Definition of Hard/Soft Bounceback

[rchen]

I haven't used your product yet, but I'm really interested in using it sort out my mailing lists. We send out literally thousands of emails weekly, and are not able to process the invalid email addresses.

My question is:

How does Email Verifier categorize a bounceback as Hard or Soft? What is the reasoning behind each reason? Are there bounceback reasons that exists as only as "maybe"? Can a "Soft" email bounceback becomes "Hard" or "Valid"?

Essentially, do you utilize a type of list that separates bounce back reasons into three categories - Valid, Hard, Soft?

Thanks!

 

[rchen]

I just found the page definition of hard/soft, but do you have the list of bounceback reasons that Email Verifier uses to categorize by Hard/Soft, like the sample you provided at:

http://pobox.com/bounce-common.mhtml

 

[stanbusk]

That list is not public, it is a set of custom Regex statements. We simply follow RFC specifications. Look at Appendix E of RFC 821 for example.

 

[rchen]

Thanks. I'm just curious as to how "good" the results are, in regards to the types of bounce back reasons would be categorized. I suppose there is no way to custom-define which reasons should be categorized to which grouping, is there?

 

[rchen]

I just found the page definition of hard/soft, but do you have the list of bounceback reasons that Email Verifier uses to categorize by Hard/Soft, like the sample you provided at:

http://pobox.com/bounce-common.mhtml

 

[stanbusk]

No, there is no such list right now. We simply followed Appendix E of RFC 821. Then we tested the application with hundreds of thousands of addresses to analyze responses. We found out that a same response codes are ambiguous and can have different interpretations so we added more filters. In other words we can threat error 503 in 20 different ways based on the comment following that code.

eMail Verifier is rather prudent. It is the reason why we have the 'Look good' and the 'Look bad' categories. Sometimes it is not easy to categorize a response. The server postmaster can changes both the response codes and descriptions freely.

The idea is to mark as 'Bad' an address when the server response is clear and unambiguous. Consider eMail Verifier as a tool for tracking bad e-mail addresses instead of validating good ones.

 

[rchen]

Thanks - I guess due to the lack of standard SMTP protocol, there is no way "surefire" way to categorize these emails.

I just tried the trial version, and pasted a list of about 60 emails. On every single one of them, except for @yahoo.com, it returned a "No response" under the Server column, and a "Unable to test" in the Result column. Looking up the console, I see a series of:

[00] 12/16 21:08:17 > MX Lookup <[email protected]>
[00] 12/16 21:08:17 MX Result <[email protected]> -> [[Non-existent domain
]]

I am using 3.0.2 (5) in Windows XP SP2 with 1 GIG Ram, P4 2.4 Ghz, if that helps.

 

[stanbusk]

You should check whether port 25 is not blocked by your ISP. You can easily find out following the steps listed here: How to find out whether port 25 is blocked on my account?. Also you have to make sure your system is set properly with your DNS server address(es).

 

[rchen]

Thanks, my port is blocked. OK will look for additional ways of getting out, then.

Thanks!