Audience Bridge Insights #4 - Email Deliverability Secrets with Kyle Renfrow (CEO of OptiPub)

Foreign.

Welcome to another episode of Audience Bridge Insights. I'm really excited about this one. We're having our first ESP on the show. I've got with me Kyle Renfro, the CEO, CTO, co founder of OptiPub. Kyle, how's it going?

Good. What's up? Thanks for having me promote.

Um, to begin and get started so people can kind of understand who OptiPub is. Can you give us a little history or how you decided to. To build optipub? Where it came from?

Yeah, sure. I mean, so basically I started at Angel Publishing, a financial publisher at Shit19. I was a web developer there and when I came on board, they were sending their email with XM servers and I guess maybe a year in, the system administrator left and I took over the sysadmin stuff and honestly didn't know a ton about email at the time. So the. One of the first decisions I made was to move our email to an ESP. And we basically bounced around a couple ESPs and eventually as we scaled, decided to bring email back in house and just started building out these. This platform essentially for Angel. And we'd go to conferences and some of our partners and friends just pretty much just asking us for it. So we just, it kind of just naturally happened. We bootstrapped OptiPub out of, you know.

And what was the. What were the ESPs we on before you guys built your own?

Oh, let's see. Alterion. I don't even think they exist anymore. I don't think so. Yeah, yeah, that was really old school. You, you know, you literally. Their UI relied on running a desktop application.

Oh.

So after that we went to Blue Hornet. Blue Hornet, they recently shut down, maybe like six months ago or something like that, but they've been around a long time. And we were on Blue Hornet for a while and, and then we came back in house. We were on Blue Hornet for years, basically. Our mail volume kept scaling. We didn't have the insight we needed. Their platform, we kind of found ourselves, kept doing a bunch of hacks, like using custom fields. As we added more and more publications, it became a struggle to manage. You know, at one point we were at like 20 publications, right. So we found ourselves with almost a thousand custom fields, you know, managing all these different variables for all the different publications and. Cool.

So, I mean, you kind of explained it a little bit there as far as why you're switching from the old ESPs and building your. No, your, your own. So what was the, you know, I guess what, what was the big problems you were solving. Outside of the variables and stuff, was there any like, deliverability issues? Is it. Yeah, tracking. Tracking. Anything else as far as monetization, things like that?

Yeah, I know a big, I know a big one was our send time. As our list scaled, you know, into the hundreds of thousands of subs, it was taken 30 minutes plus to get an email out. And so that was one of the big things. When we, when we started building optipub, we really focused on being able to keep our send time under 10 minutes. And you know. Cause they're paid publications in some cases those weren't hundreds of thousands, but you know, they would stack up with the queue. We find our paid mallons will be sitting behind our free newsletters. So some of these things are timely, right? You're sending out, you know, this trader education, so they're sending out like stock picks and stuff like this. So I think from memory that that was one of the big things. And I can remember one specific thing actually. It was just control. We had. We identified a bug in BlueHornet's platform, which was basically when we were syncing data, random data, they would tell us the data was synced, give us a 200, say, hey, this is random work successfully. And they're just losing data. And I mean, we went back and forth with them for weeks and they just kept telling us, this bug doesn't exist. It's on your end. It's your. It's, you know, it's on your end. Until I literally had to write a script, give them the code, and they're like, oh, yeah, okay, this bug does exist. So it's just, it was just a real struggle and kind of frustrated they were.

They were losing actual subscribers.

It wasn't actual subscribers, but it was subscriber data attached to it. Specifically status. Right. So we were, we were updating somebody from active to inactive. And the way we were doing that, where we have, you know, pub code status, you know, a separate field for each publication. And yeah, so people were essentially unsubscribing and staying active on the list, or vice versa, you know, subscribing and staying inactive. So obviously that's a big deal if you're trying to unsubscribe people and. And you know, they're saying, hey, okay, we updated it and then the segment is still including them. So that was, that was the main issue there. But yeah, ultimately I think it was just the whole control thing. That's always been our thing. And even outside of email, just, you know, we've been through some stuff with, you know, credit card processors. Hey, you got to get a new credit card, credit card processor in 30 days, stuff like that. So we've always been. We've learned the hard way that the more control you have over your stack is usually better. Cool.

With that being said, let's talk a little bit about, you know, the infrastructure and deliverability you guys have going on over at optipub. What does your sending infrastructure look like kind of under the hood? Are you guys running some shared IPs or SMTP hookups? Or you got your own dedicated IP sending through MTAs? What does that look like?

Yeah, good question. No, we're fully, you know, we run our whole stack. We own the whole infrastructure all the way down to the IP ranges that we have. Optipub owns those IP ranges. We own MTA's physical servers that we host in our own data center space. We're running E Celerity mta, but going all the way back then, you know, we're out of Baltimore, Maryland, and when we were choosing on MTAs, we were looking at both Power MTA and. And E Celerity at the time. E Celerity won out primarily because of something called adaptive delivery. They had a patent on it, and it's essentially an automation that will listen to bounces coming from MTAs and automatically throttle how many messages they're sending per hour. Right. Which had a huge impact on deliverability. And it was just one thing we didn't really have to think about as much. The MTA actually would take care of that for you. So, yeah, we run E Celerity still. We have a perpetual license. Through them was Message Systems. That was back there in Columbia, Maryland. They were like 20 minutes away from us at a time. So that was awesome too. Then Message Systems got. They ended up building their own esp. You might remember Spark Post. And then that was before Message Bird acquired them. And then they kind of went, you know, it's a European company acquired them and they kind of lost sight of their MTA platform and just went the ESP route instead. So essentially the MTA we were using started competing with us. And yeah, so we're still running Celerity, we're actually working on. And soon we'll be migrating a lot to Kumo mta, which I talked to you about. Some. Yeah. Built by a couple people from message systems, some OGs from message systems, their dev, actually, ironically, he's the guy who built out, I believe, Adaptive Delivery, so. So, yeah, they're pretty, pretty solid so that's our, our plan there is to go to Kumo, but. Yeah.

