Stop Sending Junk E-Mail

Stop sending junk e-mail is good advice for anyone, but you may inadvertently be sending junk e-mail when you don’t intend to.  If you find that surprising, read on because you’re in for some major surprises.

None of us likes junk e-mail, which is e-mail that we have not requested.  If you wish you can add a spam filter to your e-mail service and weed out what you do not wish to receive.  For example I use K-9, which learns as you teach it what you regard as spam and what you believe is acceptable.  I find its performance excellent and it does a fine job in filtering out over 40% of my incoming e-mail, which is spam.

What you may not realize is that your Internet Service Provider (ISP) has already filtered out some other spam e-mail that you never saw.  Indeed any e-mail message passes across a number of ISPs in travelling from the sender to the receiver.  That means that other ISPs on the route may also have cut out some spam e-mail by their definition of spam.  They do this to help their clients but also to act as a deterrent to others who may wish to tie up their bandwidth with automatically generated floods of spam messages. 

What exactly is their definition of spam?  In fact the answer will always be complex and it is most unlikely that you can get a clear answer.  What follows is based on some extensive work done over 6 months ago.  In the interim, despite further analyses, the situation is no clearer.

Your IP Reputation

The first and probably most important factor in determining what constitutes spam is the Internet Protocol (IP) of the originator of the e-mail.  There are a number of services that evaluate the reputation of the originating IP.  One such is If you evaluate my own IP,, you will see it has a Poor reputation. 

You can find out your own IP by using a service such as The reputation of your own IP might be rated as either Good, Neutral or Poor.  Even with a Good reputation, there is no guarantee that any given ISP will not reject your e-mail message as spam or junk e-mail but the odds are low that this would happen.

In the worst case, if you have indulged in deliberate e-mail spamming then it is possible that the IP address of your mail server is currently listed on the SpamCop blocklist.  This is a sure-fire guarantee that your e-mail messages will likely not get through.  In this case, you will need to read the SpamCop FAQ  for more information on getting de-listed from this SpamCop blocklist.

There are two other tests you may wish to do on your IP to determine how reliable it may be as an originator of an e-mail message.  The first is to do what is called a Ping test which determines how well your IP communicates with other IPs.  You may find that even with a neutral reputation IP, no pings are reported, which is not a good sign.

The problem is compounded when the your e-mail message must pass across several ISPs.  To get an indication of what is involved you can do what is called a TraceRT test.  This will give you an indication of the places your e-mail message must pass through to get to its final destination.  Remember that each ISP may possibly delete your message if it seems that it could be a potential junk e-mail.

Your own e-mail service will have a certain policy on what e-mail messages it defines as junk and what e-mail messages it allows to pass through to its e-mail clients.  In some cases you may be able to modify the default policy and allow either more or less junk e-mail messages to get through to your Inbox.  If you are aware of the IP of a source of messages that you wish to see then you may be able to whitelist the source with your e-mail service provider.  However in practice this does not always work.

Is Your E-mail Message Spam?

The reason why the reputation of the IP is not an infallible indicator of junk e-mails is that the message itself is also analyzed.  A simple text message will usually get through from a good reputation sender.  Anything more than that may be questionable.  Here are some of the reasons why e-mail messages are deemed to be spam:

  • The same e-mail message is sent to a large group of people, particularly using the blind copy approach to hide their e-mail addresses
  • He e-mail message is in HTML
  • He e-mail message contains a large number of links to other websites
  • The e-mail message contains images
  • The e-mail message has attachments such as Word document files.

Whether any given message is deemed to be a junk e-mail message will depend on a combination of factors above and the reputation of the e-mail originator.  Different e-mail services will be more or less stringent in approving these rules.  For example the Google Gmail service is fairly strict in this regard.

Making sure your e-mail message gets through

There are a number of factors you may wish to consider to ensure your e-mail message gets through to as many as possible of your intended audience:

  • Ensure your originating e-mail address has a good reputation.
  • Limit the number of attachments to your e-mail messages and ideally avoid them altogether.
  • Avoid too many images in your e-mail messages
  • Avoid creating large numbers of URL links in your e-mail messages
  • If your audience accepts this, use text messages rather than HTML messages

Monitor How Many E-Mail Messages Get Through

Given that no e-mail message is guaranteed to get through, it is useful to monitor whether all your messages get through or whether a certain proportion of your audience does not receive them.  If this proportion is unexpectedly high, then you may need to change the content of your messages or the e-mail address you use to send them.

The more of these you can cover, the greater fraction of the audience that will see your message..  However there are never any guarantees and the ideal is that your recipients should also whitelist your originating e-mail address.  Even that is not surefire so in critical situations, it is better to seek some confirmation that your readers have indeed received your e-mail message.

