Note: This post was written on 2/3/2021 and posted today.
The Messy Homes Tour was last night. I think it went well and multiple people wrote to tell me it was fun. Despite testing nearly everything ahead of time, we had some tech hiccups.
Tech issues in meetings seem to be commonplace, both online an IRL, so I want to examine this phenomena.
The MHT was, essentially, a Zoom meeting that was streamed to Facebook Live. To test this, my co-host Rebecca made a fake Facebook event in a Facebook group and streamed a test Zoom meeting to it. It worked. We had already tested all the various Zoom/Manycam/Snapchat configurations we would use as well. Every scenario we could think of was tested.
But when we tried it live it didn’t work.
Why?
Because we didn’t test her streaming on the exact same page it would be streaming to live. (The Big Weekend Facebook page). It took a while but eventually we found that we had some permissions issues.
But we shouldn’t have permission issues, I thought. Earlier in the week I realized she would need permission so I sent her an invite to be an admin of my Big Weekend Facebook page.
It turns out that you don’t just make someone an admin on Facebook. You send them an invite and they have to *accept* it.
The invitation was mixed in with the slew of notifications Rebecca sees every day. Most of which are not worth reading. She missed the invite.
I could have checked to see if she had accepted the invite, but since I’m used to things like Google docs, where you send an invite and that’s it, I didn’t know to check. Also, finding the spot to check is nonobvious.
So when we kicked off the show and she tried to stream on the live page, it sent me a notification (buried, of course) that I had to approve her stream. (Because as far as FB was concerned she was not an admin yet). Of course I didn’t know this was an issue let alone to look for how to fix it.
A typical reaction to this problem might be: “Well, now we’ve learned what to do so we know what to do next time.”
But I think a better response is to ask: “What’s the cost of doing an actual true live soup-to-nuts test?”
The “next time we’ll know” reaction might prevent future instances of this exact problem. But the help it provides doesn’t transfer from this scenario.
The proactive response is far more useful. It gives us a tool we can use elsewhere. Testing soup-to-nuts will find “unknown unknowns”.
We See This All The Time
Not testing the real deal costs us all the time when we take our setup out into the real world:
Their projector doesn’t work with my laptop.
There’s not enough outlets in this room.
We should have brought that adapter.
Sure, it’s potentially embarrassing to show a livestream test on your Facebook page. But how does that compare to having it not work at all when you do it for real? Weigh the pros and cons of a real-life test vs. controlled environment test.