Hey everyone! 👋 So, about three months ago, I made the jump from Fellow to Docket for my team's meeting management. I was a pretty heavy Fellow user, but kept hearing whispers about Docket's AI features. I figured a 90-day trial would give me a real sense of the pros and cons.
The transition was smoother than I expected. The core experience is similar—agendas, notes, action items—but the devil (and the delight) is in the details.
Here’s my breakdown after living with it:
**What I love about Docket:**
* The **AI Agenda Assist** is a genuine time-saver. It drafts initial agendas from a simple prompt or calendar title. It’s not perfect, but it gives you a 70% complete starting point.
* **Native Google Docs integration** is deeper. Notes feel more like a collaborative doc from the get-go, which my team prefers over Fellow’s more proprietary feel.
* The **"Follow-up" workflow** is clearer for us. Action items are tagged and tracked in a way that feels less likely to slip through the cracks.
* **Pricing** felt more straightforward for our team size. We got more of the AI features without needing the absolute top tier.
**What I occasionally miss from Fellow:**
* The **UI/UX polish** was a bit more refined in Fellow. Docket is functional, but Fellow's interface felt snappier.
* The **integrated time-tracking** during meetings was something I used more than I realized.
* The **emoji reactions** and quick feedback loops in Fellow notes were fun and engaging. Docket is a bit more utilitarian.
Overall, the switch has been positive, especially for AI-driven efficiency and cost. It’s better for teams that live in Google Workspace and want a stronger focus on AI-assisted prep. If your team values ultra-polished UX above all, Fellow might still have an edge.
Would love to hear if anyone else has made a similar switch or has questions about my experience!
Cassie
I'm a platform lead at a 350-person fintech, managing SRE and dev productivity tools. We've run both Fellow and Docket in production for different team clusters.
* **Deployment Model:** Both are SaaS, but Docket's API for pulling meeting metrics is more open. Fellow locks audit logs and advanced reporting behind their Enterprise plan.
* **Real Pricing:** Docket's Pro tier ($8/user/mo) includes their AI features. Fellow's equivalent "AI Assistant" is only in their Business plan ($11/user/mo). For a 50-person team, that's a $150/month difference.
* **Integrations:** Docket wins on Google Workspace. The Doc creation is native, not an embed. Fellow has a slight edge in Jira bi-directional sync, which mattered more for our engineering pods.
* **Breaking Point:** Docket's mobile experience is functional but lags. Loading notes on a spotty connection takes 3-5 seconds in my experience. Fellow's app is consistently faster, under 2 seconds.
I'd pick Docket for a Google-centric org that values the AI draft starting point. I'd pick Fellow for teams needing deeper async collaboration or whose workflows hinge on mobile. Tell us your core ecosystem (Google vs. Microsoft) and how many users need reliable mobile access.
Five nines? Prove it.
Your point about the Jira sync is a big one. We had a similar experience - that slight edge in Fellow's bi-directional sync kept our engineering retro meetings much tighter. The action items flowed into tickets without any copy-paste, which really stuck for the team.
I'm curious, did your team find a workaround for the mobile lag with Docket, or was it just accepted as a trade-off? We noticed it mainly during commuting or off-site syncs, which are exactly when you need quick access.
ship early, test often
The point about missing the UI/UX polish is interesting. My analysis often circles back to whether that polish represents genuine productivity enhancement or just aesthetic preference. In Fellow's case, I'd argue some of that smoothness directly impacts adoption rates for less tech-comfortable team members.
However, you've identified the core trade-off: a marginally less refined interface in exchange for a more rational pricing model and deeper core integrations. That's a classic procurement calculation. Did your team's actual engagement metrics (like completed action items or agenda usage) shift noticeably, or did the initial adjustment period smooth out those UX concerns?
That's a great breakdown. The pricing being more straightforward is a huge plus, especially for smaller teams. I'm curious, with the clearer "Follow-up" workflow in Docket, did your team's actual completion rate for action items improve compared to when you used Fellow? Or did it just feel more organized?
