Hotmail Feedback Loop update

Our attempts at joining the hotmail feedback have been delayed due to lack of foresight, mainly on my behalf. we attempted to sign up for Pure, Loudvision and Betzoom all at once, this proved too complected for Microsoft who do not seem very organised.
The actual form itself is not clear either which is really inconvenient as we expected more clarity from Microsoft, especially in comparison to AOL.
I have since sent many emails trying to get a response but they are not in a rush.
We may have to re-apply three times but I am waiting for an email reply from them.
I have tried to phone them but after over 6 hours on the phone to America last week I have given up.

Masking - Custom Reply Email

Any customer can have a "Custom Reply Email" address in their account.

The "Custom Reply Email" is where admin will write in an email address to be the customers' From and Reply address when the email is sent. This overwrites the profile's From and Reply email address.

If they do not have masking this is called a spoof and that email address HAS to exist as the replies will go there. Additionally the customer will not get full reporting on the send.
This is usually used when a customer will not pay for or cannot get masking.

If they do have masking the address does not have to exist, in fact it should not exist.
This way the emails will go through us as usual. As it is masked what ever the address is all of the reporting will work. It is better to keep with the same sub domain though as this way the domain check will match the header records!

Optout List

If you want to opt out and large number of email addresses in PureRepsonse simply put them all into a csv. or get a list, then either upload them to Pure Lists or paste them into the upload new list text area and then you call the list

__OPTOUT__


as in : 'underscore-underscore-OPTOUT(call-caps)-underscore-underscore'

The system will then upload all of the addresses and add them to the suppression list for that profile.
you can then delete that list as the suppression has already been actioned!

If the customer decides not to delete the optout list. They can add more addresses to optout to it by appending it. This will also optout the address
 

Outlook 2007

There was a big scare yesterday when someone saw a blog article saying that Microsoft Office Outlook 2007 uses the HTML parsing and rendering engine from Microsoft Office Word 2007 to display HTML message bodies. The blog stated that this could take email design back 5 years.

This is the original blog post

We maybe not...in fact quite doubtfully! (If that's a word?)

This move from Microsoft will just force people to adhere the html and emailing conventions that the industry has been trying to spread.
There are still many confusions about xhtml and now html emails are taking off matter have been getting worse.
The way it's always been:
External Style sheets sre stricly a no go,
Absolutely no javascript - internal or external.
Style tags in the html is not the end of the word.
Absolutley no CSS positioning, everything should be in line. Which would mean using tables to structure the email.

The problem with this is that web page conventions in the last three years has pretty much abandoned tables and started using css positioning instead. Then the html email revolution starts and people making web pages are being asked to make emails cos they think it is now the same!!!
---
The defintative Email Marketing Newsletter: No Man's an Island Email Marketing Reports newsletter for 29/01/2007 - Now sponsored by Pure. Had this to say:

Microsoft themselves published an article describing the various things that Outlook 2007 will not be able to cope with. These include:

  • Forms
  • Background images
  • Animated GIFs
  • Flash
  • Float or position commands
  • Alt tags in images
In a nutshell, this places a lot of restrictions on email design. Needless to say, email designers are not happy.

The fires of protest first sparked into flame via an article in Sitepoint's newsletter. This led to a now-famous blog post by David Greiner, with the memorable title "Microsoft takes email design back 5 years."
---

Our most knowledgeable leader and Technical Director, Mr Mark Hla had this to say:
"As always with these things, it's best to read several articles to get a balanced view. Having read this article, the Microsoft supported CSS document and a pro-Microsoft partner website, I would put forward the following:

1. This change has been forced upon Microsoft due to the legal issues of making their other software dependant on Internet Explorer.
2. CSS are supported, but only a sub-set. We would generally recommend that clients keep their use of CSS to a minimum anyway, and the basic elements are still available for use.
3. This change is accompanied by a "this is what you'll see" type application that can be plugged into one of a handful of HTML editing applications. It may be beneficial for some clients who send to a large number of Outlook uses (i.e. B-2-B).

I would recommend allocating some of Ash's time to review the following article and provide advice for clients on what CSS elements they can and cannot use, and indeed if there is any change to our advice."

