1

Looking for case study on effective 1-on-1s

Profile picture
Machine Learning Engineer at Startup9 months ago

I'm a new grad. I watched the 1-on-1 masterclass and i understand the theory -- be vulnerable, use it as a time to grow and get feedback on improvement, keep docs and an agenda

But I've only ever had 1-on-1s in my 1 internship where every 2 weeks I'd meet with the manager and it would be 5-10 mins. When I ask how can I improve, they would say yep everything is good, your perf is good, keep doing what you're doing and I'd say ok.

If I told my manager that hey I'm not happy with the current scope and I'm not learning enough they said rn keep doing what you're doing and later in the internship I'll give you other stuff and then I never got different work

In my new different company where I'm working in an early stage startup I want to set up a strong 1-on-1 culture with the founders who I am always working/talking consistently with but I have no idea what actually to talk about?

Can someone share case studies on what your recent 1-on-1s are like? what did you talk about -- breaking down in time components (e.g. first 5 mins small talk, then next 10 mins talking about X, then 10 mins on Y). If someone could share like a blue-print/actionable framework or guide that would be really helpful

111
4

Discussion

(4 comments)
  • 2
    Profile picture
    Tech Lead/Manager at Meta, Pinterest, Kosei
    9 months ago

    Two ways to force more reflection and discussion:

    1. Share your highlight and lowlight of the week
    2. Share an observation or problem and share tentative language about what you think about it

    (Taken from my YT video about 1:1s)

    I've found that sharing some level of emotion beyond just the work is the key to having deeper 1:1s.

    If you feel like you're not having a deep enough conversation with your manager, try recurring 1:1s with another teammate.

  • 1
    Profile picture
    Tech Lead @ Robinhood, Meta, Course Hero
    9 months ago

    I understand wanting more of a guide to follow, but I feel like that semi-defeats the purpose of a 1 on 1. These meetings are largely for relationship building, and that's best done when you're sort of flowing with the vibe and not following some rigid structure.

    I have had many 1 on 1s recently, and they all went largely like this:

    • [5 - 10 minutes] Ask them how they're doing and go through "small talk"
      • Bring up threads from prior 1 on 1s (this should be easy if you're taking meeting notes). An example would be "How did your trip to Paris go?".
    • [25 - 55 minutes] Go through the relevant discussion topics on-hand, which were ideally brought up and shared before the meeting.

    For that 2nd bucket, here's what I talked about with Rahul in our last 1 on 1:

    For your conversations with the founders, you can try talking about the following on top of small talk and tactical issues:

    • What their vision for the company is (both on the product and hiring side)
    • New project ideas (try proposing some)
    • How the company is doing (funding, sales/revenue, growth/retention)
  • 1
    Profile picture
    Tech Lead @ Robinhood, Meta, Course Hero
    9 months ago

    It's entirely possible that you don't even need a formal 1 on 1 since it looks like you talk to the founders quite regularly. The 1 on 1 meeting is more for bigger companies (>25 people) where each engineering manager has 10 reports and is swamped. It's especially relevant in Big Tech where managers spend so much of their time dealing with politics and bureaucracy that they can easily completely lose track of what a report is doing and the 1 on 1 meeting is a forced realignment mechanism.

    I think my challenge is how to come up with an effective meeting agenda?

    1. Figure out things you care about where you need more input.
    2. Figure out if the things from #1 apply to the relationship between you and the other person in the 1 on 1 specifically.
    3. If #1 and #2 are both true, talk about it in the 1 on 1.

    It's very possible that you don't have anything here as you're regularly covering important topics in ad-hoc 1 on 1 chats.

    If it's still not clear, here's another manufactured example: Let's say that one of the founders is specifically an industry-leading expert in LLM latency reduction. You also want to become a thought leader in this space, so you ask for this founder to mentor you in 1 on 1s and assign you resources to deepen learning. You can use the 1 on 1 to check in on this progress.

    How do I initiate depth in the 1-on-1s?

    It just sort of organically happens over time - It is hard to force. As you gain trust, the topics naturally mature. You each start giving more personal and more thorough answers to the "small talk" questions. More tactics here: https://www.jointaro.com/course/effective-communication-for-engineers-ae5a2/effective-communication-guide-part-3-building-relationships/

    Should the 1:1s be also a space to discuss higher level agenda and plans? like a 3-6 months roadmap and see how were tracking on it?

    This material seems better for a team-wide meeting actually.

  • 0
    Profile picture
    Machine Learning Engineer [OP]
    Startup
    9 months ago

    Thanks Alex and Rahul. This helps a ton. Rahul's video was very helpful

    I'm trying to reconcile how 1-on-1s should not be a scrum/but also related to the work im doing. Rahuls video helps in this regard

    I think my challenge is how to come up with an effective meeting agenda? I tend to discuss things with the founders as they come up so I feel like there's never a backlog of things to discuss?

    My 1-on-1s have always been super short -- 5-10 minutes. It's always a yep, we're on track or smth like that. How do I initiate depth in the 1-on-1s?

    Should the 1:1s be also a space to discuss higher level agenda and plans? like a 3-6 months roadmap and see how were tracking on it?

    So sorry if these sound kind of obtuse! Thanks so much for responding too. This actually helps a lot. Its hard to know how to have a good 1-on-1 if i dont have any point of reference