Reblog this post [with Zemanta]

Stop Your Email Newsletter Being Junk – A Case Study


AWeber provides a fine email newsletter marketing service, so you might question whether what Tom Kulzer (AWeber CEO) writes in an article on Email Deliverability Tips may represent a slightly biased view of the situation.

Ensuring requested opt-in email is delivered to subscriber inboxes is an increasingly difficult battle in the age of spam filtering. Open and click thru response rates can be dramatically affected by as much as 20-30% due to incorrect spam filter classification.

Having found that my own email newsletter delivery system has encountered even worse problems than that, I can state that Tom Kulzer is not in any way exaggerating.  This article sets out what I found and what you need to do. It all concerns an email version of the SMM Newsletter, which was sent out monthly. 

This has been delivered to just under 1,000 subscribers for many months.  Originally it was a text newsletter but for some months has been an HTML newsletter.  It is issued using the Pimex Mail Express newsletter software.  It was sent out in blocks of 20 with a 5 second gap between blocks.  Suddenly this month something triggered the Shaw Spam filtering system and although it appeared to have been sent, it never got to its recipients.  Explanations and solutions follow.

Although it describes a multi-delivery email message, you should note that some of these problems can affect single email messages that are junked on the way to their recipient, even if they wish to receive the communication.

What Is Junk Mail

You should be clear that there are two definitions of Junk Mail.  You may believe it is mail that you do not wish to receive.  Your email delivery system has a different definition and different email delivery systems have slightly different definitions.  Each has a computer algorithm that applies a host of factors to determine whether any given email message should be deemed Junk or not.  Part of it is determined by who is sending the email message and part by its content.  In some spam filtering systems, particular senders can be whitelisted.  In other cases, this is not possible, for example Gmail.

Your IP Reputation

The key parameter in the system used by many ISPs, the Cisco IronPort SenderBase Security Network, is the Reputation of the individual IP.  This is determined by the Senderbase Reputation evaluator.

The Reputation can be shown as Good, Neutral and Poor.  Only the first Good reputation gives an assurance that email will be delivered.  For both the other categories, mail may be caught in a Junk trap.  The Reputation for the SMM IP ( at the time of writing is Poor.  By ensuring no continuing transgressions, a weak reputation may gradually be restored.  There are other databases that may be consulted.  SpamCop is another but that has no indication about the SMM IP.

What weakens the IP Reputation

The SMM IP ( is used for only minimal email apart from the batch of emails once per month.  The ISP tech support felt that this highly bunched up activity might have triggered the Poor Reputation value.  This could have the appearance of virus or trojan related email activity. 500 emails per hour may well be a possible acceptable limit although it is probably prudent to have no more than say 200 going out per hour.  This is undoubtedly affected by whether each is a simple text message for a few hundred bytes or multimedia or HTML messages that could be 25 kb or much, much more.

What Aspects Make Junk Mail

In testing different variants of the SMM Newsletter to try to avoid the Junk trap, some interesting facts emerged.  It should be noted that each ISP or service receiving an email message may have different degrees of stringency.  Indeed with the Shaw email service you can select the level of spam you are willing to tolerate.  It turned out that Gmail treated as Junk what was acceptable to the medium level of stringency for Shaw.

The fact that an email is HTML is perhaps the biggest factor affecting how a message will be assessed in Junk terms.  Another is the quality of the URLs (hyperlinks) that may appear within the message.  Initially a URL shortening service had been used for the link to the online version of the SMM newsletter.  The domain for the URL shortener turned out to be on a URL black list ( and this was sufficient to cause the email to be Junk for Gmail (although not for Shaw).  Another black list service is to be found at and you can check particular  URLs you are not sure about  at

One way of avoiding that problem is to use your own URL ‘shortening’ service.  The shorter URL for the SMM Newsletter is  The page n.htm does not exist.  Instead a 301 redirect is arranged in the .htaccess file on the domain to transfer automatically from n.htm to the actual online newsletter.

The Junk Mail solution for the SMM Newsletter

Given that the IP being used for transmitting the SMM Newsletter is still rated Poor by some services, the decision was made to use a text version of the Newsletter with a minimal number of URLs included and using the 301 redirect approach for the link to the actual online Newsletter.  Even so there were 1% of bounces due to possible spam content of the Newsletter, although there was no questionable content in the Newsletter.

Multiply Profits AND Automate Your Business

AWeber’s email marketing software makes it easy.

Learn how they can do it for you, too.

A Surer Solution To Avoid Junk Email concerns

Another way of avoidng these Junk concerns and many other concerns of publishing a successful email newsletter is to use one of the commercial services.  One which is highly rated by its users is that provided by AWeber (for which I am an affiliate).  If you would like to check out what is involved, why not sign in for that Free Test Drive in the form on the right.  You won’t regret it. The service provides all you could possibly need.

