"Message to Nowhere": How not to send a letter to a non-existent addressee
By sending mailings to non-existent and one-time addresses, you risk falling under the "sanctions" of postal services. For example, the presence of invalid addresses in mailings can result in emails sent from your domain to the Spam folder or even blocking them.
And by sending letters to incorrect email addresses, you will cause a negative audience that has not even heard of you. As a result, not only your reputation will suffer, but also the effectiveness of your email campaigns.
Photo bnilsen CC / Flickr
For this reason, we do not recommend buying address databases, since it is impossible to verify in advance the authenticity of the addresses contained in them. The mailbox of the recipient of your mailing list can be blocked or deleted: on your own will or because of a long period of inactivity of the mailbox, so it is important to have mechanisms for checking the existence and correctness of addresses.
')
To begin, let's clarify what email addresses might look like. The left part of the address (up to the @ sign) can consist of upper and lower case letters of the English alphabet, numbers from 0 to 9, symbols! # $% & '* + - / =? ^ _ `{|} ~, Symbol“. ”( dots) if it is not the first and last character. Space and "(),:; <> @ [\] characters are allowed with restrictions .
The domain part of the address may consist of uppercase and lowercase letters of the English alphabet, numbers from 0 to 9 and a hyphen, or be an IP address in square brackets. Quite a few examples of unusual, but valid addresses can be found in the thematic article on Wikipedia.
Often, users deliberately or mistakenly provide an invalid email address. They can skip a character or, conversely, write an extra character, mix up characters in places, or accidentally press a neighboring key. Of course, each such typo makes email incorrect, but what are the chances that it will turn it into invalid?
As web developer and blogger David Gilbertson rightly notes (David Gilbertson), even if, due to some misunderstanding, you type in the address #!$%&'*+-/=?^_ERM :||@@example.com, he anyway, it is more likely to pass validation checks. Based on the arrangement of the keys on the keyboard, the probability that a user accidentally types an address that is truly invalid is relatively small (most of the letters on the computer’s keyboard are surrounded by the same letters — the chance that you miss and put an “invalid” character is estimated low).
So, incorrect addresses, as a rule, turn out to be valid - and no “basic check” can track it. David even built a statistical model (based on 117 million email addresses) and calculated the probability with which checking for address validation can "calculate" an incorrect email address. It is equal to 0.625 * 10 -38 %. By and large, checking the database of addresses for validity (from the point of view of possible misprints) does not make sense, which cannot be said about validation.
Photo by Jenni Douglas CC / Flickr
1. Strengthen site checking or use special services / API
As we already understood, users who entered an incorrect but valid address will never receive your messages later. At the same time, “incorrectness” can be different - the user can skip one or another letter, accidentally swap them, enter an extra (or incorrect) character.
For example, Pechkin has a Pechkin.Fix service that helps to calculate errors: checks email addresses for existence and corrects detected typos. We have to constantly work with a large number of postal addresses, so we regularly update and update the dictionaries of incorrect domains, corrections and typos. The service can be used online (just enter an email address for a single check), or you can use the API.
Of course, the service from Pechkin is not the only one with which you can solve the problem: both companies and independent developers are involved in checking incorrect addresses (here are some solutions from GitHub: 1 , 2 ).
2. Work on a double confirmation model
To form a “white” base of subscribers with a minimum number of random addresses, use double confirmation (Double Opt-In). With this system, the user enters his data on the site, after which he receives a letter with a link to confirm his consent to receive the newsletter.
Of course, such a subscription process takes more time, but it will save your email database from bots, non-existent and incorrect addresses. And to increase the percentage of activations of subscribers, the letter can be made more friendly and personal. For example, Pechkin's clients have the option of setting activation letters .
3. Filter one-time addresses
Some subscribers use special services to create temporary email addresses that are valid from 10 minutes to several hours. After this time, it is not possible to send a letter to such an address. Therefore, Pechkin, for example, offers his clients to use the filtering function of one-time email addresses, which will allow you not only to improve the numbers of discoveries and returns, but also to bypass spam filters.