Nice.

And their entire infrastructure, you were telling.

Me they were open source too.

Yeah, yeah. Well, E. Solarity is not open source. They're proprietary. So that's one of the issues is, you know, we have a perpetual license, but we're still paying them yearly for the licensing. And they kind of stopped like I said. They kind of. Their priority shifted. Right. So they're putting most of their, their effort into Spark post and into the ESP side and kind of, you know, went another direction. So Kumo is, you know, that's, that's open source. So that's going to be a great thing. I think. I think if you flash forward five years, you're going to allow. You're going to see a lot of people on Kumo, if I had to guess.

Got it. So E. You're saying probably they're just more like maintenance mode and they're not really making any upgrades updates and I release anything new. So.

Yeah. Yeah.

Cool. As far as your, your general deliverability philosophy, is it aggressive pruning of lists? You go engagement based routing. Like what are your thoughts on.

I mean, yeah, definitely, definitely aggressive pruning. I mean we definitely, I mean, which is kind of just the same thing as engagement based. I gu. Yeah, I mean, you know, just not sending to, you know, I guess you can kind of say prioritizing quality over quantity, you know, which is even more important as, you know, nowadays where, you know, just no sense in building up your list with a bunch of people that aren't engaged. But you know, so yeah, definitely keeping your list tight and engagement based, I guess.

And then how do you guys support your clients when they are having inbox issues?

Yeah, good questions multiple ways. So all of our clients, you know, they have a dedicated Slack channel with us and, and myself and my team were all in there and go back to October of 2023. Actually a couple months before that we, we started building out an application called Postmaster Plus. So that's integrated with Gmail Postmaster tools. Look, SMDs validity. If your, if your IPs are certified with validity, we also pull down that data. So all of our clients, we give them free access to Postmaster Plus 2. So you know, that's like a compliance deliverability application. So we use that to monitor our clients deliverability. We usually know about they do. So we'll communicate that with them in their Slack channel. If they're ever seeing any issues, they'll obviously hit us up too. But yeah, we, you Know, we monitor their campaigns as well, so by domain. So if we see anything that looks kind of off, I will reach out and ask them, hey, do you changed, you know, you started adding any new names from some new sources or did you change your segmentation? We work pretty closely with our clients on that front.

Cool. The. Let's. Let's get to the juicy stuff here. That was probably one of the first people started like pushing this info out or publishing this stuff more, I guess you could say mainstream, where people didn't really know what was going on. But the good old atomic bomb that Yahoo dropped earlier this year, where, you know, they changed their algorithm a little bit, they did some things, and then caused major delivery problems across all different types of industries. I mean, obviously hit the finance space very hard. Yeah, you know, I talked to. I had actually on my first podcast, Matt Paulson, and we kind of talked about that a little bit. You know, it obviously hit a lot of the, we'll say the affiliate newsletter people, you know, you know, sending out a lot of affiliate offers within their lists, as well as just, you know, regular newsletter publishers just sending out content. You know, I even see it got hit on some of our stuff that was just pure content plays, so. And I know we talked about it and then you did deal with this and we had some good conversations. But can you share kind of that. What that looked like for you guys and how you were dealing with it and then how you guys are, you know, where you guys are at now, how you guys kind of worked your way out of it?

Yeah. Yeah, you're right. It was definitely a big event. I mean, you couldn't really go on any. Any platform and not see people talking about it at a certain point. I mean, and to be honest, I mean, there's still. It's still an issue with some people, you know, early it was. It was down to engagement. That's where we saw the most success, is really tightening up engagement. And, you know, I read your blog post back then, and I know that you recommended you just really, really tighten up and then loosen back out. And that's essentially what we were doing too, with our clients. I mean, we brought them down to even in some cases, like 14 days, you know, engagement, and kind of just saw where they were sitting and slowly, slowly, you know, brought it out. But it was interesting though, because, you know, initially you were just getting all the 400 bounces and, you know, they just weren't accepting your email at all. Um, so, like, I was talking about adaptive delivery. Our MTAs were just backing off and queues started going kind of crazy. I just have millions of messages sitting in queue, retrying, retrying, retrying. So fortunately I, you know, I, I know the Yahoo postmaster, so I was definitely bothering her a lot. She was, she was great. So basically we, I think it took a couple weeks and we, we got through the bounces but you know, then they're accepting the email and putting it in spam. So it was kind of like two separate issues. And that still is going on in some cases for some senders if they're just haven't been aggressive enough on the engagement and just, you know. But yeah, I mean it's, I mean to be honest, there was a lot of theories out there. I think everybody kind of had their own theory of what was going on. Hey, they're, they flagged affiliates or you know, we're, we're sending certain partner offers that they don't like. But for us ultimately it was just about tightening up engagement, you know and I think some of it was post, you know, getting spf, DKIM and DMARC required. They were able to do that because you know, before DKIM and DMARC were total requirements, they couldn't entirely isolate their senders. So once they had that in place, they were able to really to crack that down. But with that said, I mean, you know, some of, some of our clients, they're validity certified and one of the benefits of that is you get Yahoo inbox placement data, like true inbox placement data, you know, not C list based. And we're actually have higher inbox placement at angel than we did before the, before the thing happened. Like, I mean we're averaging like 70 or 80% inbox placement before and now we're hitting the 90s. So we're considering high rep, you know.

