Blogs

By Spencer Kollas, Director of Delivery Services

Every morning when I wake up, I have a couple of emails from Twitter telling me that “Joe Somebody” has decided to follow me. Most of the time, I then click through to see what that particular person is really all about. Are they someone who I would like to follow? Do they have information that would be relevant to me or my role at StrongMail?

Many times they are people who I want to follow, but just as often they are simply trying to tell me how to get more Twitter followers, or sell me real estate, or anything else. If they are interesting, I will usually follow back, but for those who aren’t relevant to me, what are my choices? This is where I struggle.

If I look at it from a marketer's standpoint, the more people that follow you the better. As most marketers would agree, there are always a percentage of people who might not seem like a good fit right now but things could change and they could be a client in the future. It is all about reach right?

Another option is to block the seemingly irrelevant follower from seeing any more of your content. The final option is to simply do nothing, which could result in the person eventually un-following you once they realize you are not going to follow them back and increase their list of followers, and reach.

So let’s look at this from a couple different points of view - first from a marketer’s standpoint, and then from an overall social media ecosystem point of view.

Marketing 101 would dictate that the larger your list, the better—this means you are reaching more people and potentially selling more of your product. After a number of years, and a lot of arguing back and forth, many online marketing experts have realized that this is not true when it comes to email. With email it is all about the quality of your list, not the quantity. By focusing on the quality of your email list rather than the quantity, you are less likely to have a high complaint and unknown user rates, which are the two biggest factors that ISPs consider when filtering email.

In the early days of email marketing, many marketers assumed they could just adopt the philosophies of direct mail, and it seems like a similar thing is happening here with email best practices (e.g. list quality) getting pushed on social marketing. However, in the world of social media marketing, is there really any negative marketing impact from having some random person following me and receiving my message? Instead, I only see a positive impact from being able to reach more people and potentially sell more products. Of course, this is only true if the random followers aren't all obvious spammers and also happen to make up the majority of your list.

Now if we look at this issue of follower list quality from the social networks side, we might get a different perspective. I wouldn't be surprised if social networks would rather you block any user that is not a true fit from following you, even if you aren’t following them. From a capacity perspective, this would help users from seeing the dreaded Fail Whale, as the fewer followers you have, the fewer number of profiles the networks need to send updates to.

So as a good marketing citizen, what should you do? What are the best practices that you should be following when it comes to social media marketing? I believe that there can be a happy medium between what marketers want and what social networks can support. I believe that as a good marketer, you should accept those followers who are not overtly spammers or simply trying to get more followers. If a follower's past postings are all about getting more followers, than it's pretty safe to assume that they are not interested in your content – instead, they are most likely trying to prove to potential customers that they have a lot of followers and therefore must know what they are taking about.

The social media marketing space is definitely still evolving, and that means we have the opportunity to look at every aspect of it and determine for the best way to leverage this dynamic medium. As we start to analyze items such as list hygiene, we as an industry can began to build best practices for all to follow.

So please let me know your thoughts and comments.

update 2/5
From Yahoo
Our connectivity levels have been stable since Wednesday noon (PST). That said, we are still investigating reports of timeouts when trying to open a connection, as well as dropped connections during the DATA portion of the SMTP transaction.

update 1/27
It appears that once again Yahoo is having issues as they recently send out this message

Sorry for the late update, but as I'm sure most of the senders have noticed, we're again experiencing low connectivity on our MTAs, starting very early this morning.

We're working hard to restore the connectivity to normal levels, and I'll post an update as soon as I have some good news.

Please stay tuned

1/26

For those of you that might have seen some issues with delivery into Yahoo yesterday, those issues have officially been resolved. If you are still finding issues you may need to look into other reasons.

The official word from Yahoo yesterday was

If you're seeing some queuing in your outbounds today when sending to Y! users, we're currently experiencing low connectivity across our MTAs, which started this morning. We're working to get things back to normal.

The SMTP error message being generated by the issue should be:

"420 Resources unavailable temporarily. Please try later "

Our apologies for the inconvenience.

Just got the official word from my contact at GoodMail that they are offically live at Verizon--congrats to them

This notice was sent out to ReturnPath customers and I wanted to make sure everyone saw it.

New data sources have been integrated into Reputation Monitor, and the scoring models that drive the Sender Score have been updated. The new score was put into production yesterday. The results are impressive:

Sender Score coverage increased by 25%! Return Path is now able to generate scores for more IPs due to the increase in the number of unique IPs for which we have data. This means there will be fewer scenarios where an IP does not have a Sender Score.

