Don't hire too quickly
If you haven't found PMF, it's almost always a bad idea to hire people unless doing so meaningfully increases the chance that you will find it.
This mistake can happen in many ways, but the most common is to hire a sales person because you can't sell yourself. When this happens, the issue is almost always that you haven't made something people really want.
About 9 months ago, I nearly made this mistake a third time. We attended office hours with our YC group partner Jared Friedman, and Jared said something that has stuck with me since:
"The early work of finding PMF is not parallelizable."
He's right. To increase the chance of finding PMF, you need see the problem you're tackling in its entirety and be able to connect all the dots. Here's what happens when you hire too early:
More abstraction layers. When you put people in between yourself and the user, information that reaches you gets diluted. Specific anecdotes get lost in summary notes. Emotions get lost in bullet points. Nuggets of information that initially seem irrelevant get skipped altogether. All of this is bad – you want to hear raw insights, be able to follow up on the spot, and build direct rapport with users so that you can follow up personally. It's crucial to hear the precise language people use to describe their problems because you will want to use that same language when you speak to your users.
Wasted time. In the early stages, you should be building product and talking to customers. Both (1) the process of hiring people (writing job descriptions, reaching out to candidates, scheduling, interviewing, offer negotiations) and (2) managing people (training, 1/1 meetings, etc.) are distractions that you will likely underestimate.
More excuses. As soon as someone else is responsible for a job, when things go wrong, it's no longer clear whether the person is bad at their job or your product is bad.¹ Because founders are so attached to their creations, your inclination will be to believe the former even if there is evidence to the contrary.
In addition to the above points which directly hinder progress towards PMF, there are other reasons to initially grow your team slowly:
You will hire better people if you're further along. A startup with PMF and a growth chart that goes up and to the right is much more compelling to candidates.
You maintain flexibility. The larger the team, the more momentum you have. It becomes much harder to pivot or make adjustments when the team is large. Most people don't handle uncertainty well.
¹ When you do hire people, this is why for some roles the standard advice is to hire in pairs. If you bring on two account executives, for example, you're more likely to understand how sellable your product is.