Some good moolah for, for the clients. I said good moolah for the clients. If you're inboxing better.

Yeah, I mean it, it, I think it just kind of forced people to clean up their segments. You know, I don't, we didn't really change or like stop like content wise. I think everything from, from my point of view, everything's pretty consistent with how it was before. Just it was really for us down to the engagement and tightening up engagement and just, you know, getting pretty aggressive. Like even, even still staying at 30 days or 60 days, you know, the client I was referencing that's got 90% they're, they're out at 60 days and still seeing high inbox Placement. So I think it's case by case. As usual. There's no like blanket rule. Kind of, kind of depends on the sender, what their volume is, what their list acquisition process, you know, how, how they're acquiring names, who they're getting their names from, stuff like that. But yeah, I mean, did you guys.

Ever stop sending period for an extended period of time, for like a 24 hour period or a couple days to try to reset?

We didn't actually, no. We just really tightened up the segmentation and fortunately, I mean, I think having that contact was, was helpful because, you know, she was, she was, she was able to mitigate and we would see immediate, I mean, difference, you know, in the, in the, the 400 bounces slowing down and then they would kind of fire back up again and we would reach out again and we just kind of, it was kind of just like a loop, you know. And it got to the point where, I mean we don't, we haven't seen any of those bounces in quite some time unless we're warming up a new IP or something like that.

Awesome. And so let's talk about what makes you gu from some other ESPs. How's OptiPub built differently than traditional ESPs like Sail Through, ActiveCampaign or Mailchimp?

So one thing would be we're publication based, right? OptiPub, you know, optimized publishing. But since day one we've, we've been publication based. And kind of going back to the Blue Blue Hornet thing I was talking about, there's no concept of that. Right. So we found ourselves in order to keep track of at you know, like subscription statuses, we were, we were just kind of manipulating custom fields, right? Like we literally just have the pub code underscore status 25 times and that's how we were doing our segmentation by manually matching on field values like status is active or status is inactive. So right out of the gate we. Everything's publication based. Everything's isolated by publication. A publication just being, you know, your list. So I think that's one thing that, that's pretty different. At least back then there's some that have publication based. But yeah, everything for us is publication based. I think that's probably the big difference for us as far as, as far as like what other details are you looking for there?

No, just anything different. I mean, I know. So I, I guess here's a couple things I guess I could bring up that we did have a chat a week or so ago and you kind of walked me through gave me a little demo of the platform. You know, I think a big part of it is the way you guys have your, your monetization kind of integrated within the platform. You also manage landing pages and different things. It's more than just a newsletter publishing thing. I mean it's a full suite of almost a full. To. To be able to monetize your newsletter.

Well, yeah, we call ourselves a marketing automation platform, not an espn. ESP is one of the acronyms that we are. You know what I mean? So, yeah, that's, that's, that's one of the things that like when I said, you know, our partners, their eyes kind of lit up when they saw our platform is the fact that, yeah, you have your revenue tracking directly next to your engagement tracking too. You're not kind of bouncing between systems. So, you know, aldipub is built to run an entire publishing business. Like you said, we have landing pages, we have a whole, you know, text ad system. We actually have an entire recurring billing system that we're porting into optipub still, you know, upsell chains, all of this. We're integrated with some platforms too, like SamCart, because a lot of our clients use SamCart. So we've, we've integrated that. But I mean, yeah, we, we're just kind of focusing on just like the email specific side. But. Right. Yeah, yeah, I mean we, we have really powerful funnel automations. Our customer service dashboard, our customer service agents are logging in optipub to manage customers when they're on the phone. So it's, you know, you could call it a CRM too. MAP is what we really go by. We're a marketing automation platform is what we call ourselves.

So I know a big part of your clientele and that's where you guys came from. The space camera are fin pubs. Are there any other types of niches or styles of mailers that may be using OptiPub?

Yeah, so naturally most of our clients are finpub. We haven't spent a dollar on marketing yet. So like, you know, all of our clients are word of mouth. So we do have a couple political newsletters as well and we also have a couple SaaS applications. So we're not tied down to FinPub. You know, we can run any, any pub, any business that sends a lot of email or SMS or you know, any marketing business, publishing businesses can use our platform. But yeah, naturally, you know, we, most of our clients are finpl for sure.

Got it. And as far as like, if someone moved over to optipub like what does the migration process look like? Is it smooth? Is it very hands on?

