According to various studies conducted by MarketingSherpa, Assurance Systems and SilverPOP, at least 4% and possibly as much as 25% of your broadcast email is filtered out before it reaches inboxes.

When a filtering system stops permission email by mistake, it is known as a "false positive."

The problems with filtering and false positives are too numerous to discuss here. Instead, I am going to outline your options for avoiding filtering altogether.

==> Three Approaches to Avoiding Filters

1. Get on the whitelist of the ISP, MSP (mail service provider, such as Yahoo or Hotmail), or organization doing the filtering.

Unfortunately this does not scale well. For B2B mailers it is a complete impossibility; there are simply too many organizations receiving your email.

2. Run your outbound email through a content-evaluation system such as that offered to their own clients by CheetahMail, or those available from Lyris, SiteSell or Assurance Systems, and change your content to avoid the rules in place. (Links below.)

This assumes that email blocking is based entirely on content filtering. Unfortunately, content is by no means the only reason your email may be filtered.

AOL, for example, will not deliver mail if your mail server is not configured the way they expect it to be. Yahoo, on the other hand, requires good list hygiene. If you start ignoring their bounce messages, and keep trying to deliver to what they have already told you are bad addresses, expect to end up in their bulk-mail folder.

3. Get your email whitelisted by joining a group or system that has already negotiated whitelisting with most filtering programs for its members or users.

Below are my evaluations of the three best-known services that claim to help you with this: Positiva's Trusted Sender Program, Habeas, and IronPort's Bonded Sender Program.

===> The Short Version: Quick Summary of My Findings

A. Postiva’s Trusted Sender Program, the technology of the ePrivacy Group bundled with the privacy assurances of Truste.

The ePrivacy Group’s system was announced a year ago with beta rollout to a handful of very large mailers. Today, it is still in beta. Not a single major ISP or MSP is whitelisting based on the Trusted Sender postmark. Currently the system does not require that lists be permission-based, so it is unlikely to be whitelisted ever without significant changes.

This is not a useful solution for mailers wishing to avoid filters at this time. More on this below.


B. Habeas, a combination of haikus hidden in the header and an IP whitelist guarantee delivery to a number of major ISPs and MSPs.

Habeas has come storming out of nowhere to gain clear market dominance in the filter-avoidance space. The good news about Habeas is that they are broadly whitelisted, currently assuring you will reach around 250 million email addresses without being filtered.

The bad news (for most marketers and publishers with existing lists) is that they require lists be confirmed opt-in in order to be covered by Habeas protection. More on this below.


C. IronPort’s Bonded Sender Program, the offshoot of an email hardware vendor, created to address the needs of its users to see their messages delivered

IronPort may ultimately have the best solution.

They keep a publicly accessible whitelist of the IP addresses of companies such as email broadcast firms that have agreed to their terms. The penalty is clear: $10 per spam complaint against any IP address on their list, payable by the IP owner who posts a bond to be allowed to use the system.

While their IP list is not yet part of the in-house whitelist at AOL or Yahoo, it is checked over 10 million times a day by corporations and universities trying to provide balance to the information they are using from publicly available blacklists. More on this below.


===> Extended Play: More Details About Each System


A. Trusted Sender: A Solution Chasing Last Year’s Problem

When Trusted Sender sent out its launch press release almost exactly a year ago, their goal was to stop filtering by solving the “falsified header” problem.

At that time, ISPs and others receiving lots of email would rely heavily on information in the header of the email message to determine whether the mail was legit or not. Trusted Sender provided an encrypted seal in email promising incoming email servers that the email really came from the source that it claimed to come from.

Most of the header can be falsified, so it seemed like Trusted Sender's solution might be welcome; if it ever got out of beta.

In the meantime, content-based and IP-based filtering became more pervasive. Unfortunately Trusted Sender's $4000 solution does not currently address either of other types of filtering.

Although Trusted Sender covers a handful of other things in the seal, such as that the sender is still in good standing with Truste, they do not require that the senders prove names on their lists wanted or requested to get mail. This means whitelists are unlikely automatically to admit mail with the seal anytime soon.

In fact, Trusted Sender does not anticipate it will ever get guarantees from ISPs and MSPs to have their email pass through without filtering. As their CEO Vincent Schiavone said to me, “No one can promise you that.”

