How spam filters work (& how to differentiate with cold emailing)

Post by
Lara Bingel
How spam filters work (& how to differentiate with cold emailing)

Get more responses from your prospects. Make more sales.
Get Started for Free Today

How spam filters work

Spam mails are not only a permanent nuisance for e-mail address owners - in many cases they also represent a concrete danger. In so-called "phishing", criminals try to obtain information about critical data such as log-ins, passwords, credit card numbers or TANs through targeted spam. Therefore, in addition to convenience, there are concrete security advantages to using a reliable spam filter.

What is SPAM?

The term spam derives from a sketch by the British comedy troupe Monty Python, in which a couple in a restaurant - completely unrelated to what they want to order - consistently receive only spam - an abbreviation for Spiced Pork And Meat or Spiced Ham. Initially, the main course includes Spam only once - later, the menu ends in a plate of ten Spam with baked beans.

From this context, the term spam evolved to describe unwanted, disruptive and unsolicited email. Unlike viruses and Trojans, spam is not dangerous in itself, provided it is deleted immediately. However, it is very important not to call contained links and in any case not to open attachments. Both of them can contain malicious software that infects a computer unnoticed.

Why is spam dangerous?

Spammers are not sweet and nice, even if it sometimes seems so when you read through these emails. In fact, if you respond to spam emails, you always get yourself into trouble.

Spam messages can contain dangerous scripts. With these scripts, the sender can spy out credentials and passwords or install viruses on your system this way.

Financial transactions:

For example, if you are supposed to send money abroad for someone else, you are liable for the damage and make yourself liable to prosecution.

Counterfeit products:

For example, if you respond to a spam email selling luxury watches at a special price. The watch is most likely a fake.


You don't get the million prize and you still make yourself liable to prosecution, as the sweepstakes is even usually organized illegally.

If you enter your data there, they will be sold or you will get a virus or Trojan on your PC.

How do spammers get my e-mail address?

It is only a matter of time until your e-mail address gets unwanted advertising or similar. There are many ways for spammers to get their hands on your email address:

Websites with security vulnerabilities:

If a website is not up to date no hacker can get data through security holes.

Buying email addresses:

Address data, as well as email addresses, can be bought by spammers from shady companies.

Guessed e-mail addresses:

A large part of the email addresses that get spammed were guessed using a lexicon of names. From this lexicon, the spammer uses software to create a list of tens of thousands of randomly generated email addresses. A large portion of the addresses will not exist and the email will bounce back, but for a small portion the email will arrive.


In this, the spammer gets an email address by searching websites for email addresses using certain software. If he finds one, the e-mail addresses end up on his distribution list, to which he either sends e-mails himself or resells them accordingly.

How can I respond to spam?

Do not reply to a spam e-mail. Otherwise the sender knows that the e-mail address is being actively used and may send you even more e-mails in the future.

- Pay attention to what data you disclose and where.

- Do not follow any request of the spam e-mail.

- Do not click on links or images in the message.

- Do not open files such as images or other attachments.

- Attachments may contain dangerous scripts that can install malicious software. Dangerous file extensions are .exe or .doc.

- Remember that the sender address can be easily faked and check if the message is really typical for the sender. More about this under Phishing Mails.

- Use a spam filter, this is the best means against spam.

What is a spam filter?

A spam filter has the task to sort harmful or unwanted messages to a certain address and to judge them according to their content of information. Different approaches exist for this purpose, ranging, for example, from sorting according to origin to artificial intelligence. In general, a filter analyzes each incoming message based on its available data and assigns it a probability that it is a "real" message with relevant information or spam. Based on this screening, it decides whether to move the email to a special folder or even block it completely.

What methods does a spam filter use?