Well hopefully it's smooth. Yeah, it's a, it's very hands on too. I mean so like as mentioned we have a Slack channel. So the first thing we do is we give a client like onboarding form and they tell us who they want in the Slack channel. We get them in Slack and excuse me, we do all the DNS management as well. So rather than having them set up their DKIM records, stuff like that, we just have them delegate domain or subdomain to us and we have a lot of automations in place. We have a whole separate another application we call OptiPub Manager that we use to manage our deployments and our sending domains, our MTAs, everything but it's super white glove. We, we help give them a custom warm up guide, we help them trans pull in their templates, you know from, from their external system, help them set up some base segmentation. I mean it's. If you've been an OptiPUT before you could totally do it all on your own but we kind of use it as an opportunity to find and fix mistakes also. So a lot of times before somebody even signing we've already sometimes fixed a platform issue that they've had like, like recently a client we just are onboarding they are running into the unsubscribe issue. Right. So you know, because of Postmaster plus we'll have them set up Postmaster plus connect their Gmail postmaster tools so we can see all their message history and stuff like that or reputation history and kind of go from there. Yeah, it's definitely super boy club and hands on. Absolutely.

And you, I mean you just talked about when with the whole Yahoo issue that happened and you were able to kind of reach out to Yahoo Postmaster as an ESP and then obviously each ESP is different but from your experience as ESP, do you have any other relationships or access to other postmaster thing to help you figure stuff out and deliverability versus you know, just to publish themselves or something else like something that stands out. Like for example I can tell you like I do a lot of email deliverability consulting and you know I'll do some remediation requests and stuff and reach out to the postmasters but I'm reaching out as a deliverability consultant, not an ESP and I know like just for instance on like the Yahoo, you know, center hub contact form thing you can check off at your ESP or non ESP so I don't know if that gives you a different, you know, level of response time and things versus just someone else submitting stuff.

Yeah, it depends on, it depends on the mailbox provider because yeah, to be honest, if it's Yahoo, it's been a while since I've submitted a ticket. Um, just because I have direct access to the postmaster. Hope she doesn't get mad at me for saying that. But you know, on the Comcast side, I know the postmaster there as well. He's a really cool guy. So, you know, I just hit him up on Discord. If anybody tells you they have a direct contact, the Gmail probably fronting, right? The Gmail is pretty, keeps their stuff pretty, pretty closed. I don't know. I mean, maybe, maybe they give us some priority because we're in esp, but I'm pretty sure when we submit tickets we're going into the same bucket as other people. But you know, on Gmail, Yahoo, Microsoft, we get pretty good response times. But yeah, I mean we're, we don't have any other than Yahoo and Comcast. We're pretty much going through the same streams as others. But I think, you know, we've been doing this for years, right? And I, my presence is there, right? So like I'll go to like ESP Coalition for example, and these, these people are there and hopefully they know that everything in OptiPub, as we talked about a little bit, is, is completely isolated and dedicated. Right? We don't operate any shared IP pools. Everything is dedicated. All of our clients are sending enough volume to, to sustain dedicated IPs. Right. We have them all in their dedicated domains. So when we talk about that publicly, so you could probably argue that, you know, they, these, they keep track of ESPs and they know which ESPs are allowing shady centers on onto their platform or not. Right? I'm very well about that. You know, we don't allow cold email on our, on our platform. Right. We, we vet our clients. We don't let just anybody send through optipub. You can't just sign up online and start sending through optipub. You know what I mean? We don't have, we don't need a dedicated abuse department to identify these things because we work so closely with our clients. So I think that stuff shows and naturally I think you're going to get better responses based on where you're coming from in that sense. Right? Yeah.

Nice. The. Yeah. I mean, when, I mean Gmail, you definitely. There's no direct contact to them, but they are the most transparent out of all the ISP providers. As far as like what's happening with your stuff, right. You know, Postmaster Tools are giving you even though they don't have a traditional feedback loop like you would think of for other people so that the complaint is showing up in your, I guess ESP or whatever. But you know, they'll give you the history what your reputation, what they think your domain's currently at, which IP Reputation is currently at feedback loop IDs to show you which campaigns are causing issues and all that stuff. So wish everybody was like that. Be like even like the Yahoo when this Yahoo thing happened. If they tell you like this is.

The problem, be nice, be helpful, right?

Yeah, try to figure it out. But I'm hoping, I mean they built this whole sender hub thing which really doesn't do anything in my opinion at this time. It's just really you're, you're verifying domains that, that you have access to and stuff but there's not much on the reporting side. I mean, I mean they got the feedback, feedback loop obviously. But as far as like your, your reputation stuff, I mean at least Microsoft STNS has given you back IP based ratings reputation that you can use to help stuff out. So I don't know if you've heard anything from any contacts at Yahoo on the dev side or thing if they are looking to anything that they may be coming out with on Sender Hub.

In the future or so. I mean actually just a couple days ago and full transparency that we've been waiting for a while here. But yeah, Sender Hub is basically just the FBL management portal right now, but the data is there. We've reached out and we were told that unless you're sending a billion plus messages a month that's not even accessible to you again with the exception being if you get your IP certified through validity, you can get Yahoo inbox placement data and complaint data. It's not as nothing like you know, gnmo Postmaster Tools. But Marcel Becker, you know he's, I don't know what his title would be over there but you know, I guess like a head product guy, head email guy, he probably calls himself over there. He just a couple days ago told me to keep an eye out on, on Yahoo's blog because it's basically a tease saying something's coming for the sender hub. So fingers crossed there. They've been teasing it for a while but I think something's, something's coming. So we'll see you know, in the next, in the next few weeks or a couple Months what happens there. But I think they, they got to be working on something. You know you can see they just are consolidating AT AT&T merging at Yahoo. Most recently Xfinity and Comcast merged into Yahoo. So I feel like they, just as they, as they take on more of these domains they, they have more and more responsibility I think in my opinion to, to give us that data. And it's pretty clear that's, that's their plan with Sender Hub that right now it's just the FPL complaint loop thing. It was just all they can really offer right now. But I'm pretty positive we'll see something soon.

