Email security policy: statistics have not changed in 4 months.

6 months after launch, the technical correlation statistics performed by LetzRelay-MX during the first phase of analysis, among the three main ones, show very few changes compared to the data collected 4 months ago (See our LinkedIn posts on that matter). However, the number of emails passing through the platform has increased significantly.

A quick reminder of the 3 main analyzes performed in real time by LetzRelay-MX

1. Technical correlation

• Reconstitution of the entire email journey in terms of sending and/or transfer servers. Each transfer results in electronic signatures applied by different relays. This allows, among other things, geographical traceability of the route.
• Verification of each electronic signature to ensure that the email complies with the official security policies issued by the sending organization (See our blog posts Why is email authentication important? and What is the difference between DKIM signature and PKCS#7 signature?).
• Reputational verification at both the individual and sending organization levels, cross-referencing a multitude of information such as spam listings, social networks, etc. Hundreds of different sources are currently used for this verification.
• Scanning and sandboxing of each link/URL contained in the message, as well as each attached file, to check for any malicious content or traps set for the recipient(s) through these elements. If necessary, deactivation of such "bad" elements.

2. Semantic correlation

At this stage, artificial intelligence takes over to create a context for understanding the message: Is it a professional email, an email issued by a governmental organization such as the police or a tax administration? Does a link in the message redirect the user to a fraudulent page requesting confidential information? etc.

This semantic analysis is now capable of processing in 4 languages (EN/FR/DE/PT).

3. Cyberscoring

Once the first two steps are completed, artificial intelligence gathers all the collected information and derives a cyberscore. For example, an email from the police that has passed through one or more suspicious countries will receive a low cyberscore and therefore be disarmed.

Furthermore, if the artificial intelligence engine detects that the sender is not making their first attempt, it will directly intervene on the blacklists to immediately reject any next attempt.

The statistics of technical correlation haven't unfortunately changed much

Among the various statistics that could be established, two seem particularly important to us:

1. The number of organizations still only partially or not at all implementing basic security policies: It appears that barely over half have implemented a valid security policy.
   
   
   
2. The proportion of legitimate emails compared to the total number of messages sent: Just over 40%! This means that 60% of the emails sent are polluting the recipients' inboxes.
   

LetzRelay-MX users were able to benefit from a considerable time savings as these unnecessary emails were clearly identifiable and therefore did not require their attention.

Between the data from 4 months ago and today, the graphical representations have, so to speak, remained unchanged.
Organizations, on the other hand, were effectively protected as the dangerous elements were disabled; everyone continued to work peacefully.

Nevertheless, a serious question must be raised regarding the environmental impact due to these illegitimate messages, which maintain a significant energy consumption burden on email servers.

Learn more on how LetzRelay-MX can help your organization be secured with Internet inbound emails.