| Sustainable email sending programs in an | | | | email systems that activate blocking and |
| inherently hostile environment now require | | | | filtering systems routinely run unquestioned |
| great care and planning. Before considering | | | | at both corporate and ESP data centers. Many |
| technical complexities and marketing tactics, | | | | ESP operations are being pushed into |
| email senders must adopt this basic paradigm | | | | unfamiliar technical territory in an attempt |
| shift. | | | | to improve delivery performance. If you don't |
| | | | probe past the comforting superficialities |
| The five guidelines included in this series | | | | your outcomes may be diminished. |
| should become watchwords for ezine emailers | | | | |
| as they incur the risk and responsibility of | | | | More information on email sender tools and |
| sending newsletters or any other repetitive | | | | tactics can be found in the Email PhD Sending |
| type of email. | | | | Signature Management section. |
| | | | |
| Part 1 of 5: Treat Email as a True Risk and | | | | Upgrade and modernize your systems - whether |
| Cost Center | | | | internal or external |
| | | | |
| Part 2 of 5: Avoid Collateral Damage | | | | Delivery tactics, drop-box testing, high |
| | | | granularity reporting, trends analysis all |
| Part 3 of 5: Use the Available (Legitimate) | | | | these are necessary features for email |
| Tools and Tactics (M2M) | | | | program control. If you still think that |
| | | | there is something simply called a "bounce" |
| Part 4 of 5: Build Strong Relationships (H2H) | | | | (as opposed to a "server level response |
| | | | message") then you may be lacking sufficient |
| Part 5 of 5: Continuously Evaluate | | | | data to influence your sending outcomes. |
| | | | There is a lot more to managing a successful |
| Part 3 of 5 | | | | sending process today than just pushing the |
| | | | "send" button. |
| Use the Available (Legitimate) Tools and | | | | |
| Tactics (M2M)within your email sending system | | | | In line with these needs, new levels of |
| | | | control are now being built into many sending |
| One of the most basic systematic distinctions | | | | systems, from the MTA package through to the |
| influencing email performance is between | | | | total messaging system. These can allow |
| those barriers operating under human control, | | | | managers to avoid accidentally activating |
| and those that are under machine control. | | | | many routine volume and rate blocking traps. |
| Email senders today need to understand and | | | | And virtually all system developers are now |
| work within this dicotomy. Remember: | | | | providing improved reporting and data |
| | | | integration. The use of controlled sending |
| * Human solutions (such as ISP relations) | | | | policies and scheduling can reduce your |
| tend to work for human-run systems (H2H) | | | | sending signature, and improve your reception |
| | | | and delivery at most major recipient ISPs. |
| * Machine responses (such as sender | | | | |
| authentication) tend to work for machine | | | | The same level of expertise should be |
| systems (M2M) | | | | expected from external service providers. |
| | | | Presenting an easy to use interface, or good |
| * Not understanding the differences between | | | | database integration features is important, |
| these types of systems tends to create at lot | | | | but it is not sufficient. Investment and |
| of frustration | | | | effort to control and improve email results, |
| | | | from ISP relations to technical delivery, is |
| Most of the systems that stop your email are | | | | essential in a good ESP partner. |
| automated. This means machines, not people, | | | | |
| determine whether your email gets through. | | | | Good decisions are based on data |
| From client-side boxed software packages to | | | | |
| ISP email profiling systems, it is usually | | | | Senders have been subject to a litany of |
| the things that machines can see and measure | | | | aphorisms about how to improve delivery. A |
| that count the most. | | | | few of these suggestions are entirely |
| | | | useless, and others may have only an |
| Many ESPs rely almost exclusively on H2H "ISP | | | | extremely narrow field of effect. Effective |
| relationships" at major recipient ISPs for | | | | email managers know that the only accurate |
| delivery improvement. Email managers need to | | | | way to tell what is really happening to your |
| remember that only a few ISPs even offer such | | | | sends is to test and to collect data. |
| channels, and that a large portion of their | | | | |
| email list addresses are at ISPs that don't. | | | | Many online publishers have a key metric that |