Yeah, I just actually just looked this morning and all the domains are now migrated over to Yahoo from @t except for one web, the weird one that probably no one even uses. Yeah, everything else SBC Global is there now obvious at Bell style. Everything, everything's over there now. So yeah, they're making up obviously a major play in the email space. Yeah.

The last week or two they've had a couple outages. So I mean maybe it's related, maybe it's not but whenever I see allergies I'm like oh, maybe they're working on something.

Yeah, I'm sure it is. Is that the whole migration of all these other emails and stuff, who knows what they, you know, an accidental hiccup on bad.

Yeah, right, right.

Import or pipe or something that's that they mess up on trying to do something.

Yeah. So keep an eye out on, on Yahoo blog and I'm sure we'll write about it as soon as it happens. On the postmaster plus side, we're just waiting. Hopefully it comes with an API because we're going to integrate it right away. It's be great to have. We already got Gmail and, and Microsoft in there. Yahoo's just kind of a lone child.

To give some of the listeners a little insight. We've been talking about a lot of this, you know, postmaster and you know, reaching out to some stuff. What stuff? Like we said, a lot of, a lot of deliverability is just comes to engagement, managing your list, making sure your list clean, getting rid of people that aren't engaging, only sending to the most active people things, you know, all that stuff, not stuff that you know, publisher or someone can, can handle on their own and obviously if they see like their domain rep, you know, they tighten up their segments and there's stuff you can do. But what are things that happen on an ESP side that an issue that you know A publisher can escalate that. You might have to escalate to Yahoo or something. Like, you're saying Yahoo's not having emails aren't going in. Right. They had the down downtime, an outage. Like, how does, how does ESP handle that?

Well, honestly, just like any other bounce, the recent outage is just a 400 bounce, a temporary transient failure. So the message was like, resources not available. So the way our MTA handles it is it's just going to wait and it's going to retry and then if it gets a bounce again, it's going to wait a little longer and retry again for, for a couple days, you know, until it will actually permanent fail. So unless they have like a really long outage, which we haven't seen, it doesn't, you know, it's just going to delay your messages. You're not actually going to lose any messages. Yeah, I saw some people kind of freaking out when that was happening, like thinking it's going to impact their deliverability in some way because their open rates are going to be down. But no, because they're not getting the messages in the first place. So that's MTA's, you know, email. That's one thing that's great about email. It's super resilient. You know, it's been around for a long time. So these things, these systems are put in place for a reason. So kind of just let the MTAs do their thing. Obviously, you watch your cues because if it happens long enough, your queues might start stacking up. So that could, you know, you could maybe have a kind of cascading issue if you're, if your queues were to stack up too much. But from our, from our point of view, we just kind of let, let the MTA do its thing.

I had a note here to ask you about how do you protect the customer's reputation stuff, but you kind of explained it already because everything, everyone's on their own dedicated ip, everyone's on their own dedicated domain. So, you know, everyone's basically totally isolated.

Yeah, they have their own servers. They have their own servers and queues too. So if one client's queue backs up, it's not a situation where, hey, our queue is backed up. So one client's waiting on another client to deliver. It's isolated across the board for that reason. Right. So that way if one, if one thing goes wrong for one client, we don't have to worry about another client getting messed up.

So on getting into like engagement and Things and things of that nature, you know, after the iOS 15 update and, you know, MPP mail, privacy protection, all that stuff. Obviously open rates are, you know, shaky is not even a good word. They can't really be necessarily validated on, on an initial thing. You have to do a lot more behind the scenes to try to see what's, you know, real open versus a fake open or, versus a, you know, a bot open or a security open or whatever the damn open is. And I've always told people that I don't really care about opens as much, that I'm just using opener, you know, open rates and stuff as a baseline. Obviously every niche is different. Every, you know, list is going to be different. And if you're, you know, you can't say like, oh, you need to have a 60 open rate or 30 open for you. You have no idea. So I've, I've just always been using open rates as, as a baseline of what looks like normal deliverability for, for a client. Right. Like if they're at 45 consistently, then that's their baseline. And obviously if you see something go down in the 30s, you know, there's something's an issue. And then that's when we start looking. Do you guys, as far as your metrics go, transparency wise, do you guys. Is your opens remove the mpps? Do you have things that you, you know, some other stuff that you're doing to try to, you know, I guess codify what an actual true opener could be, you know, based on like an open happening within, you know, one second of a delivery? We know that's obviously not going to ever happen. So kind of removing those people, if they open again, maybe that's a sign that they are true opener. Do you guys have anything in, in that respect?

