Here's a common question I am hearing lately: "Everybody buys business intelligence data from business intelligence data providers. Usually including email addresses. Can't I just take that list and upload it into Mailchimp (or some other email marketing platform) and start sending marketing messages?"
I wouldn't, personally. I think it's a bad idea. You might think I'm a stick in the mud, an old school anti-spammer whose take on things needs to be updated for these modern times. You could be right. I'm just this random guy with an opinion. But I don't think my take on this is wrong. Allow me to explain where I'm coming from.
Legitimate email service providers still prohibit sending mail to purchased lists nowadays. They've got a strong incentive to identify (and prohibit, or at least react to) sends to purchased lists and data. Why? Because of a very important practical issue. Bulk email sends to purchased lists cause deliverability problems. They get ESP clients in trouble -- you, the marketer, sending to that purchased lists -- and they can get ESPs into trouble as well.
Back in the old days, when IP reputation was king, Spamhaus closely monitored their spamtraps (bad addresses used to identify bad senders based on volumes of mail received -- addresses that haven't opted-in) to identify who the bad senders are. There was (and still is) a strong correlation between use of purchased list data and mail to spamtrap addresses. Sending to lots of bad data, because you're sending to a purchased list, was a quick way to end up with a Spamhaus blocklisting issue that would result in a 60%+ bounce rate for your email campaigns. It, to put it bluntly, caused big pain.
Spamhaus still exists today, and their IP reputation data is an important tool to help internet service providers (ISPs) and mailbox providers (MBPs) block connections from bad servers trying to hand off bad things. Nowadays, those efforts are seemingly more focused on the worst-of-the-worst -- electronic gangs involved in phishing, malware, and fraud -- and perhaps less focused on plain old email marketing, good (opt-in) or bad (spam). But spamtrap hits (mail to spamtrap addresses) still cause Spamhaus SBL and CSS blocklistings and still cause deliverability problems today.
And newer filters can cause you even more trouble.
Purchased lists have higher complaint rates, higher bounce rates, and lower engagement rates than opt-in email lists. In this modern era of engagement-driven spam filtering and domain reputation, these poor metrics are quicker to drive spam folder placement (Iow engagement) and even blocking (high complaint rates) at Gmail, making it hard to succeed at email marketing, given that mailboxes hosted by Google so overwhelmingly show up in every email marketer's database. You can't do email without sending to Gmail, and Gmail's pretty good at figuring out which mail is wanted and which mail isn't.
If your domain ends up with a bad reputation due to sending purchased (bad) data, Gmail will have the data to show it. And other mailbox providers will figure it out, too.
And as I noted above, ESPs themselves have been known to get into trouble because of client sending to purchased list data, too. Some blocklists list whole blocks of IP addresses. Some "network neighborhoods" end up with broad, poor reputations because of it, and it can impact clients who aren't doing anything wrong. Broad IP block-based DNSBL issues are a bit rare, but default ESP-wide shared domains used for DKIM authentication can get tainted with bad reputation thanks to bad senders, affecting deliverability negatively even for good senders. That's actually somewhat common nowadays.
And there was that one time Spamhaus blocklisted 255 IP addresses -- 250+ clients -- of a single email service provider, because they were fed up with Spam from ONE of the ESP platform's clients. The sender was a nationally known clothing retailer, who decided that they didn't have to follow the rules. It didn't go well for them. I'm pretty sure they were invited to switch email service providers after that happened. This might be rare, but it does happen.
Can you imagine? You're an email marketing automation platform, serving up millions of marketing and transactional email messages. Maybe you've got some marquee clients, like a top 50 financial institution. Do you want that rather large Bank to have trouble, on your platform, sending valuable transactional email messages, because of something bad some other client is doing? It happens, and it leads to really uncomfortable conversations when it does.
So, for these reasons and more, that is why legitimate email service providers and marketing automation platforms don't want you to purchase email lists.
There likely are a few email marketing platforms who look the other way, or are a bit "wink wink" about it, even if their policies on paper prohibit use of purchased list data. I think they're playing a game, and it is a bit like contamination in the water supply. If it is only very small drops of bad stuff (a few clients, or maybe .01% of a client's list), it might go unnoticed. It doesn't always rise to a high enough level to trip the alarm or make people sick. But short term success here is based on hope and luck, which I find to be questionable strategy. It's a model based on dribbling in contaminants and crossing your fingers that you keep it just "clean enough" that nobody comes down with symptoms.
It is a methodology that doesn't scale up very well and often, regularly, I get approached by people looking for deliverability help, who seem surprised that their sending to bad data caught up with them eventually. Why am I having trouble now? Well, either because the mailbox providers you're having trouble with just got better at detecting unwanted contaminants in the email stream (filters are forever improving), or it took a while for the potential poison to build up to a level to be noticeable. Either way, the solution is: get clean -- don't send to purchased email list data.
