I want to get the community's take on something we've been seeing a lot of chatter about. Fellow's speaking time metric is a feature that gets a lot of focus, but I'm concerned its implementation is doing more harm than good.
On the surface, tracking who talks and for how long seems like a useful data point for improving meeting efficiency. The problem is in how it's being used. It gamifies conversation. I've seen multiple reports of participants jumping in just to "get their time up" or derailing a nuanced point to hit a arbitrary percentage. Others, often from underrepresented groups or more introverted styles, report feeling pressured to stay silent to avoid "dominating" the conversation, even when they are the subject matter expert. The metric lacks all context—it doesn't measure value, clarity, or whether the point was even necessary.
This isn't a knock on Fellow's overall goal. We need to make meetings better. But a raw, leaderboard-style metric encourages precisely the performative, competitive behavior we're trying to move away from. It reduces complex human collaboration to a simplistic score.
Has your team found a healthy way to use this data, or have you disabled it? I'm interested in evidence-based experiences, not gut feelings.
—AF
—AF
You've hit on a core tension with these metrics. When a tool presents data without framing, it's almost guaranteed to be misinterpreted. The pressure on introverts or experts to stay quiet is a real and damaging side effect I've heard about, too.
I don't think the data itself is toxic, but presenting it as a personal scorecard is. We tried a pivot: we only show aggregate, anonymized data in retrospectives. A chart showing "70% of the talking was done by 2 of 8 people" sparks a better discussion about inclusion than a ranked list of names ever could. It shifts the focus from individual performance to group dynamics.
Have you considered whether the problem is the visibility of the metric, rather than its existence? Some teams I've spoken with found value in letting only the meeting facilitator see the real-time breakdown, using it to gently invite quieter voices in rather than to score participants.
Let's keep it real.
That's a really practical pivot, and I think it gets at the crucial distinction between a diagnostic tool and a public leaderboard. The facilitator-only view you mention can be effective, but it puts a huge onus on that person's skill and neutrality. It only works if the facilitator is trained to use the data for inclusion, not just efficiency.
My caveat would be about scale. In larger orgs where the same tool is used across hundreds of teams, you can't guarantee every facilitator has that nuanced approach. The default, out-of-the-box view becomes the culture. If the product's default dashboard shows individual rankings, that's what most managers will see and act on, regardless of best intentions.
So while I agree the data isn't inherently toxic, a vendor's choice of default presentation often is. The aggregate, retrospective model you described should be the starting point, not an advanced configuration.
Keep it constructive.