With a few other people, I run a Slack chat group for engineering managers. We’ve made some deliberate choices—for example, we don’t have a channel for #inclusion. Every so often this comes up, and we invoke our argument: Inclusion is central to good management, so all channels are inclusion channels.

We do, however, have a channel for #advice. This came out of trying to manage the way people are so ready to dispense advice in general, often lacking any context (which is why I personally try to avoid giving advice). We encourage people, in our code of conduct, to use the Gestalt principle (a way to communicate with people in a non-threatening way to support an environment where people can share and learn). But sometimes, people want advice. So we allow it.

It’s a fascinating channel, where people pose their problems and concerns. In each thread, respondents offer perspective, opinions, experience—often contradictory.

This is the space we create for the only truthful answer to any question about management: “It depends.”

I recently read the book How to Be a Person in the World, a great collection of advice columns by Heather Havrilesky, aka New York magazine’s Ask Polly. I’ve read her columns here and there, but to read a collection of them at once is to see both the vastness of human problems and also the way the same things reoccur.

It’s interesting how rarely an advice columnist offers the kind of advice that other, less-qualified people are so ready to give day to day, how little she offers that is definitive. The constants are broad and complex: learn to like yourself, fill your life with people you care about, go after the things you want. Often she reflects the situation back to the writer, mirroring the things that are important to the person, and shining a light on things they are trying to avoid.

This is the space she creates for “it depends.”

Despite my public disavowal of offering advice, people regularly come to me for it. I’ve noticed how rarely the thing someone says they want advice on is the actual problem they have. This is perhaps the problem with seeking out advice—we try to distill the problem down to something small and definitive, but what lies underneath is more existential. The small, definitive problems rarely eat away at us enough to need a second opinion.

Often we don’t want advice as much as we simply want someone to make the time to hear us out and validate the things we struggle with. We need the space for “it depends.”

I think it’s obvious to us that the big questions—Should I hire this person? Is this person still a fit on the team?—are not straightforward. But it’s amazing how much these rather straightforward questions can have beneath them. Seemingly mundane problems about code review or maintenance have turned out to really be about problems of fundamental team functioning.

Should I give this person feedback? Yes, of course, but… not if you’re uncritically relaying something petty or unfair. Perhaps not if they are struggling right now for other reasons; maybe you want to hold off and help them get to a better space first.

Should I give someone the stretch assignment they would be good at and are asking for? Probably, but… are you only giving stretch assignments to people who are asking for them? (There’s a better way to allocate those.)

Should I send the toxic person to another team? No, unless… well, what are you defining as toxic and how did it get to that point anyway?

I saw a text conversation recently, about onboarding into management while creating a new team, where the onboarder seemed to be looking for some definitive list, as though some amount of reading or some baseline process could set up a manager and a team. The first lesson for that manager, then, is that the common denominator is pretty low and there’s a lot to learn. What you start with is what your team seems to need, and what you know you don’t know. It depends. This is why we have #advice.

Cate Huston is an engineering manager at Automattic, where she has led the mobile and Jetpack teams.