As things stand now, I do not expect Trusted Sender to gain traction.


B. Habeas: It is Everywhere Confirmed Lists Want to Be

According to CEO, Anne P. Mitchell, “Major ISPs and MSPs that accept Habeas warranted email without content filtering include AOL, Yahoo, and Juno.” The Habeas warrant is also whitelisted by SpamAssassin, DeClude, MessageLabs, and others that provide content-based filtering for many corporations.

SpamAssassin alone is huge because its open-source rules are used by many other filtering tools. Total email addresses covered by the Habeas Users’ List: ~250 million.

The founders of Habeas were not willing to wait for laws to catch up with email technology. They decided that since copyright laws carry ample penalties, they would rely on copyright laws, rather than hardware-based encryption to keep anyone who was not Habeas certified from using the Habeas warrant.

The advantage of the Habeas' non-technology approach is that implementation is nearly instantaneous, once you have agreed to their terms and paid the fee.

Habeas actually offers two methods for mailers to use to get their mail through. For marketers and publishers who do not have their own IP addresses for outbound mail (most mailers who use broadcast email providers) they can include the Habeas warrant in their email headers. For anyone who complies that does have a dedicated IP address, they can be included on the Habeas Users’ List, which is what Yahoo, AOL, and Juno rely on.

A list of IP addresses available in real time from a trusted third party offers the same quality of certification to the receiving email server as a hardware-based encryption, without the implementation hassle. Realize that most receiving email servers are already checking 3rd-party IP address lists for blacklisting information, so this is not a stretch technically for the receiver to implement.

The pricing for Habeas is trivial if you are a publisher ($200 per year) but more significant for marketers, sliding scale up to $3,000 per month, depending on volume.

(Note: Do not rely on the vague language on the website for whether you qualify as a publisher or a marketer. Which payment schedule you qualify for is subjective.)

The major implementation hassle for wanna-be Habeas users is the requirement of a confirmed opt-in list. Any list older than 18 months is unlikely to be confirmed opt-in. Also, attempts at re- confirming existing subscribers typically see dismal conversion rates in the thirties.

Analysis: A good solution for double opt-in lists.


C. Bonded Sender: The SpamCop Model for the Good-Guys

Bonded Sender is most easily understood as a publicly available list of IP addresses of companies that are willing to guarantee with a cash bond that email sent to their lists will not generate spam complaints.

Just as SpamCop offers a freely available list of IP addresses of companies associated with alleged spammers, IronPort keeps a list of the good-guys. Any company relying on publicly available blacklists might want to begin including the IronPort list of trusted IPs in order to help reduce their “false positive” rate.

IronPort is a hardware vendor. According to Tom Gillis, they do not see the Bonded Sender program as a business model for them. They created it to address the concerns of their corporate clients whose email, coming out of their hardware boxes, was being filtered.

Bonded Sender has yet to achieve the kind of roll-out that Habeas has achieved. That is a shame because this is a solution that anyone with an opt-in list that they know to be relatively complaint free could be using to help get around poorly configured content filters in many organizations.

The cost for joining Bonded Sender is simply the cost of the bond. (Existing IronPort customers with a good credit history can skip the bond.)

The penalty is simple. Every complaint of spam reported to IronPort against your IP address costs $10 out of your bond.

Of course, in order to implement Bonded Sender, you or your broadcast email service must have a dedicated IP address that is used to send your email only.

According to Gillis, many major mailers are using Bonded Sender only on their opt-in lists, then using an alternative IP address for new-customer acquisition to rented lists.

Bonded Sender fits nicely with the way that inbound email servers currently make decisions about what email to accept and reject. In short, Bonded Sender, with its understanding that even single opt-in lists can be good ones, is the solution to watch.

==> Related Links:

Content Checking Resources:
Lyris’ ContentChecker: http://www.lyris.com/contentchecker/
Assurance Systems: http://www.assurancesys.com
SiteSell’s Content Checker: http://spamcheck.sitesell.com/

Services Reviewed Above:
Trusted Sender: http://www.postiva.com
Habeas: http://www.habeas.com
IronPort Bonder Sender Program: http://www.bondedsender.com