Yeah, we, we do. We can identify MPP opens. I'm totally agree with what you said, by the way. But yeah, some people go too far and kind of say, oh, who cares what your open rate is? I don't agree with that because like you said, it's, it's kind of like a baseline thing because, you know, watch your, watch your open rates when you're, when your Gmail rep goes down from high to medium or, or medium to low, you'll see it, you know, that it drops. It is a, it is, it is a metric in that sense, you know. But yeah, it's more like a baseline thing. And opens are really hard to detect the bots on. I mean, almost impossible in some cases. Right. So honestly, we don't Do a ton there because in my opinion it's not really worth it. On the click side is where we try to focus, right. And more so on like website visitors and using any other metrics you have that's outside of that to really gauge your, your engagement, you know. But on the clicks, you know, you can, you can identify the bots a lot easier. But you know, some of them, they're, they're trying not to be detected, right. Faking their user agents, they're purposely not hitting it right away. But most of the clicks, they're, they're security bots. So some of them are just straight up tell you it's a bot, like Barracuda, they have a user agent that straight up says it's a, it's a bot click, you know, so we identify what we can. On the qlik side we keep track of data centers. Most of the big ones they'll, they'll make public their IP ranges. So we keep track of all those in our web application firewalls and we'll automatically tag our clicks as they come in as a data center click and what data center they came from, stuff like that. But you know, it's, it's, it's, it's not, there's, there's nothing perfect that's going to identify 100% of the bots, you know.

But yeah, I mean it seems like you guys are definitely on the forefront, at least the top of it going, you know, as much as you guys can do to, to get true data for your, for your clients.

Yeah, we still have work to do. I mean, but it's, it's especially on the click side, it's important and we try to tell our clients, you know, a lot of clients will be using these like one click opt ins, you know, and it's like dude, you're, you're just adding bots to this thing. Assuming that a bot clicked or a human clicked it, you know what I mean? Like they didn't actually subscribe or even worse, you've seen people with one click unsubscribes, you know, thinking that the one click unsubscribe means hey, the link in my email needs to just be one click. And they have bots just following the links and they're unsubscribing people, you know. Right. So it's, you know, you gotta be careful for sure.

As far as, let's talk about some features here. What's the most powerful feature for segmenting, you know, segmentation or automation or engagement.

Optimization within Your platform most powerful feature for segmentation. You know, our dynamic segment tool is super powerful. It's basically a database query builder. You know, you can segment on anything from, you know, the basic stuff like opens and clicks. But we also have full campaign management and like, like you noted, we have landing pages. We host all that too. So you can, you can build chains of landing pages in optipub so you can segment based on, hey, they landed on the promo but didn't click. Continue to the order page. We also have like video hosting through Wistia so you can track, hey, they, they didn't watch the entire video. It's like, let's see. We also have a whole partner management system. So these campaigns are tied back to partners. So you can, you can segment based on which partner slash affiliate this name came on from. So yeah, I mean it's our segmentation super powerful in that sense.

You guys experimenting with any AI?

Yeah, we've been, we've been sitting back and watching too because you know, I feel like a lot of people, they wanted to be the first one to add AI to their title, you know, and just kind of building in my opinion, useless AI features just because they're AI features. And we have, we are working on some cool AI stuff. On the content generation side, we're building out Personas because we're a content management system. Also it's pretty typical that our clients, you know, they have months or years of articles written by their editors. So we're creating Personas around that. So it's actually custom agents that are trained on previous writing by this editor. So you can create content as if it was an editor that's in your system, which we've seen is, can drastically improve the quality of the writing versus just, you know, putting a prompt into to an agent that doesn't have any context. On the campaign management side, we're working on something, we're kind of called smart, smart campaigns that will look at the engagement of subscribers, but specifically what they're engaging in and cross reference that with other content and try to prioritize what campaigns they're seeing based on what campaigns they previously engaged with. So just like, just like a social media platform would if by seeing what you stop scrolling on, try to give you more of that type of content. Stuff like that as far as just on the segmentation side basically, basically being automatically build your segments with the AI prompt. Right. So say for example instead of dragging and dropping, you know, clicked in the last 30 days domain is Gmail, you know, and also visited this order page, you could just say 30 day clickers from Gmail, you know, and it would just basically auto build your segment for you. But yeah, we're doing a bunch of stuff with AI but like we've been kind of playing slow with it and make sure that it actually adds true value before we put it out there.

Nice. So obviously you guys, you know, have a lot of fin pub people and so a lot of the stuff is, you know, driving people to, you know, financial offers and, and all this stuff. So on the monetization side, ads, kitted emails, even affiliate native ads or drops or whatever you want to call them, how do you help balance that without the list or you know, burning out the list or hurting deliverability with possibly using third party offers or not?

Well, I mean we, you know, we obviously monitor their complaint rates and stuff like that. But back to, on the segmentation side we have the ability to basically you can segment based on if they've seen an offer before. So you know, where we can auto rotate content based on what they've been exposed to, which can be big. So you're not just blasting this, blasting the same thing to everybody, you know, as they come on. If they've already seen this offer, if you know that they've watched the video hit the order page multiple times, maybe you shouldn't send them that anymore. So we have that built into our segmentation so you can kind of segment based on what specific campaigns they've seen or even what partner they've seen or even, you know, if you want to get down to it, what landing page they've seen. But that might be a little bit, you know, pretty granular to try to manage that.

But yeah, I think you talked, we talked. You brought this up already earlier. I have a thing here. Do you support revenue optimization tools like dynamic ad insertion, affiliate integration or ad tracking?

