115 - Personalization at Scale

Rich: 82% of B2B marketing leaders say buyers expect tailored experiences, yet only about half of companies have a documented personalization strategy. So can B2B truly personalize at scale, or is this still just an aspiration? Let's find out. Welcome back to Cocktails, Tangents, and Answers. I am Rich, your host, and Zach, our other host today. I don't know how you match those up. Prost. That's a whole different thing. Yeah. Good to see you again. My favorite times. Yeah. Yeah. Yeah. Oh, yeah. That's a toast. Prost. Prost, yeah. It's Octoberfest. Maybe. I think it's German. Yeah. Is it German? Okay. I don't know. We're going to have to look that up. All right. Well, regardless, as we kind of said in the output opening in this episode, we're going to dig into one of BW marketing's biggest challenges. Um, how do you personalize at scale? And we're not talking about just, you know, it's your name, it's your industry, it's your title. We're talking about actually getting deeper than that and personalizing at scale. Retail does this great because they know everything that you bought. They know what size you are, what colors you like, like all of that stuff. Um, but what happens with B2B? And if you use AI and you let AI take over, do you lose all your personal touch? You know, do you want the human part of it? Mm-hmm. Yeah. Um, that can be really painful for people. Um, so you've got to balance that personalization with authenticity, and we'll we'll try to figure out how we do that or how we can at least get closer as B2B marketers. Um, and then we can uh we can go from there.

Zac: Yeah, awesome. Well, uh, we can get right into this week's cocktail, which is a last-minute mold wine. Uh, this is really cozy and warming. Uh, for me personally, I remember last year around Christmas, I bought like a bottled mold wine that you like warm up on the stovetop. And I got cinnamon sticks and put those in there with some straw and east. And I loved it. It was awesome. And this is perfect for this time of year. Uh, we're kind of getting towards the end of fall, getting towards mid-fall and going into winter. So, especially in the Midwest, the weather's gonna be getting colder. So, a nice mold wine is awesome. This one in particular. Uh oh, go ahead.

Rich: So, the thing I like about this one, Zach, is mold wine usually takes a while, right? Like you put it in like one of those old coffee pot percolators or you put it in a crock pot or something, and you just do it like low and slow, like all day, just and the smell comes through and it's perfect. Last minute mold wine indicates this is gonna be speedier than that. So I'm very curious, like if you forgot and you're like, oh, we were gonna do mold wine, how quickly we can get this done. So, um, our ingredients, we're gonna have one bottle of fruity red wine. So just look for anything that has the tasty notes of fruit forward. Like you're looking for berries and cherries and all that stuff. Uh, four ounces of brandy or bourbon, uh, two teaspoons of sugar, four star anise pods, uh, two long strips of lemon peel or orange peel, or get crazy, do both. And then for garnish, you're gonna do six cinnamon sticks. Um I usually put the cinnamon sticks in while it's also moling, just because you get that extra cinnamon through. So maybe get like eight cinnamon sticks, throw two in while it's mold. Who knows? Whatever you want to do. So red wine, bourbon, or brandy, sugar, the star anise, and lemon or orange peels go into a small saucepan, um, just enough to hold uh all of that stuff. Heat it over low heat until warm. Very important, do not boil it, or you'll boil all the booze off. And then you will not have boozy mold wine. You'll just have uh, I don't know what that is, even. It's just non-alcoholic warm wine. All right, so then you remove from the heat, let it stand for five to ten minutes because it is going to be pretty warm, even though it's not boiling. Divide it between six mugs or punch cups and garnish each with a cinnamon stick, and there you go. So this serves six in mugs. So adjust that accordingly for your taste and how boozy you are and how boozy your friends are. Um, this is a super easy one to like double or triple, um, just get a bigger pan. Um, but yeah, it's I mean, we have an induction cooktop, and that thing will boil water in about 20 seconds. So I would imagine I could make this really fast. Oh, yeah. Which is kind of cool.