What else can you read in our blog on the topic of improving mailings:
And by sending letters to incorrect email addresses, you will cause a negative audience that has not even heard of you. As a result, not only your reputation will suffer, but also the effectiveness of your email campaigns.
Photo bnilsen CC / Flickr
For this reason, we do not recommend buying address databases, since it is impossible to verify in advance the authenticity of the addresses contained in them. The mailbox of the recipient of your mailing list can be blocked or deleted: on your own will or because of a long period of inactivity of the mailbox, so it is important to have mechanisms for checking the existence and correctness of addresses.
')
Why and how the problem with incorrect addresses occurs
To begin, let's clarify what email addresses might look like. The left part of the address (up to the @ sign) can consist of upper and lower case letters of the English alphabet, numbers from 0 to 9, symbols! # $% & '* + - / =? ^ _ `{|} ~, Symbol“. ”( dots) if it is not the first and last character. Space and "(),:; <> @ [\] characters are allowed with restrictions .
The domain part of the address may consist of uppercase and lowercase letters of the English alphabet, numbers from 0 to 9 and a hyphen, or be an IP address in square brackets. Quite a few examples of unusual, but valid addresses can be found in the thematic article on Wikipedia.
Often, users deliberately or mistakenly provide an invalid email address. They can skip a character or, conversely, write an extra character, mix up characters in places, or accidentally press a neighboring key. Of course, each such typo makes email incorrect, but what are the chances that it will turn it into invalid?
As web developer and blogger David Gilbertson rightly notes (David Gilbertson), even if, due to some misunderstanding, you type in the address #!$%&'*+-/=?^_ERM :||@@example.com, he anyway, it is more likely to pass validation checks. Based on the arrangement of the keys on the keyboard, the probability that a user accidentally types an address that is truly invalid is relatively small (most of the letters on the computer’s keyboard are surrounded by the same letters — the chance that you miss and put an “invalid” character is estimated low).
How to solve the problem with incorrect email-addresses?
So, incorrect addresses, as a rule, turn out to be valid - and no “basic check” can track it. David even built a statistical model (based on 117 million email addresses) and calculated the probability with which checking for address validation can "calculate" an incorrect email address. It is equal to 0.625 * 10 -38 %. By and large, checking the database of addresses for validity (from the point of view of possible misprints) does not make sense, which cannot be said about validation.
Photo by Jenni Douglas CC / Flickr
1. Strengthen site checking or use special services / API
As we already understood, users who entered an incorrect but valid address will never receive your messages later. At the same time, “incorrectness” can be different - the user can skip one or another letter, accidentally swap them, enter an extra (or incorrect) character.
For example, Pechkin has a Pechkin.Fix service that helps to calculate errors: checks email addresses for existence and corrects detected typos. We have to constantly work with a large number of postal addresses, so we regularly update and update the dictionaries of incorrect domains, corrections and typos. The service can be used online (just enter an email address for a single check), or you can use the API.
Of course, the service from Pechkin is not the only one with which you can solve the problem: both companies and independent developers are involved in checking incorrect addresses (here are some solutions from GitHub: 1 , 2 ).
2. Work on a double confirmation model
To form a “white” base of subscribers with a minimum number of random addresses, use double confirmation (Double Opt-In). With this system, the user enters his data on the site, after which he receives a letter with a link to confirm his consent to receive the newsletter.
Of course, such a subscription process takes more time, but it will save your email database from bots, non-existent and incorrect addresses. And to increase the percentage of activations of subscribers, the letter can be made more friendly and personal. For example, Pechkin's clients have the option of setting activation letters .
3. Filter one-time addresses
Some subscribers use special services to create temporary email addresses that are valid from 10 minutes to several hours. After this time, it is not possible to send a letter to such an address. Therefore, Pechkin, for example, offers his clients to use the filtering function of one-time email addresses, which will allow you not only to improve the numbers of discoveries and returns, but also to bypass spam filters.
What else can you read in our blog on the topic of improving mailings:
Source: https://habr.com/ru/post/315004/
All Articles