You will be able to build inbox rules and automations that label, forward and file email by type.
Arif's shared inbox at the renovation firm in Ubi gets around sixty emails on a busy day, by his own count. Supplier invoices, quotation requests from homeowners, site photos from the supervisors, condo management notices, delivery confirmations, newsletters, and the occasional complaint about noise from a neighbour. Everything lands in one list. He spends the first half hour of every morning reading subject lines and forwarding things to the right person, and on the days he is on leave, things get missed.
Sorting email is one of the most common automations people want, and one of the easiest to get wrong by reaching for AI too early. This lesson builds the sorting in layers: cheap, predictable rules first, AI only for what is left.
A large share of incoming email can be sorted by who sent it or what the subject says. Arif's suppliers always send from the same addresses. Condo management offices send from their own domains. Delivery confirmations come from a handful of logistics companies with predictable subject lines, and the newsletters all come from addresses he can list.
For mail like this, use the rules and filters built into your email app, as lesson 3.2, Use the automation you already pay for, described. Match on the sender's address or domain, or on words in the subject, and apply a label or move the message to a folder. These rules cost nothing extra, run instantly, never vary and are easy to check. If a rule misfires, you can see exactly why.
Arif wrote his rules by going through two weeks of email and grouping it. Every group with a clear sender or subject pattern got a rule. That covered most of the volume. What remained was mail from people he had never heard from before: homeowners asking about renovation, a neighbour complaining, a job applicant, a new supplier pitching.
The remainder is where an AI classification step earns its cost. General enquiries arrive from unknown senders with subjects like "Hi" or "Question" or no subject at all. No rule can sort them, but a model reading the body can.
Set it up the way lesson 5.2, Make the model pick from a fixed list, taught. Arif's labels for unsorted mail are quote request, complaint, job application, supplier pitch and unsure. Each has a one-line definition and an example. The answer is checked against the list before anything routes on it, and unsure goes to him.
Only send the unsorted mail to this step. Trigger the automation on a label such as "unsorted", which your email rules apply to anything they have not already sorted, or filter at the start of the flow so mail that already has a label stops there. Running every email through the AI step would cost more, and it would be slower and less predictable for mail that a rule already handles perfectly well.
Once a message has a label, from a rule or from the AI step, route it to whoever deals with it. The destination might be a person, by forwarding or assigning it in a shared inbox. It might be a folder. It might be a task list, by creating a task with a link to the email.
Arif's routes look like this. Quote requests create a task for the project manager with the homeowner's name, the email subject and a link to the message. Complaints go to the boss and to Arif, flagged. Job applications are labelled and moved to a hiring folder. Supplier pitches are labelled and archived. Unsure stays in the inbox with a label so Arif sees it.
Two habits keep this safe. First, never delete, edit or move the original email in a way that makes it hard to find. Label and file it, but keep it in the account, so anyone can go back to what the sender actually wrote. Second, the routing step only ever forwards a link or creates a task, and it never replies on the firm's behalf. If you want drafted replies, add them the way lesson 5.3, Draft replies a person approves before sending, described, with a person in charge of sending.
New routing rules make mistakes that only show up with real mail. A supplier changes the address they send from. A homeowner writes "invoice" in the subject of a question about a quote. A newsletter rule catches a real message from the same company.
For the first two weeks, check the sorted folders and labels every day. It takes a few minutes. Look for anything in the wrong place, and when you find one, fix the rule or the label definition that caused it, so the same kind of message lands correctly next time. Keep a short note of each misroute and its cause. After two weeks with few or no misroutes, drop to a weekly check.
Arif found two problems in his first week. A supplier who also did renovation work for the boss's relatives sent personal emails that his rule filed as invoices. And the AI step labelled a homeowner's angry follow-up about a delayed quote as a quote request rather than a complaint. He narrowed the supplier rule to emails with attachments and added an example of a frustrated follow-up to the complaint definition.
In the activity below you will write the routing rules for your own inbox, split them into simple rules and the cases that need AI classification, and set up the first three. Start with the rules that cover the most mail.
Write the routing rules for your inbox, splitting them into simple rules and AI-classified cases, and set up the first three.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).