Leave a comment

Brian Balfour shares his growth meeting process from HubSpot/Sidekick.

  • SE

    Sean Ellis

    about 5 years ago #

    Good article on a very important subject. The weekly growth meeting is critical for holding the team accountable and managing a good testing/learning cadence. One thing that surprised me in the article is that Brian doesn't think this is a good time to be debating priorities. For our team we follow the lessons learned analysis with a discussion about the tests to run the next week. Generally each person on the growth team nominates a couple of growth ideas that have a clear hypothesis and have been score on their potential impact, confidence they will succeed and ease of testing. Each person does about a 30 second "pitch" on why they think their nomination is something that should be tested that week. After everyone has pitched the ideas, we select 4 - 5 experiments to run. By having everyone in the meeting when we select the experiments, people are able to voice their concerns if the experiment conflicts with another experiment or relies on a resource that is already stretched. We allocate a total of 15 minutes to this discussion and it seems to work pretty well.

    Besides this one point, I thought the article was really good. And even on this point, it's always useful to contrast different approaches and consider ways to improve your own.

    • BB

      Brian Balfour

      about 5 years ago #

      Hey just saw this. Sorry for the slow reply. Agree contrasting different approaches is good. Every team works a little differently.

      I think team size and team composition determines what works well.

      At some point there are too many people in the room where a priority discussion on a particular item isn't relevant to a number of others. Ends up causing more friction vs being productive. For example, a discussion around a virality experiment should really only involve a pm and maybe a couple engineers. But at the team meeting I might have paid acq members in the room where the conversation isn't particularly relevant. They can get the "whats going on" and identify any potential conflicts just by briefing over the experiment pipeline or experiment doc.

      The right amount of people for a priority discussion heavily varies. It is a hard balance. Involve too many people in the convo and you end up with a lot of wasted time. Don't involve enough people and you run into communication conflicts. Thats why I think it is better to have those outside of the team meeting where the number of people involved is set.

  • ES

    Edward Stephens

    about 5 years ago #

    Hallelujah, meetings without structure suck. This is a really good way to maximise the value of them. It's interesting to see them broken down as a sunk cost or ROI.

    The only issue we face is, tell another employee that 'it's reducing ROI' and you'll get a dim glance, namely because most of us end up working late to make up for the hours the meetings lose. So it normalises.

  • SM

    Shanelle Mullin

    about 5 years ago #

    Great read. Thanks for sharing this, Morgan!