Global Email:FAQ

From Melissa Data Wiki
Revision as of 22:25, 23 February 2022 by Admin (talk | contribs)
Jump to navigation Jump to search

← Global Email

Global Email Navigation
Basics
Introduction
Licensing
Domain-Only Verification
Privacy and Global Email
Best Practices
FAQ
Service URLs
Input/Output
Request Fields
Response Fields
Examples
REST JSON
REST JSONP
REST XML
Batch XML
Batch JSON
Interpreting Results
Deliverability Confidence Score (Basic)
Result Codes (Advanced)
Global Email Result Codes
Sample Code



What Features does it have?


Available as

  • REST (Single, Batch)
  • XML (Single, Batch)
  • JSON (Single, Batch)
  • SSIS Component


Required Fields

Mandatory:

  • CustomerID - "Your License Key"
  • Email - "Email to validate"

Optional:

  • RecordID = "Internal Reference"
  • TransmissionReference = "Internal Reference"
  • Options = "Options to set"


Deliverability Confidence Score

Deliverability confidence score was added to Global Email V4. It is intended to be a quick and simple way to interpret and use our results. Please see Deliverability Confidence Score for more information.


Result Codes

Our result codes allow you to evaluate emails at a more granular level of detail. All the information represented by result codes is encapsulated in the Deliverability Confidence Score, but some may wish to dive deeper.

While we recommend most users look to the Deliverability Confidence Score, certain power users, such as those who need to pass on granular information to their own clients, will be able to use result codes to great effect.

Global Emails uses two types of result codes: status (ES) and error (EE). ES codes are returned if the email is valid, or was fixed and made valid, and they also give more information about the exact type of email (disposable, mobile, etc.).

Any EE code means the email is in some way invalid. Since we have removed the old result code ES02 from Version 4, invalid emails will only have EE codes and either ES21 or ES22.

For more information on our result codes and how to use them, please refer to Result Codes (Advanced).


Mailbox Validation

Why should client use email validation?

To reduce bounce rates (rejection) and avoid being labeled a spammer from mail servers. Once you are blacklisted, your email deliverability campaigns will suffer a lot and you have trouble sending out emails in the future (most mail servers subscribe to a form of spammers list). Integrators should consult with us to in order to understand what they need to look out for to avoid a bad campaign and why certain emails should be flagged/inspected. Even if you avoid spam traps, sending out too many invalid emails will cause the mail server to flag you as a potential spammer (which is why mailbox validation is important).

What is an accept all mail server?

Accept All means that the server stated all emails queries are considered active and valid regardless if they actually are. It is a setting by the responding mail server and we have no way to know if the email actually exists or not. We only know that the responding server says that the email exists regardless if the email actually exists on its server. It is up to you to decide whether or not to risk sending the emails out. Accept All emails generally have a higher bounce rate.

When should client use premium or express?

Premium

  • Should definitely be used for any email marketing campaign or where having the right email in a web form is completely necessary for it to function.

Express

  • Can be used for simple web forms where it would be nice to have a correct email but it isn't absolutely of utmost importance or if there are no consequences for a bad email in real-time. You can validate with premium mode in post batch processing.
What is the difference between mailbox and normal email validation?

Currently, our email object only performs normal email validation which is essentially just a simple spell check, mail server and domain check. Our Global Email Web Service does both mailbox and normal email validation. Mailbox validation means that the mailbox in the mail server is actually active and exists.

What is mailbox validation?

Mailbox validation is more comprehensive beyond than a MX Domain check. Mailbox validation actually asks the mail server to see if that mailbox account is active and if it exists. Otherwise we will only check to see if the mail server is active and if the mail server exists.

Why does an email have no mailbox validation?

Mailbox validation will only occur if the email passes our domain level and syntax level checks.

What does mailbox validation do?

It performs what is known as an SMTP check in the industry. Our service avoids being labeled as spammers for performing this check through proprietary technology. Your emailing reputation will be safe since we are the ones doing the mailbox checks for them. Of course, your emailing reputation will depend on our results when you start your email campaign on your end.

What does confidence level mean?

ES22 means that we performed a mailbox check on this email. When we say confidence level, we mean the certainty of the result code we are getting back. For example, ES01, ES22. Means that "with high certainty, this email is an valid email". These ES2x are the identifiers for our mailbox validation calls. Otherwise, we can't keep count of performed mailbox validation calls. So to summarize, the ES2X dictate how certain we are of the ES codes coming back.

  • ES23 is reserved for unicode detection for emails. We do not have UTF-8 Support yet for validating these emails. So for now we just identify these unicoded emails.


Verify Mailbox Option

What are the differences between the different modes (Express, Premium)?

Verify Mailbox Options


TimeToWait Option

What exactly does this do? What number should I be using?

TimeToWait is our newest option designed to give you more control over the process flow. It’s up to you and your team to decide if this option is necessary for you to use. The minimum value for this option is 5 seconds, while the maximum value is 45 seconds. The Default value is set to 25 seconds. Setting TimeToWait to a lower number will give you faster, but less accurate results. On the other hand, setting this value to a higher number will give you slower, but more accurate results. It’s up to you and your team to decide what you value more for each project – accuracy or speed. We recommend running a small test batch with a large variety of your emails at different TimeToWait settings. Every ‘ideal number’ will be different; and may change dependent on your email batch. The best way to find your ‘ideal number’ is to get a good spread across your email data base, and test that sample selection at different TimeToWait settings until you find the number that gives you the most accurate results in an acceptable time.

