Every reply, sorted before you open the inbox
Sort campaign replies by what each one needs from you, not by when it arrived: problems and buyers first, questions next, housekeeping last, sales pitches to you at the bottom. Email Digit does that sorting for you as replies land, ranks them by an explainable priority score, and drafts a response, while anything risky waits for a human.
Why replies pile up
A campaign that goes to a few thousand people comes back as a mixed bag: a customer whose card was charged twice, a buyer asking for a call, an out-of-office, three unsubscribe requests and someone selling you SEO. They all look the same in a mailbox, sorted newest first. The urgent one sits under the auto-replies until somebody scrolls far enough, which is often too late.
How to triage replies by hand
You can get most of the benefit with a few folders and a rule for each:
- Today. Complaints, billing problems, refund requests, anything mentioning a lawyer or a chargeback, and people who said yes. These cost money or a customer if they wait.
- This week. Questions and “not right now, maybe later”. Answer them properly rather than quickly.
- Housekeeping. Out-of-office notes, wrong contacts and unsubscribe requests. Treat an unsubscribe sent as a reply exactly like a click on the unsubscribe link, and process it promptly.
- Not leads. Cold pitches from vendors, however enthusiastic. A reply is not interest if the person is selling to you.
The weakness of doing this by hand is that it depends on somebody reading every message, every day, and remembering who the sender is.
Where Email Digit reads replies from
You connect the mailbox your campaigns reply to: Gmail, Outlook, or any IMAP mailbox (presets cover providers such as Zoho, FastMail, iCloud and Yahoo, and you can enter a custom server). Email Digit checks the mailbox on a schedule, reads each new reply, and puts it in one inbox.
Six readings of every reply
Each reply is read on six separate questions rather than dropped into one bucket:
| Reading | What it answers |
|---|---|
| Party type | Who is writing: a lead, a customer, a vendor, a partner, press, an applicant, spam |
| Category | What it is about: sales, support, billing, a complaint, a refund, scheduling, a contract |
| Intent | What they want: interested, not now, a question, to unsubscribe, out of office, wrong person |
| Sentiment | Positive, neutral or negative |
| Urgency | Low, normal, high or critical |
| Intensity | How strongly they feel it, from 0 to 100 |
When a reading is not confident enough, the reply is marked for review instead of being tagged automatically, so a guess never quietly lands a contact in the wrong segment.
A priority you can explain
Every reply gets a priority score from 0 to 100. It is a fixed calculation over five inputs, so you can see why a reply ranks where it does: how urgent it is, how heated it is, your existing relationship with the sender (their engagement score), any money mentioned, and the sentiment. The same angry sentence ranks higher from a valuable customer than from a stranger.
Risk flags sit on top of the score: legal threat, chargeback risk, escalation, VIP, churn risk, payment overdue and unsubscribe requests. A serious flag sets a minimum priority, so a calm, politely worded legal threat cannot sink to the bottom of the list.
Context, so a pitch is not a lead
A message read on its own can fool anyone. “Love what you're doing, can we talk this week?” is a hot lead if you emailed that person, and a sales pitch if you never have. Email Digit checks the conversation first: whether you have emailed this person, whether they are a known contact, how many exchanges you have had, and the reply goal you set for your workspace. Someone selling to you is read as a vendor, and their priority is capped low unless the message carries a real risk.
Corrections stay in your workspace
If a reading is wrong, fix it in the inbox. You choose whether the correction applies to this message only, to everything from that sender, or to everything from that domain. From then on it is applied as a rule to future messages in your workspace. Nothing you correct is used to train anything shared with other customers.
Drafts that wait for you
Open a reply and pick the situation; the most specific playbook for it assembles a draft in your brand's voice. You never write a prompt. Where the draft needs a fact it does not have, such as an order number or a refund amount, it leaves a marked gap rather than inventing one, and the reply cannot be sent until you fill it.
Email Digit never sends a reply on its own. Every draft waits for a person, and a reply flagged for review, carrying a risk flag, or matched to an escalation playbook asks you to confirm again before it goes, however confident the draft looks.
What this changes
The folders above still describe the right order. The difference is that the sorting has already happened when you open the inbox, the reasons are visible, and the reply you would have found on Thursday is at the top on Monday. Read more about how replies are read, or see plans and pricing.