This is the Microsoft knowledge page

There is no panic about this, it will just force everyone to create best practice emails

Are Aquision Campaigns Dying

From my own experience, research and my interaction with Mimecast and Email Systems(British Academy), more and more commercial email providers are not just doing the usual html and content check, they are focussing more on the spam sender itself rather than the spam email.

They will have groups of safe-lists; global ones and personal ones. If the email coming in is not on any safe lists, emails will get put in a global or personal bulk folder or the email servers will ask the sending server to validate itself, like Mimecast.

I expect this to spread.
Hotmail will already put everything in the bulk folder that is not from a known contact.
More companies are using professional email providers rather than dealing with it in house I believe it's likely this will spread to internal email server software - ie exchange server etc or there will be pluggins.

Could this essentially kill off cold emails; Acquisition Campaigns etc.

Should we start getting our customers to put a now popular phrase in their emails:
"To ensure this email gets in your inbox please add this address to your email contacts list"
Obviously this is more newsletter related than anything else.

Should we therefore also try to get our customers to build there own lists. ie: listbuilder interface, web interaction, web focus community and forum actions etc.

What to do, what to do!

Greylisting

Spoke to a customer called Rapier, they use Mimecast for their emails and it is quite a new company. Their way of handling spam, as well as the usual key words and email content and html checking it checks the speed of the servers' send.
It doesn't accept new email addresses first time. What it will do is reject the first email and ask the sending server to send it again. It expects the sending server to re-send it in 15 seconds and then 15 minutes but ours waits 4 seconds and then 24 hours.
So anyone using this system will get delayed receipts for the first send.
So if you are testing campaigns a lot in a day emails keep getting delayed.

Our customer's head tekki said to me: "
Just going back to your server timings, I understand that your servers will respond very quickly on the second attempt but why do they wait so long before a third attempt? Most servers try every 15mins?"


This is the link to mimecast's official pdf

---

This is Spam Assasin's pov:
-

Greylisting

Instead of a 2nd fake MX you can use greylisting, which returns a temporary "Come Back Later" error for users currently not known. It has the advantage of helping you on the primary MX directly, and rejects about 60% of the connections here. This is because spammers only try to send once, and if there is an error, they drop it. Real mail servers retry later.

A disadvantage could be that e-mail is delayed a bit, as some users seem to demand that e-mail arrives immediately, and cannot wait some minutes. Either you can tell your users to wait, and save lots of SPAM, or don't use greylisting *g*.

Very good greylist server for postfix are: postgrey: [WWW] http://isg.ee.ethz.ch/tools/postgrey/ (uses DB style files, easy to configure, good support) sqlgrey: [WWW] http://sqlgrey.sourceforge.net/ (uses SQL databases)

exim: Marc Merlin wrote exim-sa, running SA during smtp time. With adaptive greylisting:

  • mails with a low spam score are accepted without delay

  • mails with an average spam score are greylisted, and only those are delayed

  • mails with high spam scores are rejected regardless (no greylisting)

[WWW] http://marc.merlins.org/linux/exim/sa.html

milter-greylist ([WWW] http://hcpnet.free.fr/milter-greylist/) is an excellent greylisting solution for Sendmail. I've been using it for almost two years now, and the difference in the amount of mail SpamAssassin has to worry about is amazing.



Hotmail services

Hotmail has a slightly different set-up to AOL.

Hotmail has a service called Smart Data Network services (SNDS).
This is an online statistics service where you register your sending IP addresses and you will have access to an online graphical reporting system, where you will be able to track the reputation of your sending servers in Hotmail. There is a graph for the all the server which have had any action for each day and a report for a specific IP address which covers 32days.Hotmail also has a feedback loop which we are on the verge of being part of, this should work in a very similar way to the AOL feedback loop.One other point is that Hotmail was initially involved in SenderID. This is where an TXT record on the sending server will contain an the SenderID. Hotmail will then check this ID against our sending servers to make sure that we are who we say we are etc.Recently Hotmail have increased this service but for a Price with their Sender Score program. This is a paid for service where members are almost guaranteed entry into the inbox. This is a service I am investigating along side AOLs Goodmail service as an optional extra to sell with new Unlimited packages.