Uh, yeah, yes and no. I mean we're, we're working on more of it on the dynamic side but we have a whole, we actually have a whole text ad system in optipub so you can load your text ads in and, and they'll auto rotate or you can drop them into the emails, into your autoresponders or into your websites as well. So on the AI, on the AI front that's kind of what I was talking about where we were, we're working on tracking what they've, what signals we have of things that they, they would be interested in versus not be interested in. And so yeah, we have something called custom content also which are essentially Content blocks. And so, yeah, we're integrating AI into that so that it's basically. And it's not, some of it's not even AI, it's just behavior based, but, you know, making sure that they see an ad for a product that they're not already active on. Right. Stuff like that. Because we might have 20 paid publications and they're going to drop in an ad and odds are a percentage of those people are going to, you know, potentially see an ad for something they're already active on. So I wouldn't call that AI at all, but, you know, definitely optimization.

And so I want to talk about what the future looks like on a few things. So, you know, we're in this big, obviously creator economy. Everyone's creating stuff, everyone has started newsletters. You know, there's, and then you have, you know, what are your thoughts on like Beehive, convertkit substack, you know, and their influence on sender behavior, how that's impacting stuff?

Yeah, I mean, they've made it. I feel like they've, you know, you give them credit, they've kind of made email cool again. You know, they've made it. You see a lot of people creating newsletters that probably wouldn't have had a newsletter a couple of years ago, like, you know, like athletes and stuff like this. But I don't know. I mean, there. Yeah, I don't know what to say about that. I mean, yeah, I mean they definitely, they've definitely made creating a newsletter easier. But I think it's also contributed to our inboxes being fuller too, especially with, with the recommendation networks. Right. You sign up to a list and all of a sudden you just sign up to five lists, you know, so it's, it's flooding the inbox, I think a bit. But you know, and depends, I mean, I guess, you know, you, you sub stack, I think is unique out of those three in a sense because it's, you know, it's more so like they're kind of turning into more of like a, like a social platform in my opinion than the other two. Right. But yeah, I mean, which, which, which.

Which kind of defeat defeats the purpose of having a real newsletter?

Right? Yeah, you know, they're, you're not owning, owning your audience in that sense. Right, right. But you know, it's interesting because, you know, some of them call themselves email service providers, but if you ask me, they're not ESPs. You know, their service provider is a pretty important part of esp. And a service provider, in my opinion, actually Owns the infrastructure and you know, is sending the email themselves. Right. Because you can kind of call them ESP wrappers in a sense.

Right? Yeah, yeah. I mean they all, they're all using some sort of SNTP like sendgrid or mailgun to send all their.

Yeah.

Their emails from and manage all the IPs and things like that. What I, what I find crazy is, you know, I mean just between Beehive and, and ConvertKit, I mean there's tens of thousands of accounts on there on their platforms. Obviously a majority of them are going to be very small, you know, under 500 or a thousand subscribers on their list I would assume. But just that amount of volume of people and accounts, it's crazy, you know, I don't know, we'll see what happens as far as that goes. But I know that's, that's another tens of thousands of people knuckleheads that you got to make sure aren't messing up deliverability. So on their, on their side, as, as the esp, we'll say rapper, like you said, having to manage that whole thing, you know, with all their shared IP pools and all the stuff to.

Make and there's, and they're, you know, they're sending over, at least in some cases like sending over a shared domain too, not just ips, which is interesting, you know, because I know mailbox providers. Well, kind of the whole point of enforcing dkim, dmarc, whatever is, you know, to isolate a sender, you know, so all of these senders sending over the same from domain. I'd be curious, I'm really curious what the mailbox providers think of that and if they like it or if they don't like it. And you know, I've heard, I know that they'll restrict what you can send, you know, like if you're on their share, you can't send dedicated promos, stuff like this or you can only send a certain ratio like one promo for every whatever amount of non promotional content, stuff like that. So it kind of, that kind of goes back to. I'm a huge fan of owning, owning your reputation, right. Having dedicated IPs, dedicated domains, you kind of, you have more say in what you can do. And you know, being the ESP yourself, you also have more say in what you can do. I mean going back a few years, like especially in finpub, you know, part of the, part of the thing is always be on the cutting edge for like investment opportunities, you know. So I know we had people coming to us because they couldn't send email about like marijuana stocks for example, stuff like that, you know. So I think that just all goes back to having that shared infrastructure. They kind of have more or less lenient buy ins and more. Yeah exactly. They have to deal with keeping everybody the whole pool happy.

You know they got, they got slick features though. They built them, they built them good. I mean my audience Bridge Insights newsletters on Beehive but that's a good platform. It's just a small little guy.

Yeah, that's what we say. It's like you know when if you outgrow one of those providers that we're here waiting for you, you know or especially if you're running multiple publications is when, when we have a lot more value because you can easily cross promote between your publications and stuff like this and, and you can see you know all of your metrics across publications not isolated by publications, stuff like that.

And then what. What's one thing you think as an ESP or ESPs in general we'll need to offer in the next two years to stay relevant?

I think you touched on it some aspects which some of them are doing is really getting the whole picture right of being a full platform like having your revenue and engagement all in one and being able to have your automations and as I think clicks are going that way too. You know as the opens and clicks become less reliable it's like you need other metrics to segment off of and trigger automations off of and stuff like this. So I think integrations and the more data you have to be able to segment off of and you know automate it's going to be big, you know.