Whether you’re looking to get your first email campaign off the ground, or you’re now ready to dig into advanced tools like detailed email web analytics, activity based segmentation, geo-targeting and broadcast split-testing, you will find that AWeber has all you need to make email marketing work for you.

Reblog this post [with Zemanta]

Stopping Twitter Spam

Almost any online service can be overrun by spam and Twitter is proving to be a lucrative field for such activities.  Twitter Search that identifies Trending Topics is a major target here and very recent actions force the question: Should Twitter moderate Trending Tropics to prevent spamming? 

This is the specific case that triggered this:

For the second time in a month, offensive terms have made it on to Twitter Trending Topics. In both cases, Twitter didn’t remove the offensive terms until after they spent some time at the top of Trending Topics. This time, the attack appears to have been carried out by the infamous 4chan group.  ReadWriteWeb asks a legitimate question: Maybe Twitter Trends Shouldn’t Be Entirely Automated?

If you wish to do your bit in avoiding spam, then there is good general advice available on Avoiding Twitter Spam.  If you suspect that a particular account is a spammer then you should report them to Spam Watch

Saying no to Spam is even easier if you use the TweetDeck platform and use their spam button.  With just one click, you can delete the message from view, block the user and report them to Twitter.

The tools for Avoiding Twitter Spammers are getting more effective all the time.  Firstly, you can use Topify to give you more information on your followers, that will allow you to more quickly identify potential spammers.  It is in beta at the moment so you will need to wait for an invite.  Once you get the invite, Topify will ask for your e-mail and make sure it is you, then give you a new e-mail via Topify to enter on your Twitter account.

A more precise tool to identify spammers is TwitChuck. You simply fill in the name of the person you are wondering about, and TwitChuck goes through a surprisingly detailed list to arrive at their spam grade.  Here is the information for bwelford:

twitchuck bwelford

Perhaps you should follow the TwitChuck recommendation and follow me.
I assure you TwitChuck has it right: I am not a spammer.

Someone is wrong on the Internet

Someone is wrong on the Internet

Matt Cutts of Google has an intriguing slip in his post, Something is wrong on the internet!. He is referring to this cartoon by xkcd.

Matt Cutts said something rather than someone. He went on to say:

That comic sums up the internet in one sentence: the scrum of jostling opinions on the web and the optimism that truth can still win out. I was reminded of that comic when someone asked me about a particular way that someone recently tried to get links.

His spam group is perhaps one key way human intervention comes into the Google search process. So his comments later in the post are particularly interesting.

If a website claims to have high-quality information and then deceives the user and serves up malware or off-topic porn, Google considers that spam and takes action on it. Likewise, if a site says that they completely made up a story to get links, Google doesn’t have to trust the links to that site as much.

I really don’t view Google’s role as judging the truthiness of the web. … But if someone is sloppy enough to get caught (or to admit!) making up a fake story, I don’t think Google has to blindly trust those links, either.

It sounds very much as though Google will be acting as the judge. This prompted me to add the following comment to his blog post.

This all seems to be shaking out as it should, Matt. It raised one question in my mind. You did say I don’t think Google has to blindly trust those links, either. I believe Google’s policy is to try to do everything in its search process by computer algorithms since this is scalable. Human intervention should therefore be very limited. Your spam group does that human intervention with an on/off button, I presume, as it applies to clear spam content.

I’m sure many would be interested to know how you treat websites you are no longer blindly trusting. Do you apply the off button for these with a reminder to check again in say six months? Or is it more like a volume control where you apply a down weighting factor? Or again, is it one of those minus X penalties in the SERPs that some talk about?

Since Google is now suggesting it will be more open than it has been in the past, I hope we will get some clarification on this.

Removing Spam From The Web


Neither Hormel, the maker of Spam, nor Monty Python’s Spamalot would appreciate the sentiment in the title. However the rest of us would very much like to see the end of that other spam that accumulates on the web.

One major influence in creating all this rubbish was Google with its search view that the number of inlinks to a web page could be a measure of the importance of that web page. Once every one knew that, the name of the game was to create as many other inlinks as you could. Even though this was against the Google Terms of Service, for a time it seemed to work. The view that inlinks are what counts persists and is as strong as ever even though Google has been improving its ability to root out the spam-producers.

Now Matt Cutts, a Googler with some authority on this topic, has written a very clear explanation of what they’re doing about spam. This was triggered by a new service to create “Undetectable” spam. The follow-up to that is well described by Loren Baker in a post, Matt Cutts vs. V7N Links : Matt Wins. Another sign of the times is that Wikipedia seems to have gone the “No Follow” route in adding this tag to all its outlinks. So these will no longer count as inlinks conferring authority in the Google search process.

One would hope that the message will get around and the mindless creation of inlinks or the search for irrelevant reciprocal linking will cease. Unfortunately too many people currently waste too much time and money on these pursuits and spoil the scene for the rest of us.