Spam Filters


Where do they come from
Every recipient of emails today has a spam filter on their account.
This is either supplied by the ISP or it is a third party plug-in to an email client ie Outloook or Outlook express.
These filters will check every detail of the email before it gets to the inbox and decide whether or not it is bulk, spam or a proper email.
Many corporate filters have different levels of filtering which can be decided by the network administrator and/or the direct user.
Many of the higher levels will directly target and block newsletters.

Safe List
The best way to get in the inbox is to have your sending email address in the recipients' address book with the ISP and/or specified safe list, these are often the same thing. It is corporate filters which will have an additional safe list/
This means to the email handler that any email from this sender is expected as they are already a contact. So it essentially skips the deep filtering.

Pure Spam Tester
Our spam tester with Spam assassin picks up most of the issues and reports them but some of the explanations are not clear - I have yet to compile an explanation of these issues.
Normally it's OK to send when there are one of two low warnings. It is best to resolve the medium warnings as they are usually easily fixed. Try to avoid sending any email with any high warnings as these will get a very poor delivery rate but will still spend credits.

Plain Text Comparison
One of the most important things to remember is to make sure there is a plain text version and that the amount of text content in the html version is reflected in the plain text version.
Many filters will take both versions, strip the html out of the html version and then count the amount of characters in each and if they do not relate fairly closely they will not get in the inbox.
A good example of this is reflected in our own esales setup. The plain text version of a lot of the esales templates can only have the hyperlink to the html version because the content can be changed in the VE and will there for not be the same in the plain text version if there was any predefined content. This is often why quotes do not get in the inbox. There is a solution to this but that is for another post!

Max Lines Error
"Message lines too long - MAX 1000".
So either there can only be a maximum of 1000 characters before someone presses enter in an email. OR onlt 1000 lines of text? Can't find it on Google though!
This was illustrated in the LetsRecycle report on 1st December 2006 16.30.
This was a perticularly long email too!

Low Errors
Some corporate filters are now stopping all errors, even the low ones:
Technical Bit
Unfortunately, if there are any tables in the html, our editor automatically adds <tbody> tags inside the <table> tags. This is unfortunate but will always give a warning and will not get in a very small percentage of inboxes.
It is unlikely that we will be changing the VE in the near future but as we have to parse the html in order to add our mailid and tracking details to the message it may be possible to delete the <tbody> tags on the way through

Image size
Another classic warning is the size of images.
We all know that people like to click images as links instead of hyperlinks but if there is a large picture it will get a warning. Large Images should be broken down into little images - like a jigsaw.
A good example of the wrong way to do it is Eurodivision's 'Offers final'. - a giant picture with all of the offers on it. They have been informed that on the rare occasion that they do get in the inbox no-one will click through and it looks like spam!

AOL


We are on the Feedback loop with all of our sending servers and these are also on the whitelist.

The AOL Whitelist is a list of IP addresses which we register with AOL to say
"we are a responsible email marketing company, many different people are sending double opt-in best practice emails from our servers, please do not immediately black list our servers if you members mark any of them as spam, please let us know and will sort it out so our other customers don't suffer."

You can only achieve the AOL Whitelist by joining the AOL Feedback Loop (FBL) amd maintaining a good reputation.
This causes every email from our registered servers, that are marked as spam to be automatically forwarded to us.
Using our email tracking capabilities we can find out who marked the email, who sent the email and when.
This enables us to manually unsubscribe the complainer.
AOL states that marking an email as spam is as good as an unsubscribe.
Especially as once someone has marked email as spam any email from that sender will go into their spam folder and negatively effect our reputation.

http://postmaster.info.aol.com/

Subscribe link

Subscribe to Get In The Inbox by Email

Intro

Hello,
This blog exists for the Pure people to have access to everything I learn about Getting in the Inbox.

This includes:

  • Details of how ISPs deal with bulk email senders
    • legitimate and illegitimate senders;
  • Processes and natures of:
    • Whitelisting servers with ISPs
    • ISP Feedback loops
    • The nature of Spam Filters;
  • All the processes of how I have implemented and found the info;
  • Other Spam filtering organisations and how they work - or don't work;
  • and anything else relative;

There is a project on Sugar where most of this info is logged called "S&M Paid for Whitelisting" (don't forget to login to Sugar first)