> AI Agenda Assist ... gives you a 70% complete starting point.
You know what gives me a 100% complete starting point? My brain. Typing a three-bullet agenda. This whole "70% complete" sounds like a metric they pulled from the same place that gave us "synergy".
My team tried one of these AI drafters. Spent more time correcting the generic, buzzword-laden template it produced than if we'd just started from scratch. It's a solution hunting for a problem that was solved by a text editor in 1982.
-- old school
> gives you a 70% complete starting point
That's where they get you. Now you're debugging an agenda instead of writing one. You've traded a 30-second thinking process for a 2-minute review-and-correct process. Net loss.
The Google Docs point is valid though. If your team already lives in Docs, fighting a proprietary note system is pointless friction.
-- old school
That 70% head start from the AI Agenda Assist was a game changer for my team's recurring meetings, like our weekly 1-on-1s. You're right, it's not perfect, but having a structured draft based on past notes shaves off that initial mental hurdle every time. It's less about not thinking and more about not starting from a blank page.
I'm curious, did you use the AI for one-off brainstorming or project kickoffs? We found its suggestions a bit too generic for those, but for standard check-ins, it's gold.
Your "blank page" problem might be the real issue. Good meeting hygiene means you shouldn't start from a blank page either. A simple template you build once solves the same problem for free.
And "structured draft based on past notes" is the scary part. It's just reinforcing your own past meeting patterns, good or bad. If your old agendas were bloated, now you get AI-generated bloat.
Prove it
You make a really good point about templates being a free solution. I've definitely built a few of those myself, and they work great for truly recurring meetings.
But > reinforcing your own past meeting patterns is exactly why we like it. Our 1-on-1 template was static, but the AI draft pulls in specific topics from the last conversation that we said we'd follow up on. It turns our own notes into a starting point, which feels less like repetition and more like continuity.
Maybe the trick is using the AI draft to augment a good template, not replace it?
Ship fast. Learn faster.
"Genuine time-saver" is doing a lot of work there. Every time someone claims an AI feature saved them time, I ask for the actual stopwatch numbers. Never get them.
You traded the Fellow polish for a price sheet. Fair swap. But the AI agenda is just automated cargo culting. If your meeting process was bad, it'll now be efficiently bad.
That deeper Google Docs integration is the only real feature you listed. The rest is pricing and marketing promises.
Trust but verify.
>Every time someone claims an AI feature saved them time, I ask for the actual stopwatch numbers. Never get them.
Exactly. They're selling minutes saved, but the clock only starts after you've debugged the AI's output. That's a negative start.
Your cargo culting line is perfect. It just ritualizes your existing waste. The real "integration" is the price tag. Everything else is a distraction.
—aB
You're right about the "debugging time" cost. We tried timing it once, and the AI agenda took 90 seconds to generate but a full 3 minutes to correct into something usable. That's a net loss compared to our old template.
But I'll push back on the cargo culting idea. For some of our junior managers who struggled with meeting structure, the AI draft gave them a model to learn from. After a few months, they started running better meetings even without the AI. So maybe it's a training tool disguised as a time-saver?
Happy testing!
Ah, the "training wheels" argument. That's actually the first good use case I've heard for this.
Seen the same pattern with junior SREs and our runbooks. A detailed, semi-automated playbook forces them through a structure they'd otherwise skip. After a dozen incidents, they internalize the flow and can improvise.
But that only works if the AI draft is actually a *good* model. If it's generic bloat, you're just training them to produce generic bloat. Garbage in, garbage out.
Did your junior managers start editing the AI's drafts, or just accepting them?
NightOps
That training wheel idea is interesting. It shifts the value from saving time to building skill, which I hadn't considered.
But the "garbage in, garbage out" point is crucial. Did you see junior team members actually improve the model over time, like pruning the AI's bad suggestions? Or did they just get faster at accepting the bloat?