Return Path’s New Year Community Challenge: Week 4
This week’s email best practice is: Get Your Infrastructure In Order
Unfortunately, marketers do not get to ignore infrastructure just because you have a good IT team or use a good email broadcast vendor. Your sender reputation is directly linked to the quality of your infrastructure. Use this column to audit your system and gain confidence that you are in good shape. And you should check in on your infrastructure on a regular basis – things change and it’s important to make sure all the wheels are oiled. For most companies a quarterly check is going to do the trick, though you should work with your IT folks to set up a maintenance schedule that makes sense for your business. And certainly any time you make a major change – adding a server, changing content management systems, and so on – you’ll want to double-check that everything is set up correctly. You will also want to review this information if you find that you are being blocked by an ISP – infrastructure may be the culprit.
Servers that are incorrectly configured present a real problem for the ISPs and other large-volume receivers. They don’t know if it’s a legitimate server that just has a glitch, or a malicious botnet, zombie or other harmful machine. In fact, our Benchmark Data report found that 34% of email servers are impossible to identify as either good or bad. Only about 20% of the servers we studied were properly configured to send email.
That puts the vast majority of us at risk! But the good news is that if you do have infrastructure issues they are generally relatively easy to fix. And, once you’ve fixed them you will usually see a dramatic improvement in your deliverability rates, absent other reputation issues like high complaint rates, high unknown user rates and spam trap hits.
So, with thanks to my colleague Greg Gould, Director, Support Services, here are top ten common infrastructure issues that you will want to verify are working well in your own system, even if you outsource to an ESP. A good service provider should be happy to review their system approach with you and give you the confidence you need to know you are protected and served well.
1. All sending IP addresses must have a valid rDNS PTR record and the host name specified in the HELO must match the host name returned in the rDNS. Some ISPs will block you for reverse DNS or HELO failures. Also, note that if you are working with an email broadcast vendor, it’s always better to get a dedicated IP address for your mail. Sharing an IP address with other customers means that their reputation affects your reputation.
2. Your sending domain must be able to receive mail as well as send mail. It doesn’t have to be the same server, but you must have a valid MX record for that domain. Some ISPs will block you if you don’t.
3. Setup and monitor postmaster and abuse mailboxes for all your mailing domains. Many ISPs require that these are working to be signed up for feedback loops or white listing. These are also common destinations for complaints for ISPs that don’t run feedback loops.
4. Set your MTA connection and throughput settings according to the ISPs published guidelines. Many MTAs will allow you to set these per receiving domain. Sending more mail faster isn’t always a good thing and in fact can get you blocked.
5. Secure your mail servers! Make sure you don’t have an open relay or open proxy. Follow industry standard best practices for network and server security. All the best mailing practices don’t matter if you don’t have control of your environment.
6. Process your complaints. It’s critical to get setup on all available feedback loops, but that’s only half the process. Make sure that you promptly suppress mailing to any addresses that complain.
7. Monitor and report on mail delivery logs. Make sure that you are properly classifying and handling hard bounce, soft bounces, deferrals, and so on. Our friend Dennis Dayman from Eloqua has a great post on Deliverability.com that explains this in more detail.
8. Don’t keep sending mail to an ISP if they are blocking your messages. Work with the ISP to resolve the block first.
9. Never move IP addresses to resolve deliverability problems. The ISPs are wise to this and so they regard new IP addresses with extreme suspicion. In fact, all IP addresses start with “no reputation,” and must be “warmed up” by your good practices. If you haven’t cleaned up whatever problems caused your poor reputation at the old IP, you will just end up with a new IP with the same old bad reputation.
10. Make a practice of quarantining data until you know it’s safe. Use your welcome message or even the first three to five messages to ferret out bounces, ill formed addresses and even sources with a high likelihood of complaining.
Working with your IT team or email broadcast provider to identify and fix infrastructure issues can be a challenge, but the payoff is usually big. Once you have these nuts and bolts issues taken care of you can focus your time and attention on creating emails that your subscribers will really value. Stay tuned for more on that in next week’s post.
Also, if you haven’t officially signed up for the Challenge, do so now. It’s the only way to get our Challenge Wrap-Up Report with all the great best practice advice you need for a prosperous 2009. Although, if you send me chocolate I might consider sharing part of it with you.
Start talking about this week’s best practice:
- Do you feel in the dark or intimidated by infrastructure? What information helps you feel confident asking questions and getting support?
- Have you encountered infrastructure issues? If so, what were they and how did you solve them?
- Have you found creative ways to work cross-departmentally in your company to get marketing and IT in synch on email issues?
- What is your biggest technical frustration when it comes to sending email?