To distinguish spam from real messages, spam filters use different features and algorithms. Some common procedures and methods are:

- Sorting the emails based on the sender

- Analyzing the content for certain keywords frequently used in spam

- Analysis of the address and metadata

- Evaluation of content by an artificial intelligence

- Trained classification by the user (Bayes or Markow filter)

- Comparison of mail address and links with a database

Each method has specific advantages and disadvantages that affect its reliability and handling. Sophisticated spam filters use several approaches in parallel to increase their accuracy.

Spam filters using data matching: blacklist and whitelist

These spam filters extract information - be it keywords, domains or server providers - and compare it with a database. Allowed or blocked parameters are stored in this database. There are several ways to configure the filter:

- E-mails must meet all or certain conditions (whitelist)

- E-mails must not contain any known spam senders or keywords (blacklist)

- Blocking all messages that do not allow classification

- Delivering unknown e-mails or moving them to a special folder

The crucial problem with this method is that it can only perform an assessment on known data. It fails as soon as no data is found in the white or black list. Depending on the database used, there is a medium to high risk that it will either fail to reliably detect spam or incorrectly categorize messages. Furthermore, the categorization is not based on probability. Most Exchange hosting plans already include a spam filter.

Spam filters based on experience: artificial intelligence and individual training

Spam filters use a self-learning algorithm that takes into account selected, characteristic features of an e-mail. It must first be learned and improves its detection over time through permanent training. The program used determines whether the spam filter must first be trained or whether it accesses an existing database. As with speech recognition, some services move the analysis to the cloud - this has several advantages:

- Reliable optimization through large amounts of data

- External servers take over computationally complex operations

- Fast response to new variants of spam

- Universal and international coverage

- Good detection immediately after setup

For the end user of a mail address, artificial intelligence also has disadvantages. When taught individually, it must first be trained - this requires at least 1000 messages before the algorithm achieves reliability with a single-digit error rate. Using a service in the cloud eliminates this preliminary work, but individualization is only possible to a limited extent. If messages with certain content are classified as spam by a majority of users, but are personally interesting for various reasons, this leads to problems.

Spam filter on several levels

Many providers such as Gmail use internal spam filters even before a message is received in order to select incoming mail and thus reduce data transfer and memory consumption. Various databases on the Internet, such as SORBS, register IP addresses, providers and domains that attract attention due to intensive spam sending. Servers or URLs that are blacklisted by these organizations are automatically blocked by some Internet Service Providers (ISPs). When identified, the mail server terminates the contact after passing the metadata and reports an error. In the case of offerings such as Exchange hosting, this happens automatically.

The next level are programs that monitor the inbox on your own computer for spam. Depending on how they work, they basically reject receipt or move a message to the appropriate folder. Plug-ins that allow integration of additional spam filtering email clients such as Outlook or Thunderbird work in a similar way. They can be installed subsequently and apply various procedures.

Spam in contrast to cold emailing

As described above, sending spam messages is very different from sending targeted sales emails that are intended to explain added value to potential customers. Unfortunately, this is often misunderstood in the context of spam regulations, which is why these emails are increasingly classified as spam. You, as a sales team, of course have a legitimate interest in promoting your offer to qualified potential customers. For this reason, it is important that you ensure that spam filters understand that your sending behavior is natural. This means that in order for your sales emails not to end up in spam, you not only have to write clean, target group relevant content and really only send to the recipients who could be relevant, much more your average response rate has to correspond to that of a regular email account. Mailivery takes care of this. By intelligently exchanging emails, the average response rate is increased and fewer of your sales emails end up in spam, and more in the inbox.

Summary: Spam filter - A useful addition

Unlike an advertisement or a letter, sending an email has practically no relevant costs. For this reason, it is not uncommon for spammers to send messages to millions of addresses, some of which they generate randomly. On the black market, it is also possible to buy lists that combine a valid mail address with additional information such as contact details. These are composed of illegally obtained and public data and allow spam to be individualized. If your sales emails no longer reach the recipient and you are sure that you are not a systematic spammer, but write qualified, targeted addresses, then you should use a service like mailivery.

Don't Land In Spam.
Make more sales.
Get Started Today For Free.