions like "how are you settling in" don't produce a usable signal. Specific ones do.
The 30-day check-in: are they still asking questions?
Ask these three, close to word for word:
- "Walk me through the last account you flagged as at-risk. What told you it was at risk?"
- "What's one thing about how we do onboarding that doesn't match what you expected coming in?"
- "Which current customer would you bet is the least happy right now, and why?"
You're listening for specificity. A hire who's working cites a usage drop, a support ticket tone change, a missed onboarding milestone, something observed, not a vibe. A hire who answers in generalities at 30 days either hasn't looked closely yet or doesn't know what to look for. Either is fixable at 30 days. Neither is fixable if you wait until 90 to ask.
The 60-day check-in: process, or just activity?
By 60 days, activity is easy to fake and process is not. Ask:
- "Show me how you'd explain our health score to a new hire in this role."
- "What's the one account-level number you check every Monday morning?"
- "If you got hit by a bus tomorrow, what would I lose that isn't written down anywhere?"
The third question matters most. By day 60 you want two things to exist outside your hire's head: a documented health score and a repeatable onboarding motion. If those live only in conversation, you don't have a customer success function yet, you have one well-informed person, and you're one resignation away from starting over.
The 90-day check-in: the number that actually matters
By 90 days, stop asking about activity and ask about outcomes:
- "Of the accounts you own, how many have a documented next step in the next two weeks?"
- "Which accounts moved from red to green since day 30, and what did you specifically do?"
- "What's your honest renewal forecast for the accounts up next quarter, and where does it differ from mine?"
That last question is the real test. A working hire's forecast will disagree with yours in specific, defensible ways, because they're seeing signal you don't see from the founder's seat. A hire who's still guessing will either parrot your number back or hedge with "I'd need more time to say" on every account, not just the genuinely uncertain ones.
What to do when the answers are wrong
Don't wait for the next scheduled check-in to course-correct. If the 30-day answers are vague, pair them on two accounts for two weeks before the 60-day conversation, not as a punishment, as calibration. If the 60-day answers show no documented process, that's usually a scope problem, not a performance one: they may not know documentation is part of the job, so say so directly. If the 90-day forecast still matches yours exactly, with no independent read on any account, that's the clearest signal you'll get before a bad renewal quarter makes it undeniable, and it's cheaper to act on now than to relearn later.
Frequently asked questions
What if my CS hire gets defensive when I ask these questions?
Frame it before you start: "This isn't a performance review, it's how I calibrate with everyone in a new role." Defensiveness at 30 days is normal. Defensiveness that hasn't eased by 60 is a separate conversation worth having on its own.
Should I use the same script for a CS hire with prior experience?
Yes, unchanged. Experience changes how fast someone gets to good answers, not whether you need to ask the questions. Skipping the check-in because someone looks senior on paper is exactly how an experienced-sounding mis-hire survives past the point where it's cheap to fix.
What if I don't have a health score to compare their answer against yet?
Then the 60-day question is doing double duty: it's also telling you whether you need to define one together. That's a normal answer at this stage, not a red flag on its own.
How is this different from just reviewing their 30-60-90 day plan?
A written plan tells you what they intend to do. This script tells you what they actually noticed and did. Review the plan once, on day one. Run this script regardless of what the plan said would happen.
Run this script even if everything before day 30 looked fine. The point is a specific, falsifiable read on the hire while it's still cheap to act on, not a confirmation of what you already assumed. Reach out if you want a second opinion on what you're hearing back.