Skip to content

How Calendly uses Zapier and AI to fight community spam

By Maranda Borchert · May 1, 2025

Ah, the spam of yesterday. It used to be easier to spot harmful messages—because so often, they were stuffed with bad grammar, janky formatting, or even outright gibberish.

But now that AI's in the picture, all that's changed. Today's AI-generated spam looks polished and personalized, with relevant context that makes it seem genuine. 

That created a unique challenge for Calendly, a company that thrives on meaningful community engagement. Jillian Bejtlich, who manages Calendly's online community for customers, partners, and employees—saw an opportunity to modernize their current spam filters. So she took to the drawing board.

Problem: AI spam comments that sneak past basic defenses

Calendly relied on industry-leading spam detection tools that proved less effective against deceiving AI-generated content. Too many false negatives allowed AI-generated spam to slip through, polluting discussions with malicious links. Meanwhile, the tools mistakenly flagged legitimate posts.

Having someone manually check every post wasn’t sustainable, as the community receives 600 posts per month. Jillian knew she needed a smart, automated system that could more accurately analyze content in real time.

Solution: An automated moderator that fights AI with AI

Calendly had already been using Zapier to automate Slack alerts whenever new posts appeared in their community. This helped their team respond to potential problems faster, but it didn’t prevent spam from appearing in the first place.

To improve their filtering, they needed a more advanced detection system—one that could intelligently analyze content, flag suspicious posts, and alert moderators before the problematic content lingered for too long.

After researching AI-powered solutions, Jillian chose Winston AI, a tool designed to detect AI-generated content and unnatural writing patterns. Since Winston AI integrates with Zapier, it was the perfect way to build an automated spam-fighting system.

Here's how the system works:

  1. When a post is submitted in Gainsight (Calendly's community host), Zapier triggers Winston AI to analyze the text for AI-generated patterns and unnatural phrasing.

  2. Winston AI scans for suspicious URLs, spammy intent, and unnatural writing, flagging anything questionable.

  3. Zapier sends an immediate alert to the moderation team via Slack.

  4. Moderators quickly review just the flagged posts, reducing their workload while ensuring genuine content isn’t mistakenly removed.

Results: A scalable system that protects the community

"We're about six weeks into our pilot so far," says Jillian, "and we're starting to see less spam in Calendly Community all around."

Spam levels have dropped about 70% because problematic content gets flagged before it gains traction. Moderators who once spent hours sifting through comments now need just minutes to review flagged items. And legitimate posts are less likely to be mistakenly removed.

"We're going to keep iterating, testing, and innovating," says Jillian. "But we're excited to have tried something new—and we're even more thrilled to see it having a positive impact."

About Calendly

  • Company Size: 400+

  • Industry: SaaS

  • Location: Remote

Get productivity tips delivered straight to your inbox

We’ll email you 1-3 times per week—and never share your information.

Related articles

Improve your productivity automatically. Use Zapier to get your apps working together.

A Zap with the trigger 'When I get a new lead from Facebook,' and the action 'Notify my team in Slack'