A few months into our internal pilot, the Sidekick team noticed that most of what reached them was bug reports. Someone started a channel just for wins, a place where people could post what Sidekick had done for them. It filled up faster than anyone expected.
We read through it, along with the pilot feedback threads and our own team channels, and picked five stories that made us stop scrolling. Everything below happened at Zapier or with a Zapier customer, and where people measured something, we kept their numbers.
1. Every Enterprise deal gets its own breakdown the morning after it closes
For most of a year, our product marketing team posted a weekly analysis of why we won or lost a deal, along with a map of how the deal unfolded. A person built each one by hand whenever they remembered to.
Now a single Next Gen Zap handles it. When an Enterprise deal closes, the Zap pulls all the deal facts, runs an analysis, and saves the report to a Sidekick page. The next morning, the team has an answer to why we won, who was in the room, what put the deal at risk, and where the next expansion could come from.
The report costs about 55 cents, and it runs the moment a deal closes without anyone nudging it. By the time the builder shared it, the page held 58 breakdowns that the team could filter by theme, competitor, quarter, or deal size. It started as a weekly chore and became something the whole company can search.
2. A BDR turned out-of-office replies into a contact database in a week
Anyone who sends outbound email knows the pile of out-of-office replies that comes back. The only business development rep on one of our teams noticed that those replies are full of useful information, such as a note that someone has left a role and who replaced them, or a colleague to contact instead. The volume was too high to read by hand, so the information went to waste.
In about a week, working entirely inside Sidekick, that rep built a pipeline of Next Gen Zaps and Tables. One step catches the reply and separates bounces from human responses. Another uses AI to classify the reply and pull out names, titles, and contact details. The last step writes the results to five connected tables, with scheduled rollups for the week and month and a nightly check on retention.
The builder's summary was that the Next Gen Zaps "worked amazingly" and that they never had to leave Sidekick. It's an early version, and the next phase needs a review of how the data can be used, but the idea works: a pile of replies nobody had time to read became a record of who to contact next.
3. Nearly 200 calendar invites for a company-wide listening tour
When our CEO and CTO decided to hold 10-minute conversations with a large part of the Build organization, someone had to schedule them across time zones, vacations, and busy weeks.
A teammate had already built a Sidekick Slack app that books meetings for whoever you tag in a message. The executive assistant team took it further for the tour. The tool scheduled most of the calls, determined which days suited each zone, and flagged time zones, working hours, and people who were out of the office.
One of the people who used it said it made a real difference for two reasons. It removed the grind and mapped out the schedule in a way that would have taken far longer to work out by hand. Scheduling is the kind of work that fills a week and shows nothing for it afterward, so seeing it handled in one tool is a pleasant surprise.
4. A team went from about 25% to about 75% coverage in two weeks
Our observability team was building an AI feature of its own and moved the work onto Sidekick. After about two weeks, they went from roughly 25% to roughly 75% coverage on their target use cases, which they described as a huge step toward launching the feature.
They named what made the difference. Sidekick's code execution gave their assistant a way to do real work. File handling enabled them to render files for users. And the refinements the Sidekick team had already made to the agent's decision-making came along for free.
There were quieter benefits too. They now work within boundaries and evaluation practices they didn't have before, which keeps them from relearning lessons the Sidekick team already learned, and they get better knowledge sharing because they're no longer building alone.
5. Customers outside Zapier archived 800 Slack channels and sped up spreadsheet work
Some of the best stories come from outside Zapier. A customer using Sidekick told us they bulk archived more than 800 inactive Slack channels with it. They had spent months trying to work out how to do that through the Slack API.
Another user mostly uses Sidekick for spreadsheet work and scripts and told the team, without being asked, that it has been awesome a couple of times now, saving so much time.
Both of them are doing real work with Sidekick because it connects to the apps they already use through Zapier.
What these stories share
Look at the five, and a pattern shows up. Each one began with a chore that someone described in plain words. Some ended up running on their own, on a schedule or a trigger, connected to the tools people already use. And in most of them, the person who built it wasn't an engineer.
One salesperson who built six connected automations over three weeks gave advice that applies to all of them. Results become more useful when Sidekick sends them to you on a schedule, in a place you already look, so you don't have to remember to ask.
Try it on your own chore
Pick something you do every week that you'd happily hand off. Describe it to Sidekick the way you'd explain it to a new teammate, and ask it to build the automation. Once it works, ask it to run on a schedule and send the result to Slack. That's the whole recipe behind every story above. If you want more ideas first, see these 12 AI automation examples from other teams.
You can start at sidekick.zapier.com.
