Summary
Drift Email classifies replies using machine learning techniques and a proprietary natural language engine that continually gets better over time as it sees more emails.
Reply Categories
Type |
Definition |
Bounce |
Auto-replies indicating the email could not be delivered. Sometimes a reason is given, such as a full inbox, or just that the address is inactive. These are known as soft bounces. |
Changed |
Auto-replies indicating the recipient has changed their email address - common examples of this are during a company rebrand or acquisition, or if they're changing their name after getting married. |
Human Direct |
Emails that we've determined are not automated, but are also not in response to something you sent - for example, someone sent a brand new email to one of your connected Siftrock addresses. |
Human Reply |
Replies that our system has determined is not from an automated source. These are the replies that you'll want to route to your team, to make sure someone gets in touch. |
Left Company |
Auto-replies indicating the recipient has left the company. |
Out of Office |
Auto-replies indicating the recipient is out of the office. |
Spam |
Direct emails (not in response to something you sent), that we've filtered as spam. |
Spam Shield |
Auto-replies from services such as Spam Arrest. Contacts using these services will not receive your emails without a manual validation step (usually clicking a link found in the auto-reply). |
System |
Auto-replies that are not a Left Company, Out of Office/Vacation, or Bounce. One example is a "thank you for your inquiry" email from a support or helpdesk email address. |
Unknown |
A catch-all for when we cannot place an email in any other category. The most common cause of this is emails in a language we don't support. Another common case is replies with no content in the email body. |
Unsubscribe |
Human replies that specifically include a request to be unsubscribed from future emails. Some users will ignore your unsubscribe links and instead respond to your emails asking to be removed from your list. |
Vacation |
Out of office emails that specifically indicate the recipient is on vacation or holiday. |
Accuracy of Classification
We strive to continually improve accuracy but it's important to understand that natural language processing is an imperfect science. Here is a focused example to help illustrate:
The word "retire" appears in various forms of automated reply. Examples we've seen include:
-
"I have retired from..."
-
"Bob Smith retired from our company last month..."
-
"I'm happy to announce my retirement..."
-
"I will retire from my position next month..."
Having seen so many of these types of emails the natural language engine recognizes forms of the word "retire" only appear in emails where someone has left (or is leaving) the company. Thus emails that look like this are classified as "Left Company".
This is a real email that came through for one of our customers:
-
"I have retired this email address and will be using this new one going forward..."
The human brain is capable of easily recognizing that this is actually a "Changed my email address" reply. However, the natural language engine doesn't have enough context or examples to understand this. Thus it gets incorrectly classified as "Left Company".
In processing more than 2 million replies we've only seen this language used once. Statistically we are unlikely to ever have enough examples to appropriately train the system to understand the difference.
Comments
0 comments
Article is closed for comments.