Your First CS Hire: What to Expect in the First 90 Days
The first ninety days don't produce a playbook. They produce a written record of what you've been carrying in your head.
The first ninety days don't produce a playbook. They produce a written record of what you've been carrying in your head.
You've made the call. The customer list has outgrown the founders and the account executives, renewals are starting to matter, and you're hiring your first person into customer success.
Most founders I talk to know what they want from that hire in a year. Fewer know what to expect in the first ninety days, and that's where the hire usually gets judged. So here's what I'd look for, written for the founder doing the hiring.
The short version: don't expect a playbook. Expect the customer knowledge that's been living in your head to end up on paper, and expect it to be a little uncomfortable to read.
Your first CS hire isn't there to make customers happy. Your customers are probably fairly happy already, because you've been personally answering their emails.
They're there to turn what you know about each customer into something the company knows. Right now, the reason a key account renews is that you remember what they bought it for. That doesn't scale past you, and it doesn't survive you being busy for a quarter.
Each stretch builds on the one before it. Skip the first and the second has nothing to stand on.
There's no platform purchase anywhere on this list, and that's on purpose.
The first month is mostly conversations. Your hire should be on calls with your top accounts, introduced by you, as the person who's now accountable for them.
The output I'd want by day 30 isn't a deck. It's two documents, each a page or two:
What you shouldn't expect yet is a health score, a tooling recommendation, or a renewal forecast with decimals in it. There isn't enough history for any of those to mean something. A hire who hands you one in week three is guessing, and you'll both believe the guess.
The second month is where the work becomes repeatable. Your hire picks one motion for the accounts that get real time (a check-in cadence, an objectives review, a quarterly value conversation) and runs it personally.
By hand matters. A motion nobody has run is one you can't hire against, can't teach, and can't tell is broken. The first twenty or thirty runs are how they find out which parts actually change anything.
This is also when the first renewal usually lands on their desk. Let them own it, with you in the room if it's a big one. How they prepare for it will tell you more than anything else in the ninety days.
A motion nobody has run is one you can't hire against, can't teach, and can't tell is broken.
By the third month, your hire knows the customer list better than anyone except you, and on some accounts better than you. The output here is a short review, and it should be a little uncomfortable.
It names the accounts at risk and says why. It says which customers you've been informally over-serving, and what that costs. It says what the motion revealed that nobody expected. And it describes the second hire, based on the work that actually exists rather than a job description from a company forty times your size.
If the review is all good news, I'd gently push on it. Ninety days in, a customer list that's been held together by founders always has something in it that nobody wanted to say out loud.
This part gets skipped, and it's usually why first CS hires don't work out. The hire can't do any of the above without four things from you.
Hand over the relationships in public. Introduce them to each key account as the person who's now accountable, and then stop answering those emails yourself. If customers can still reach you directly, they will, and your hire becomes a note-taker on their own book.
Agree the losses. When the allocation says some accounts won't get scheduled time, that's a company decision, and you have to make it with them. Otherwise they'll be the only person holding it when one of those accounts leaves noisily.
Keep them out of the support queue. At this size there's always a pull to make the first CS hire the person who answers tickets. Do that and you've hired a support person with a better title, and the renewals still sit with you.
Give them a line into product. Your hire is about to hear more about what customers need than anyone in the company. If there's nowhere for that to go, they'll stop collecting it.
There's a real pull to buy tooling for the new function in month one, and most of it is built for a team of twenty, not a team of one.
Where I've seen AI actually help this early is the boring part: capturing what a customer said they're solving for, at the moment they said it, so nobody rewrites the objectives later after seeing the usage numbers. That record is what every renewal conversation for the next three years gets built on. Get it right while there's one person doing it, and it's much easier to scale.
Anything that scores, forecasts, or decides can wait until there's a year of outcomes to check it against.
If you try one thing from this, do it before your hire starts. Write down your ten most important accounts and, next to each one, what that customer told you they were trying to solve. You'll probably find a few you can't fill in, and that's the most useful thing to hand your new hire on day one.
If you're about to make this hire, or you're a few weeks in and want to talk through how it's going, reach out. Happy to help.