DirectoryWorkflowsCase studiesPodcastsWho's Noah Take the stack assessment
← All case studies
Web-design agency

From 3.5 hours a day to 30 minutes on email

Inbox → classify, reply or draft, schedule follow-ups → Outlook + Google Sheets

Industry Web-design agency (Australia)
Stack n8n · Outlook (Graph API) · Google Sheets
Timeline Prototype in 1 day, live in 5
Compliance Australia Spam Act 2003
3.5h → 0.5h founder time per day
20–30/day emails auto-handled
Seconds lead reply time

Michael ran his entire agency out of a single Outlook inbox - no automation, no CRM - and lost about 3.5 hours a day to email. We built an n8n workflow that classifies every message, sends instant replies for low-risk cases or drafts approval-ready ones, schedules smart follow-ups for leads and invoices, and logs everything - with business-hours and compliance rules built in. Inbox time dropped to about 30 minutes a day.

Results at a glance

The problem

Michael managed the entire business through one Outlook inbox with no automation or CRM. He needed reliable categorization (leads, invoices, support, spam), personalized instant replies where safe and draft approvals for complex messages, automated lead and invoice follow-up scheduling, compliance with Australia's Spam Act, and all of it inside Outlook.

What was built

01

Classification

Sorts every message into lead, quote, invoice/payment, support, spam, or out-of-office.

02

Response routing

Sends instant replies for low-risk cases; generates a one-click-approvable draft for anything nuanced.

03

Lead follow-ups

Immediate first response, then nudges at +2 and +5 business days, halting the moment a reply or booking comes in.

04

Invoice follow-ups

Reminders pre-due, on the due date, then at +3 and +7 business days, pausing after two unanswered attempts.

05

Personalization

Pulls in business names, suburbs, requested services, and prior thread context, using local tone and spelling.

06

Operations

Sends only during AEST/AEDT business hours, with idempotent processing, retry-with-backoff, and failure notifications.

How the approach worked

Take a channel a person is drowning in, classify what's in it, and act on the safe cases automatically while queuing the tricky ones for a one-click approval. Compliance and business-hours rules are part of the build, not an afterthought.

The results

Manual inbox management was replaced with scheduled automation. As the founder put it: replies go out fast, follow-ups never slip, and he only reviews the tricky ones. Compliance was built in via the Spam Act 2003 - sender identification, opt-out, and DKIM/SPF/DMARC through Microsoft 365.

Replies go out fast, follow-ups never slip, and I only review the tricky ones.

Tools & technologies

n8nOutlook (Graph API)Google Sheets

Have a process like this slowing your business down? Tell me your stack and goals and I'll send back a build plan.

Take the stack assessment →