Meetings Are Messy: Lessons From One Month of Meet-Ting
Spoiler: it’s less about software, more about sociology.
Hey all,
This is one of those fun articles I’ve been waiting to write. On one hand, it’s a set of insights about humans and meetings. On the other, it’s a record of Ting’s first month - combing through Slack, Jira, and notes to capture what we’ve learned so far.
I love working on meetings. There’s something special about tech that helps people connect. For years I felt that with social media (before it went off the rails).
At its best, tech accelerates human connection. And I think that’s one of the most exciting things AI can do for us: strip away the inefficient admin so we can focus on actually getting together.
Meetings are kind of magical when you think about it: job interviews, new partnerships, the first time you share an idea, the company all-hands, even just watching your favourite coworker present well to the group.
And the names we give them are just as weird and wonderful: 1:1s, huddles, standups, check-ins, debriefs, touchbases… the list goes on.
One line I heard years ago that hit me hard as a manager - and even harder now as a founder:
“People don’t want your time. They want your energy.”
Fifteen minutes of full attention is better than thirty distracted minutes on your phone. We’ve all been in meetings where the other person was mentally elsewhere. Ten minutes of focus would have been worth more than an hour of drift. That’s the original meeting sin: time wasted, momentum lost.
That’s why we love LLMs and their potential to help Ting become more emotionally intelligent - to schedule not just for time, but for energy. Imagine: if you have a board meeting on Friday, Ting will detect and hold Thursday clear for focus.
When tech can understand time and context, it can start protecting energy, not just minutes.
We’re not there yet, but that’s productivity on a deeper level - not just smarter task management, but something that actually helps humans work and connect better.
Where we are now
As of today, Ting has helped 250+ people get together, averaging 2.7 emails per booking. Pretty efficient, huh?
For those not obsessing over meetings 24/7 (must just be us), here’s some context on the problem we’re solving: over 40% of 1:1s with managers get rescheduled. Even the “stable” ones bounce weekly. And the average professional spends ~3 hours a week - 7.5% of total working time - just managing meetings.
One Reddit user suggested something we might actually build: before booking, Ting could spin up a quick chat to ask “Do you really need this meeting, or could async work better?” I love that. Features that protect energy and prioritise quality time fit perfectly into our emotionally intelligent scheduler vision.
And good reminder to any builders reading - Reddit and Product Hunt are goldmines. Thousands of brains ready to share feature ideas if you just listen.
The thing I’m most proud of so far? Ting’s human touch.
We’re seeing it show up in customer feedback - even ChatGPT, when asked what Meet-Ting is these days, highlights the more human approach as our USP from what’s on the web.
Recently I had two very different experiences. A founder replied to my genuine interest with a bare Calendly link. Efficient, but cold, inhuman. Contrast that with Ting stepping into an exchange for me - picking up on nuance in the thread and replying with warmth and compassion. Same efficiency, but personal and human. That’s another superpower: Ting keeps conversations warm even when your energy is low.
What building Ting has taught us about meetings
When you drop an AI into an important email thread, the bar is high - people deserve peace of mind the experience will be incredible.
Reliability and feedback are our single product focus right now. We’ve deliberately slowed invites, added more feedback loops, and are piloting new ways to capture raw, human insights directly from our community.
Here’s what else we’ve learned (and sometimes cringed at) in the past month:
Replying to the assistant. People love replying to Ting directly instead of the group, like they would with a human EA. Saves emails, but currently breaks the thread. We’re fixing it.
Time zones are chaos. Nobody says “Tuesday 2-3 PM GMT.” They say “after lunch,” “UK time,” or “when I land.” Thinking machines struggle here. Humans live here.
Fake calendar blocks. Lunch, fake focus time, “personal” holds. Everyone books over them. Ting’s learning not all blocks are sacred.
Wild inbox setups. People still use Hotmail. Yahoo too. Gmail dominates, Outlook is huge, and yes - Gmail through Outlook is a thing.
Meetings happen everywhere. WhatsApp, Telegram, LinkedIn, Slack, even quick calls. Then someone emails “let’s move the 2 PM we talked about,” and Ting has no history to reference.
Staying quiet matters. Sometimes Ting jumps in too early. Like a human EA, it needs rules of engagement.
Async is reality. Calendars change hourly. The system must always match real-time availability with what AI sees.
Multi-person, multi-zone = exponential mess. That’s why Ting-to-Ting feels magical: both sides reference calendars server-side, and it just books.
Humans write vaguely. “Let’s do it soon” vs hyper-specific “Tuesday 3:15-3:45 PM.” Pattern: senders (EAs) are precise, receivers are vague. A fascinating human tell.
Optional ≠ obvious. Most threads have 2-4 people. If someone writes “Find time for me and Ben,” Ting used to invite everyone. Took weeks of PRs to teach it the difference.
Timezone = identity. Defaulting to UTC was a mistake. Now if someone in LA says “2 PM my time,” Ting respects it. Timezones aren’t just technical - they’re social signals, and sometimes even power dynamics.
After Friday ≠ Friday. Humans speak in ranges, not timestamps. Early Ting panicked. We had to teach “after” as intention, not precision.
Debugging with tone. When Gemini failed, Ting used to vanish. Now it says: “Sorry, I couldn’t reschedule - want me to try again?” People forgive bugs. They don’t forgive robotic error messages.
Why this matters
Meetings are still where things (or ‘tings’) happen: deals close, relationships form, projects move forward.
The inbox remains the universal layer - it’s where our calendars live, and where confirmations always land.
But the way we book meetings is outdated, rigid, and painful. The old legacy Calendly-era game of forcing your way into someone’s time box and praying it doesn’t move doesn’t match reality. Schedules shift constantly. AI finally gives us a chance to do it better.
Ting is still early. Still messy at times. But one month into closed beta, we’re so motivated that if we can give people back even a slice of wasted time and energy, that’s a win worth chasing.
-Dan
Chief Ting