The Sender Score is more robust than ever! By adding four new sources of data there's less chance that a Sender Score will fluctuate now that we have more insight into mailing behavior across more domains. Also, we've doubled the total amount of mail volume we have insight into which has a stabilizing effect. As long as your mailing behavior is consistent across data sources, your score shouldn't fluctuate.

Most Sender Scores remained the same! For the most part, Return Path customers have consistent mailing behavior across the domains they mail to so their Sender Score did not change. For others, we saw increases and decreases in their Sender Scores with the average swing being 20 points. If you happen to see this kind of point swing there are two explanations: 1) the increased volume data we now have insight into either had a positive or negative impact to your score OR 2) your mailing performance at the new data sources is very different and influenced your score.

Due to some recent changes it is important that all StrongDelivery Tools customers download their updated seedlist for proper results. Some domains were removed and consolidated.
Please contact your Account Manager if you have any questions.

As many of you might have recently heard AOL has a new error code as announced on one of their blogs.

ERROR [state=rcpt_to, code=550, text= 5.1.1 : Recipient address rejected: aol.com

For all StrongMail customers there is good news--there is nothing you need to worry about. The StrongMail system as been categorizing these as 2001 bounces in the system since the code came out.

if you have any questions please contact StrongMail support

This is a repost from the ReturnPath blog. I wanted to make sure all were aware of these changes/updates

By George Bilbrey President

As 2010 begins and many head to the gym to fulfill New Year's resolutions, here at Return Path we'll be getting our workout strengthening our data! As the world's most comprehensive email reputation data source (collecting data from over 150 million mailboxes worldwide), we are dedicated to ensuring our reputation scores are accurate, current and thorough.

We know the email universe relies on our Sender Score. For anyone who sends email, it's the foremost measure of email reputation - a direct reflection on sending practices based on universal reputation metrics: user complaints, spam trap hits, unknown user counts, and more. For those receiving mail, Sender Score can be used to inform inbound email handling and assist with the Herculean task of separating good email from spam.
These are the reasons why we'll be integrating new and improved data sources into our Sender Score calculation. Though we constantly tune the underlying Sender Score data to ensure accuracy, we have some deeper changes (adding and removing data as well as tuning) that we are making in early January that we wanted to give everyone a heads up on.
When these releases are complete, we anticipate both coverage (total unique IPs scored) and accuracy to improve. We do expect an initial, minor drop in coverage, however as the changes roll out coverage should ultimately increase (well beyond what it is today).
What does this mean for those sending mail? If you are affected by the temporary drop in coverage, when you check your score on senderscore.org (or in Reputation Monitor in our monitoring suite if you're a client), you may see a question mark instead of a numeric value. Again, this will only be temporary. Otherwise, you may notice your score (0-100) has risen or dropped from the previous average, as our analytical models are automatically adjusting to this new, broader perspective. But there's no cause for alarm. It'll smooth out soon.

Generally speaking, not having a score or having a minor shift in your score will NOT affect your email deliverability. The ISPs who use the Sender Score in their filtering decisions have been fully informed of these enhancements. We've also taken steps to ensure that our Return Path Certification program and Reputation Network Blacklist will ignore any inconsistencies in the Score.

Thank you for your patience and support as we roll out these great improvements to the Sender Score. As we continue to fortify our reputation data, Sender Score will be more robust than ever and will continue to be an increasingly crucial factor in inbox placement rates.

As some of you may know by now there is an issue with SpamAssassin scoring for all emails sent as of 2010. The bug causes an extra 3.4 point addition to your overall score. It is a particularly high scoring rule considering that the default threshold is 5.0.

A fix to this issue has been made available, but it may take some time for servers running SpamAssassin to make the necessary updates.Any emails sent as of January 1, 2010 to servers running SpamAssassin before the fix was made available, and to any servers running SpamAssassin that have not implemented the fix, will experience a higher than normal SpamAssassin score. This might impact delivery for messages where the score increase is significant enough to put the email's score above a filtering threshold.

If you are using a set of tools to test your messages before sending them out, such as StrongDelivery Tools, please make sure that your team is aware of this issue if it arises

I recently came across a great article on AOL's Blog by their Postmaster Christine Borgia. This article shows how the ISPs view permission as compared to some marketers. Make sure to read this article and post any comments you might have

Spamhaus has unveiled a new blacklist to combat a equally new spammer technique that uses static IP addresses and send in low volumes to avoid spam filters. This technique is called "snowshoe spam" because it's akin to spreading the load of sending across a large area.

If you're interested in using the new Spamhaus CSS (Composite Snow-Shoe) list to help detect snowshoe spam, you can find out more on their website: http://www.spamhaus.org/news.lasso?article=646