Here's a common question I am hearing lately: "Everybody buys business intelligence data from business intelligence data providers. Usually including email addresses. Can't I just take that list and upload it into Mailchimp (or some other email marketing platform) and start sending marketing messages?"
I wouldn't, personally. I think it's a bad idea. You might think I'm a stick in the mud, an old school anti-spammer whose take on things needs to be updated for these modern times. You could be right. I'm just this random guy with an opinion. But I don't think my take on this is wrong. Allow me to explain where I'm coming from.
Back in the old days, when IP reputation was king, Spamhaus closely monitored their spamtraps (bad addresses used to identify bad senders based on volumes of mail received -- addresses that haven't opted-in) to identify who the bad senders are. There was (and still is) a strong correlation between use of purchased list data and mail to spamtrap addresses. Sending to lots of bad data, because you're sending to a purchased list, was a quick way to end up with a Spamhaus blocklisting issue that would result in a 60%+ bounce rate for your email campaigns. It, to put it bluntly, caused big pain.
Spamhaus still exists today, and their IP reputation data is an important tool to help internet service providers (ISPs) and mailbox providers (MBPs) block connections from bad servers trying to hand off bad things. Nowadays, those efforts are seemingly more focused on the worst-of-the-worst -- electronic gangs involved in phishing, malware, and fraud -- and perhaps less focused on plain old email marketing, good (opt-in) or bad (spam). But spamtrap hits (mail to spamtrap addresses) still cause Spamhaus SBL and CSS blocklistings and still cause deliverability problems today.
And newer filters can cause you even more trouble.
Purchased lists have higher complaint rates, higher bounce rates, and lower engagement rates than opt-in email lists. In this modern era of engagement-driven spam filtering and domain reputation, these poor metrics are quicker to drive spam folder placement (Iow engagement) and even blocking (high complaint rates) at Gmail, making it hard to succeed at email marketing, given that mailboxes hosted by Google so overwhelmingly show up in every email marketer's database. You can't do email without sending to Gmail, and Gmail's pretty good at figuring out which mail is wanted and which mail isn't.
If your domain ends up with a bad reputation due to sending purchased (bad) data, Gmail will have the data to show it. And other mailbox providers will figure it out, too.
And as I noted above, ESPs themselves have been known to get into trouble because of client sending to purchased list data, too. Some blocklists list whole blocks of IP addresses. Some "network neighborhoods" end up with broad, poor reputations because of it, and it can impact clients who aren't doing anything wrong. Broad IP block-based DNSBL issues are a bit rare, but default ESP-wide shared domains used for DKIM authentication can get tainted with bad reputation thanks to bad senders, affecting deliverability negatively even for good senders. That's actually somewhat common nowadays.
And there was that one time Spamhaus blocklisted 255 IP addresses -- 250+ clients -- of a single email service provider, because they were fed up with Spam from ONE of the ESP platform's clients. The sender was a nationally known clothing retailer, who decided that they didn't have to follow the rules. It didn't go well for them. I'm pretty sure they were invited to switch email service providers after that happened. This might be rare, but it does happen.
Can you imagine? You're an email marketing automation platform, serving up millions of marketing and transactional email messages. Maybe you've got some marquee clients, like a top 50 financial institution. Do you want that rather large Bank to have trouble, on your platform, sending valuable transactional email messages, because of something bad some other client is doing? It happens, and it leads to really uncomfortable conversations when it does.
So, for these reasons and more, that is why legitimate email service providers and marketing automation platforms don't want you to purchase email lists.
There likely are a few email marketing platforms who look the other way, or are a bit "wink wink" about it, even if their policies on paper prohibit use of purchased list data. I think they're playing a game, and it is a bit like contamination in the water supply. If it is only very small drops of bad stuff (a few clients, or maybe .01% of a client's list), it might go unnoticed. It doesn't always rise to a high enough level to trip the alarm or make people sick. But short term success here is based on hope and luck, which I find to be questionable strategy. It's a model based on dribbling in contaminants and crossing your fingers that you keep it just "clean enough" that nobody comes down with symptoms.
It is a methodology that doesn't scale up very well and often, regularly, I get approached by people looking for deliverability help, who seem surprised that their sending to bad data caught up with them eventually. Why am I having trouble now? Well, either because the mailbox providers you're having trouble with just got better at detecting unwanted contaminants in the email stream (filters are forever improving), or it took a while for the potential poison to build up to a level to be noticeable. Either way, the solution is: get clean -- don't send to purchased email list data.
TL;DR? Should I use a purchased email list? No.
Comments
Post a Comment
Comments policy: Al is always right. Kidding, mostly. Be polite, please and thank you.