Here’s why marketer’s need to understand dedicated and shared IPs
Your reputation is tied to your IP and your domain, thus, both have to be properly configured and protected.
During a Q&A at a recent event, I was asked which was better: dedicated or shared IPs. This is a great question because there are many factors to consider. The person asking was fairly certain that shared IPs are better, but it’s not that simple. Not all is as it seems.
Dedicated IP addresses are awesome if you have the volume and frequency to take advantage of building your reputation. However, for smaller senders, SMBs, local businesses, and entities that send infrequently, a shared IP is probably the best route to achieve the benefits of mutually-combined volume. Senders that send fewer than 50,000 emails a month should consider a shared pool because they won’t build enough volume to help a mailbox provider understand and assign their sending IP a positive reputation.
Large senders, those that send routinely and deliver more than 50,000 emails a month, need a dedicated IP to establish and build their reputation versus simply inheriting the shared reputation of a pool of senders who might have wildly different sending patterns, styles, volumes, and audiences.
Consider your neighborhood
In addition to establishing a dedicated IP for larger senders, those brands that will send larger volumes should consider the neighborhood of IPs that they’re sending from. A bad sender could be so bad that the entire class C IP range becomes blacklisted to contain and limit the damage. A class C network consists of 255 IP addresses.
When you obtain a dedicated IP address, it might be worth having a conversation with your provider to understand the kind of network your IP is a part of. Is the rest of the network employed by other large sending entities that represent, with any luck, less risk? Or is the dedicated IP on the same class C as multiple shared pools where bad traffic spikes can drag down the entire network, thereby conferring what I like to call “digital guilt by association”?
Uncertain = Untrusted
There was a time when new IPs were worth their weight in gold (if such a thing were possible). The reason new IPs were so valuable was that they had no sending reputation. They had never been seen in the logs of mailbox providers such as AOL and Yahoo! Thus, a sender (good or bad) with a new IP address could deliver email unfettered, at first, via this new golden IP.
That was a lifetime ago.
Today, this is no longer the case for a couple of reasons. First, there aren’t any new IPs. Quite the contrary, we’ve run out of IP space globally and had to create a brand new IP schema called IPv6 to make way for the universe of IoT devices—each of which will require an IP address. Second, and this is probably the more important of the two reasons, mailbox providers have long since learned that IP addresses with no reputational history, e.g., those that they’ve never received emails from, should be treated as suspect vs. getting a free pass to the inbox. In today’s peril-fraught messaging ecosystem, new IPs have to put in the time to build a positive reputation before they are given the keys to the kingdom.
Smaller senders obtaining a new IP may have a hard time building enough volume on that IP to demonstrate their mailstream deserves to be anointed with a good-sending reputation. Without sufficient production volume, that IP will essentially remain a liability in the eyes of the mailbox provider.
More is not necessarily better
I’ve met several marketers who think that having a bevy of IPs at their disposal is a necessity. If you’re reading this, and you agree, let me stop you right there. More is not better. Snowshoe attacks take the form of lots of emails coming in small volume/batches from lots and lots of sending IPs. These attacks are tough to spot and even harder to stop. Marketers that employ too many IPs may present their mailstreams as having the characteristics of snowshoe spammers. At a minimum, every sender should maintain two IP address: one for transactional emails (because these are treated differently under CAN-SPAM) and one for commercial email.
When more is necessary
Diversification and reporting are good reasons to increase the IP addresses your mail is sent from. If you’re sending on behalf of multiple brands that have different audiences, frequency of campaigns, and cadences, you’re likely to build a different reputation for each brand on that IP. Separating mail streams by brand, product, or another logical purpose to achieve granular reporting and unique reputations is an absolutely good reason to add IPs.
Another reason to have more IPs at your disposal is timing. Some platforms have limitations on how many emails they can process and send out by IP. Increasing IPs to hit delivery windows, or, for example, to make good on time-sensitive offers, could require additional IPs to ensure mail flow happens in a timely manner. But even in this scenario, the number of IPs should be determined by experts versed in mail operations and the nuances of the platform and done so in a conservative manner to prevent the mail stream from resembling a snowshoe attack.
Fix it, don’t swap it
When deliverability issues arise, the absolute worst thing you can do is to change your IP. Let me repeat that: switching IPs in the middle of a delivery crisis is really, really bad. This is what a spammer would do: they burn through an IP and then move their traffic onto another IP or platform to avoid detection. Delivery problems must be solved—running away from your problems is bad in life and email. Don’t do it. Delivery issues most often stem from how you obtained the email address in the first place. Diagnosing your opt-in issues, or segmentation, is the first thing you should look at. Switching your IP address will only mean that the problem will follow you and compound as you begin to restart a problematic mail flow from a different source.
Reverse it transparently
Reverse DNS isn’t something marketers often consider, so if you’re not interested in how the sausage is made, go ahead and skip this bit. Reverse DNS (or rDNS) is the hostname of your IP. When you search, say, Google.com, you will get an IP address like 22.214.171.124. If you were to look at the rDNS of 126.96.36.199 you’d get the name of the IP, which might look like nrt12s14-in-f14.1e100.net. See, I told you this is how the sausage is made. I’ve spoken to some mailbox and blacklist operators that prefer platforms to employ an IP hostname that denotes the sender. Instead of using the above gobbledygook, you could consult with your platform provider to use something like brand.provider.IPxxxx.com. Now, most providers usually have a standard schema for how their IPs are configured, but if you want to take hold of how your IPs affect your sending, start with the foundation: IP configuration.
Like the tortoise, steady is best
I’m not saying that you should send your email at a snail’s pace, but you should understand the published and acceptable connection and throughput rates. Instead, what I’m referring to is frequency. Instead of delivering mail infrequently in large increments, see if there’s a way to send with more frequency and greater regularity. If you can avoid the massive spikes with giant lulls of inactivity, you will build a reputation faster. If you’ve done your homework and set up useful organic methods to grow your recipient lists, then the reputation you build will be a positive one.
Reputation starts with establishing an IP. Think of IPs as a business commodity that needs to be treated with respect, not abused. Your reputation is tied to your IP and your domain; thus, both have to be properly configured and protected.