Inbox Management
How to Auto-Label Gmail and Outlook Email with AI
Gmail and Outlook decide what's important for you and get it wrong constantly. Promotional mail lands in Primary marked important while the messages you wanted sit underneath it, and hand-written rules hold up until the next sender changes their address. This walkthrough starts in a real inbox, shows what Gmail flagged as important in the promotions tab (collaboration pitches and sales mail, none of it legitimate), and then shows what Carly's auto-label workflow did with the same messages, with no custom setup at all. The second half opens the workflow on the Carly site so you can see the machinery: every arriving email triggers a run, the content goes to an LLM with a prompt built from the default template plus your agent's context on who you are, and the labels come from the ones already in your account. You can preview real messages, run a batch to see how each gets categorized, and correct a bad call by saying so in plain language.
Step by step
- 01
See what your mail client got wrong
Look at what Gmail or Outlook has marked important, especially inside the promotions tab. Collaboration pitches and cold sales mail flagged as important is the normal case, and it's the reason built-in rules stop being worth maintaining.
- 02
Turn on the auto-label workflow with no setup
The template works out of the box. Switch it on and it starts labeling and routing incoming mail without you defining a single rule first.
- 03
Open the workflow and read its run history
On the Carly site, every arriving email triggers a run and every run is listed. This is where you check what the workflow actually did rather than guessing from the state of your inbox.
- 04
Understand how the classification is built
The email content goes to an LLM with a prompt assembled from the default template plus your Carly agent's context on who you are and what you work on. Labels are pulled from the ones already in your account, so there's nothing new to create.
- 05
Preview real emails and run a batch
Pull in actual messages and run them through to see how each one gets categorized. You're testing against your own mail, not a hypothetical example, before the workflow touches anything.
- 06
Correct a bad label in plain language
Say what should change ("anything about forestry should be action, not promotional") and test again immediately. You are never stuck with the classifier's opinion: adjust the prompt yourself, or email Carly and tell her what to fix.
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."