| For those destinations, and just as | | | | they like to use. Unfortunately, chances are |
| importantly, for all of the outside | | | | that it has been corrupted by some type of |
| monitoring systems that watch and record | | | | sending barrier. Other senders still rely |
| email flows, your technical (M2M) sending | | | | only on aggregate indicators of sending |
| behavior is the primary determinant of your | | | | success, or worse, indicators that can simply |
| success. | | | | tell you if your campaign succeeded or failed |
| | | | according to a measure other than delivery. |
| This section addresses the machine-to-machine | | | | This data, while better than nothing, does |
| (M2M) communication universe - that strange | | | | not provide the diagnostics and granularity |
| virtual place where what people think and see | | | | needed to control and adapt your sending |
| counts for little, but where algorithms, rule | | | | strategies. To control an email program today |
| sets, and thresholds control everything. Of | | | | it is just as important to be able diagnose a |
| course, people usually set those rules and | | | | problem as it is to simply report a result. |
| thresholds, but once in place the machines do | | | | Data collection at multiple levels is key to |
| the sorting, filtering, and blocking. And | | | | diagnosis. |
| they determine whether your recipients get | | | | |
| your messages. | | | | Complete reporting systems should provide: |
| | | | |
| Email delivery and communications is your | | | | * detailed data from the Internet |
| business, and your responsibility | | | | transmission level (server-to-server) |
| | | | |
| Successful email communications programs | | | | * data from monitored mailbox landing systems |
| actively use many types of M2M tools and | | | | (filter/junk-bin) |
| tactics to overcome delivery barriers and to | | | | |
| improve the level of recipient response. In | | | | * data from email-based actions (opens, click |
| fact, at a point in time where 22% of | | | | through) |
| permission-based email from a wide spectrum | | | | |
| of sources is being erroneously blocked, and | | | | * results measuring value created (such as |
| with email response rates far below their | | | | purchases or requested recipient actions) |
| historical highs, managers cannot afford to | | | | |
| just accept the status quo. From HTML | | | | Each of these data sources addresses a |
| checkers to drop-box landing systems to | | | | portion of the delivery and response cycle, |
| sending profile control systems, email | | | | and only in aggregate can delivery and |
| managers already use a range of M2M tools and | | | | response problems be identified and |
| resources first to map out their own email | | | | addressed. Each measure also has |
| delivery situations, and then to take | | | | characteristic sources of error (for example |
| corrective action. Even at the most | | | | image blocking for opens rates), so it is the |
| elementary level, investigation and | | | | relationship between these data points that |
| remediation of sending errors or delivery | | | | provides the most information for sender |
| failures can dramatically turn around | | | | control. This data is almost always created |
| difficult situations. | | | | from several different sources, and senders |
| | | | may need to build a customized system for |
| Senders can effectively use either internal | | | | integrating and displaying these results in a |
| or external solutions | | | | useful manner. |
| | | | |
| With the right controls in place, outbound | | | | Integrated reporting data recorded over time |
| email programs can effectively use either (or | | | | allows senders to measure changes in system |
| both) internal or external sending resources. | | | | and recipient behavior, indicating at which |
| The goal is to cost-effectively access the | | | | level problems (or improvements) may be |
| expertise and capacity necessary to operate a | | | | occurring. Changes in the pattern of these |
| sustainable system. To do this email managers | | | | behaviors is very informative to email |
| first have to realistically evaluate their | | | | managers seeking to control their delivery |
| internal capabilities as well as the claims | | | | profile. |
| of potential Email Service Providers. | | | | |
| | | | More information on email data collection and |
| The key to this process is to ask the right | | | | analysis can be found in the Email PhD Data |
| questions! In an era dominated by delivery | | | | and Tracking section. |
| and sustainability problems, many managers | | | | |
| still evaluate systems only on the basis of | | | | Copyright © by Email Ph.D. |
| convenience or ease of use. Old "broadcast" | | | | |