Artificial Intelligence, Calendar, Meeting Scheduling, Product Updates, Productivity, Startup

Ridiculously Efficient Scheduling Software for Everyone

Artificial Intelligence, Calendar, Meeting Scheduling, Product Updates, Productivity, Startup

Today is the day.

I believe we’ve finally cracked the scheduling-for-everyone challenge. So we’re launching a retooled product that melds our AI assistants with intuitive visual UIs. AND we’re re-positioning the company, product and price. $8! (The wonders of software.)

We’ve also added a host of new features: a personalized calendar page, so you can send contacts a link to your availability, conference room booking and integrations with Slack, Zoom, Greenhouse, Salesforce, Intercom, and others. And we’ve introduced links inside the calendar events themselves, so you can reschedule meetings Amy or Andrew didn’t even set up to begin with, just for starters.

If anyone has been following along, you’ll know it hasn’t been easy to get here (it never is!). It’s worth sharing a bit about our journey (read: the blood, sweat and tears, the toil, the mistakes), since it was through these hard and exhilarating years of experimentation that we’ve arrived at the powerful and intuitive AI scheduling software we’re launching today.

So let’s start at the beginning. Our original vision was to offer AI scheduling assistants who behaved like humans for about the price of a cup of coffee. We named our AI assistants Amy and Andrew Ingram, and they operated exclusively over email at first. Like human assistants, they took over the entire scheduling task, and would only check back in if your guest didn’t respond or if they’d successfully scheduled the meeting, by sending you the calendar invite.

You can probably gather that in 2014, when we formed the company, I was super excited about the possibilities of the conversational user interface and this notion that a machine could not only be as good as a human assistant but superhuman—responding faster, scheduling instantly (because if all meeting participants use Amy and Andrew, they know everyone’s scheduling preferences and availability), and eventually, addressing participants in their native language.

We’ve certainly been right about a lot of things. We’re superfast now, responding to almost every scheduling request in under 5 minutes, and nearly instant when scheduling within a team or for a meeting in which everyone uses our software. And we’re not too far off from a multilingual AI scheduling assistant. But we were also wrong on several counts.

Faulkner once said “in writing, you must kill all your darlings,” and we’ve found it’s just the same in startups. One of the hardest but most important lessons learned is that not everyone wants to “talk” to an AI assistant. Sometimes, interacting over email is quick and convenient. Other times, guests want to simply click a button or select from a wide range of times, as they can now do via our visual UI and calendar pages.

Overall, we’ve greatly expanded the ways you can interact with our scheduling software because we learned (the hard way!) that people do not want to interrupt their workflow to get a meeting scheduled, so switching to email isn’t always the best choice.

Then too, we learned the limits of the assumptions we had baked into Amy and Andrew’s behavior. Because they rarely checked in with the host, Amy and Andrew would sometimes set up a meeting that wasn’t to the customers’ liking. For instance, if they were scheduling a meeting with four people, and three could make a date in the preferred time range but the fourth could not, they’d select the date that was convenient for all four rather than ask you whether you could make that fourth person optional.

Over the past months, we’ve added a series of decision points along the way, so that when a meeting might not work out the way you intended, Amy and Andrew ask for your input. In many ways this makes them more human like. For when in doubt, any good human assistant would ask his boss what to do next.

Finally, we haven’t always had a well tuned price. We started high—$39/month for Individuals and $59/month for Team. After all, we could always lower our price, we thought. What happened was pretty predictable in hindsight: with higher prices, our customers treated our product like a luxury—because when it’s expensive, it is :). And that meant losing out on the benefits of having your whole team use our software. At the new $8 price for individuals, and the new $12 price for teams, we’ve seen in testing that companies do indeed choose to add the whole organization. This makes me super excited about the future of scheduling.

So here we are getting back to original vision of charging about as much as a really nice latte. At this cost, it’s easy to sign up your whole team to schedule together.

We’re super excited about offering such powerful AI scheduling software today, at a price that’s in reach for pretty nearly every professional. Let me know what you think about this new turn in our startup journey. We’re always eager for feedback, and you can check a half dozen good examples on how the NEW x.ai works here.

If you like what you are seeing, do support us on Product Hunt here.

You can find our press release here.

And VentureBeat has a nice account of our relaunch here, for more context.

Cheers
d. 🙂