And yeah I think there's, I mean I'll give credit to Beehive. I think they're doing a good job on the reporting on the subscriber reporting side for and this is, this is more you know just we'll go with like a you know, long form newsletter B2C kind of person. Not someone who's actually really sending out. Not a performance marketer, someone sending out trying to get conversions on sales and stuff. Right. But when I've talked to you know, I mean hundreds of people, you know Mike, so how, how valuable is the subscriber to you? What's the you know, what's the life term value, you know what's, what's the you know, average length of engagement like how long they actually on your list and average you know by source and stuff. And no one, no one knows donkey dick about that. Everyone just freaking clueless you know, so that's the biggest thing. Like, I don't know, just, you know, I got a feeling in my gut that it's working or it's not. I'm like, oh, that's, that's good. But Beehive has that whole cohort analysis. I don't know if you've seen it, but, you know, kind of tells you how long, like based on what, what month they came in, how, how engaged that that person is. You can kind of break it down by source too. You can type in the source and show you that. So I, I think that's a great, great feature. And I think that's something it goes back to like your whole. You just said about tying in the monetization so you can see the value and what people are doing. You know, they'll probably do that at some point too, I would assume to try to implement everything over there. But being able to show the publishers actual value of, of these subscribers, especially when you're starting to spend a lot of media, you got to know that, right? You got to know what sources are performing, which ones aren't. So that's going to be a big.

Yeah, it kind of, it's funny because they're, they're integrated with Stripe and it's feels kind of stripy in there. Their ui, it's like similar reporting features. So it's, it's. Yeah, it's pretty interesting. I think that's a, that's something we haven't really touched on much. But on the credit card processing side, I mean, I personally have dealt with issues with Stripe. There's a dedicated Reddit on Stripe. People having issues just getting dropped like this from Stripe, AI, fraud features, just closing accounts. So I think having redundancy there is crucial. I mean, at least, at least in certain industries, being able to have multiple credit card processors and doing that in a way that's not a total pain in the ass is at least for marketing. Platform is going to be valuable too. That's something that we've, we've just learned the hard way. Like I said very beginning of the talk is it's a scary thing hearing from your credit card processor that, hey, you have 30 days to find a new credit card processor.

You know, the worst is what I used to haze back in the day. I, I actually built a penny option website called Snaglow. And we grew, we grew up pretty big. I was doing well, but I remember one of our payment options was PayPal. And then they would hold a ton of your Money in like escrow almost and I'll let you touch as like a fallback in case, you know, largebacks or whatever. So they have there, so they're, they're holding like $100,000 of cash just sitting there and it's you know, kind of, kind of brings me back to that when you were talking about that we're running to the kind of the end of this. So I have a few quick little questions rapid fire just to see what, what you got. What's your favorite feature you've built at optipub?

Let's see. I think our, our campaign builder is pretty sweet. We're, we're still fleshing it out, still adding feature set to it. But you can visually and kind of like you do an automation or a funnel, you can build your entire life cycle from first touch all the way down to the upsell chain after they purchase. You can visualize this all in one view. You can see the conversion rates. Like I was saying, if you're chaining together landing pages like from a promo to an order page to a chain of upsells, you can see those drop offs, stuff like that. And it's really brought all these different components in OptiPub that have kind of seem isolated all into one view. Um, so you can see all your landing pages, your, your text ads, your lift letters, everything that's promoting the campaign, all the messages, the autoresponders, it's all tied into one. So it's given us the opportunity to really make everything cohesive. And then back to the AI thing when we, you know, I, I've, I imagine just completely automated campaigns here soon that are, you know, auto automatically, you know, whopping out lift letters that are performing better, swapping out ads like I said, giving giving subscribers the right campaign instead of, you know, so you're not just batching Blast is, has been dead for a while but people are still doing it, you know. So being automate these campaigns I think is a big one and naturally we'll.

Put a stamp that calls it AI on there for you.

Yeah, exactly. Yeah, yeah.

Well, what's one thing every newsletter operator gets wrong about ESPs?

Oh, about ESPs. Because I was going to say like as far as it's getting wrong, that that bigger is better. Right. And you can say that about ESP too honestly. Like in our case, you know, like some of these ESPs they just take on investment funding and they're just trying to add clients. Right. They're trying to increase that investor deck maybe for their next round of funding or whatever. So bigger is not always better. Especially, I was going to say, like, on a list, right? You want quality over quantity. And same deal with an esp. Like, having a true partnership can go a long way. So, you know, I think that's probably the best one you can get. You can get more value out of less a lot of times, you know.

And then what's one thing the legacy ESP still don't get, right?

Transparency. Transparency and feedback. That communication loop. Just like, literally the. The stuff I was dealing with, you know, 15 years ago, you know, just partnerships, just really being tight with your clients and. And giving that feedback loop there and. And not just, hey, cool, we signed you on. On to the next. You know.

All right, I appreciate you coming on. I think that was good. I think you got a lot of value there, like, you know, listener, to be able to get some insights on the ESP level from. From esp. So appreciate you coming on. Hope you had a good time.

Yeah, of course, man. Thanks. I appreciate it. Kind of out of my comfort zone. I'm, you know, CTO turned CEO, so I appreciate the opportunity and hopefully I was able to share some helpful information.

Awesome. Hey, I hope you enjoyed that episode of Audience Bridge Insights. If you did, go ahead and do me a solid, click the subscribe button right around here. You can also subscribe to the newsletter. I have a link down in the description below. Now, we have a lot of amazing guests lined up, but until then, keep sending smart.