This one starts with a person. A single support rep at PingPong, the cross-border payments company where I worked from 2021 to 2025, covered Trustpilot reviews, product questions, and a steady stream of tickets that belonged to other departments. I was Senior Digital Marketing Manager for growth and ops there, and the AI tooling was mine to build. Before I touched any software, that was the system: one inbox, one human, every category of work mixed together.
One inbox, four departments' mail
Before building anything, I spent a day sitting next to the rep watching the queue. Everything landed in one place. Urgent work sat next to repetitive work. Every switch cost her a beat and bought nothing. A review reply, a product question, and an accounting ticket all demanded the same attention, and only one of them belonged to support in the first place. The rep was triaging four departments' mail to answer one department's questions.
Watching her for a day ruled out the default answer. The job was to take the repetitive and misrouted work off one person's plate.
Build around the pain
I started from the pain the rep already had and wrapped the workflow around it. Nobody had to learn a new tool. I built two pieces that way. Both run on Zapier with a GPT-4o mini prompt step in the middle, no agent framework, no custom code. It is the same pattern behind the 200+ workflows I run in marketing ops.
The review responder
The build is unglamorous. Trustpilot's notification email hits a Zapier email parser, the review text gets posted into a dedicated Slack channel, and a GPT-4o mini step drafts two replies, one formal and one conversational. The rep edits whichever draft she likes better and posts it. Six steps, one afternoon to build.
The first version was embarrassing. The drafts read like a hotel apologizing. “we truly value your feedback” twice in three sentences. Nobody would believe a human wrote them, which defeats the point of replying to a review. The fix was a better prompt. I pasted ten of the rep's real past replies in as examples and capped the length. After that, the drafts sounded like us on a good day.
Once Trustpilot held up for a few weeks, I pointed the same workflow at G2 and Capterra notification emails. The same zap now covered three platforms.
The routing layer
Then I shipped a light classifier for inbound tickets. Every new ticket hits a webhook, GPT-4o mini reads the subject and body, and it returns one label. Only non-sensitive tickets go through, general inquiries, product questions and requests, with any PII stripped before the model sees a word, which is why this never needed a legal review. A CS person still approves every reply before it goes out, support, sales ops, accounting, or product. Zapier writes that label back as a tag, and a plain helpdesk rule assigns the ticket. The model read the ticket and picked a department. Nothing else touched the routing rule.
Anything the model wasn't confident about stayed in support, because a wrongly forwarded ticket costs more goodwill than one the rep forwards by hand. That rule earned its keep in the first week, when the classifier kept sending refund requests to accounting, who wanted invoices and only invoices. I added one line to the prompt, “refunds are support,” and accounting stopped getting them. Every drafted reply waited for a human before it posted.
What actually improved
Honest caveat: I never wired a dashboard to this project, so there's no headline percentage. Here's what visibly changed:
- Review replies stopped queueing, drafts were waiting in Slack before the rep even opened the review, so responses went out while the review was still fresh instead of whenever a quiet hour appeared
- Coverage reached three review platforms, Trustpilot plus G2 and Capterra, all answered by the same rep
- Misrouted work disappeared from the queue, sales ops, accounting, and product tickets went straight to their own queues without a bounce through the inbox
- We finally saw the workload, the tags finally showed what support spent time on, which no one had counted before because the counting itself was work
Do this before you open Zapier
Sit next to your support rep for a day before you build anything, and tag every ticket by hand: yours, sales ops', accounting's, product's. That tag list becomes your classifier's labels, the awkward edge cases (refunds!) become lines in your prompt, and you'll find out which part of the queue is even worth automating.
The zap took an afternoon. The day next to the rep told me which four labels to write and which tickets to leave alone.