Here is a sample of data from receivers using the Registry. For now, the numbers are hypothetical - spam is 10% of the flow from every domain. In reality, there are large differences between domains (see below). There are also large differences in total volume. This explains why a small domain can suddenly lose its A-rating when it lets a few spams slip out, and a large domain can send millions a day with no effect on its rating. It has to be that way, because when a recipient sets a threshold for acceptance of mail from an unknown domain, there is only one thing that matters - what is the chance that this mail is spam?

Small domain owners shouldn't lose hope, however! Most registered senders are small domains, and most are completely spam free. They would not have registered if they didn't care about the quality of their outgoing mail.

If the charts below are missing, try viewing in Firefox.  For some reason it won't display in Internet Explorer, even though it is an Excel Chart in a page produced by Microsoft Word!!

If the chart doesn't appear, try viewing with Firefox. For some reason, it doesn't show up in Internet Explorer, in spite of the fact that this webpage was created by Microsoft Word, and the chart is a Microsoft Excel picture.

Note: These scores are not intended as an overall rating of the domains, but rather just a snapshot of what our receivers are currently seeing in the mail they accept from these domains. A domain with a large amount of spam may be no "worse" than one with a low amount, but still show a poor score simply because one domain has not published an authentication record and the other has, or maybe they have both published records, but one has listed all their IP addresses as authorized servers for their domain, and the other just their own "in-house" mail servers. We score on everything that we must accept, and we must accept whatever the domain doesn't prohibit in its authentication record.

 

 

 

 

 

Here is some real data from a small sample of Spamcop reports.

 

There are substantial differences in the % of spam coming from different domains, like a factor of 100!!
 

 

                              Daily   Spamcop   Suppression

                  Magnitude   Volume  Reports   Ratio

      charter.net       6.7   7.9E+06     37    -5.3

      sprint.net        3.0                                 Network owner unknown. ???

      sprint            7.0   1.6E+07     70    -5.4        all 484 domains owned by sprint

      aol.com           7.8   1.0E+08     6     -7.2

Magnitude is from senderbase.org = log10( domain volume / internet total )

Daily Volume = 1.6E10*10^(Magnitude-10)

Suppression Ratio = log10( Spam Reports / Daily Volume )

Extracting spam scores from these reports is difficult.

 - Its a small sample with a lot of uncertainty.

 - We can't get an absolute ratio (% of spam) because we are comparing spam reports sent to one company vs another company's estimate of the worldwide volume from that domain.

Nevertheless the conclusion is firm.  Some domains are much better than others.  We just need to get this information into a form that email receivers will understand and take action.  Senders will then discover that cleaning up their outgoing mail was not so difficult after all.

 

Microsoft Excel Chart Object:

 

Picture (Enhanced Metafile):