For more information, see ES03’s and TimeToWait.


ES03’s and TimeToWait

ES03’s can happen for a wide variety of reasons. One of them being the particular email timed out. We can reduce the number of ES03’s by increasing the length of time before the system will give up on an email and move on to the next. The ‘ideal’ number to use for TimeToWait will change primarily based on your email batch. For example, if you are validating email domains from a small to mid-sized business type you may want to consider increasing your TimeToWait. While on the other hand, if you are validating emails from large mailbox providers or companies you may consider decreasing your TimeToWait.

Recommended TimeToWait

Our research and analysis shows that the TimeToWait option results show varying levels of accuracy. Be warned that your results may vary with your own dataset. For example, the default time of 25 seconds you can expect receive around 7% unknowns. If you change this to a time out of 20 seconds, you can anticipate around 13% unknowns. At the end of the day you need to determine how accurate you need your results, and how long you are willing to wait to receive them. For example, batch processing will have a higher accuracy percentage (Less unknowns) from setting a high timeout value. Meanwhile, for real time web form type of validations where speed is of the essence, a lower timeout with logic to handle any potential unknowns may be preferred. We are constantly improving our algorithm in efforts to improve both accuracy and speed, but this is one method for us to give you more control over how our system will handle your data.


'Good' Email

What is a good email?

A good email will have an ES01 result code, consists of valid characters for the mailbox name, an '@' symbol, and a valid domain name. If mailbox validation option is set, a good email consist of valid characters, valid domain name, and active/existing mailbox account on that mail server (ES20 or ES21 or ES22).


Accept-all Emails

What is an accept-all email?

Accept-all (indicated by result code ES07) means that the email provider sets up their servers in a way that makes them relatively opaque to external email verifiers. A verifier will be able to perform a normal syntax check and determine if the domain is valid and active.

The email provider tells external verifiers that any email is valid and active, and even accepts all incoming emails as if the mailbox is valid and active, regardless of what may actually be the case. Because of this, sending emails to such addresses is essentially a gamble. We are constantly improving our accuracy with these emails.

Why does accept-all get its own result code?

Accept-all emails are generally problematic, but perhaps the most problematic for large email marketing campaigns. Without a way to identify accept-all emails, such a campaign could send a great many emails to accept-all addresses, many of which would bounce or get flagged as spam, thus decreasing ROI.

Please refer to Result Codes (Advanced) for more information on how to handle accept-all emails using Global Email.


'Bad' Email

What is a bad email?

A bad email will return an error code (EExx) which will tell you why this email is deemed to be a bad email.


'Unknown' Email

What is an unknown email?

An unknown email will return a ES03. With mailbox validation even if it's an unknown email, we want to let you know that it's with high certainty that it's an unknown email and that we did try request to the mail server for a mailbox validation call on this email.

Here are some examples of unknown emails, the mail server for some reason during the mail validation call, triggered our ES03 code either by responding with a non standardized response, or the mail server is increasingly aggressive with their protection to the point where IP address block might be implemented against our system if we continue or mail server tells us that we need to send a message to receive a bounce code (actually performing this could cause IP Block) or Grey Listing technology has been detected (we could not proceed). Another possible reason, is that the mail server the email resides on is timing out. There are too many to list and some are proprietary in nature. ES03 is meant to convey "at this moment in time, the email's status cannot be determined", email statuses are always in flux and it could change if you were to do another check.

Some mail servers are friendly and some are hostile. We will try to do mailbox validation on all mailboxes. Hostile mail servers have been developing countermeasure technologies to fight detection services like ours. It is a constant warfare for countermeasures and detection technologies. As they constantly develop their technologies...we constantly develop ours. So sometimes, for a certain email, mail server's algorithm at the mailbox validation level did a better job countering our algorithm's moves which results in a ES03.

Like a chess game, the results could be different if we rematch. It is up to you whether you want to run these unknown emails again...our algorithm could perform different actions that might result in a win and you receive a definite ES01 or lose again with ES03. We implemented a 3 day wait period minimum before retrying.

  • Important: Unknown emails can take as long as 13 seconds to verify. If your process is time sensitive and you are not interested in unknown emails, you can timeout early and mark the email as bad or unknown yourself.
  • Aol and Yahoo mail servers are known to take the longest time to validate.


What speeds can I expect from this web service?

The Global Email Web Service is split to two major option types: Express, Premium. These options drastically vary in performance due to the nature of the service. Express will usually return a higher throughput by returning results based on cached data while Premium involves a real time check to an MX Server to validate an email address. There are other factors that will impact performance so the throughputs may vary depending on the use case.

Factors impacting performance:

  • VerifyMailBox Options Express vs. Premium
    • Selecting Express or Premium options will be the greatest factor impacting performance
    • Premium will have slower speeds due to real time validation
  • Number of threads used for sending requests to the service:
    • We recommend 10 threads for processing
  • TimeToWait
    • Default setting is 25 seconds per request. Setting this lower will give you faster performance but may affect accuracy. Setting this higher will do the opposite; slower performance and possible higher accuracy
  • Data Quality
    • The quality of data sent to the service can affect the performance substantially
    • Premium real time checks will wait for a response from an MX server. If there are issues with a domain, TimeToWait may hit its maximum, increasing processing time
  • WhoIsLookup
    • Activating this option will allow domain information lookups which has a small performance impact

With these variations, you can expect these approximate throughput rates:

VerifyMailBox Option Throughput (records / hour)
Express 150,000 - 250,000
Premium 40,000 - 100,000