Zac: I remember when I made it uh last year, it was like the first week of December, and uh I thought it was awesome, but nobody else in my family likes it. Oh I was like splitting it up between mugs.

Rich: It looks good though.

Zac: It did make the house smell amazing. And I threw a lot of cinnamon in there, and I was like, oh my gosh, like this is awesome. Like, what do you guys think? And they were just like, I guess they couldn't get past the fact that it was warm wine. They like their wine not warm.

Rich: So I mean, but it's it's mold, like so. I mean, it's yeah, I guess, I guess some people are like that. It's weird. They were just being lame, that's my opinion. Yeah, and you can, like, if you do double this and you want to keep it going, you can keep it on like the lowest of low heat, like on a one or a warm or whatever, and just let it keep going. If you've got the time, there are other recipes where you can do it in a crock pot, and then you get that smell all day, which is just really cool. Um all right, very cool. All right, personalization. As my computer fails to personalize my weather location, um, let's talk about how B2B marketers can do that. So we talked about 82% of marketing decision makers agree that buyers really want this to be tailored, personalized. They want to feel like you know them or you know something about them. Um, we talked about that at the beginning, and that's really, really hard. Um, so what about on the buyer side, Zach? How many buyers actually want this?

Zac: So 82% of global B2B marketing decision makers agree. Oh, okay, I see. So set, I'm sorry, wrong stat.

Rich: You can hit the next one.

Zac: Yes. 73% of B2B buyers say they want a personalized B2C-like customer experience, which is interesting. I mean, a lot of people complain about privacy and being like, oh, it's it's creepy when like it's overpersonalized. But I think when it comes to like making purchasing decisions, the easier you can make it for someone and the more personalized, the better.

Rich: So yeah, I get people all the time who are like, you know, we can replace your phone system. And I'm like, we use a VoIP thing through HubSpot. It costs us $5 a line, and we have like six lines, I think, maybe eight. And I'm like, I'm like, you're not gonna be able to like come in with that. So it's like just go away. And you can like you can look and see who we're using for our phone system. Like that information is out there. Um, we can use third-party databases to find technology clients are using. So we know, like, is there what is their website in HubSpot? Is it in WordPress? Is it in Webflow? Those kinds of things. Um so I think that what's really hard about this is um, you know, there's like this promise that, oh, we know you, like we get you, but then there's the reality of the personalization where it's like, you know what you can read on my LinkedIn profile, and that's about it. Um and and there are ways to get more than that.

Zac: I mean, it's it's really surface level, I think a lot of times, right? Like it's high brackets first name. I see you're in brackets industry. Like that's what I think B2B, like a lot of B2B personalization is, and that's nowhere near the level of B2C, right? It's really surface level.

Rich: My favorite is is people can now like so like HubSlide does this, other places do it. Like they bring in an about this company and they scrape the metadata on your website and they pull in whatever your homepage metadata is. And so I'll get things that say things like, um, you know, I see that, you know, antidote 71 is in the blank blank blank blank and offers blank blank blank blank, and it's just pulled from our meta description. Um, so I can tell when like Riley updates a meta description because it's it's new in those emails, and I'm like, oh, we must have updated. Um, but it's just it's annoying because it's like, no, you don't know anything about me. You know how to read my website and look at meta text. Like, whoo, congratulations. That's great. Um, so you know, I contrast that, and I'm gonna tell a couple of Lego stories because why not? I mean, I got them behind me here, wherever they are. Um so there was one vendor, and it was not something we needed at all, but I did give him a I did give him the meeting because he went out of his way. He mailed me the book, The Cult of Lego, like a hardbound book. It's like a $40 or $50 book, I think, at the time, to the office. So didn't come to my house, thank God, because if he had my poem address, I'd be like, what are you doing? Um, and basically with a note about what they did and how to get in touch with him. And so I like I emailed him and said, Hey, that was really great. I do appreciate that you went deeper than just like my name and my, you know, industry and my company name. Um, but at the same time, like we didn't need what he was offering, but he did he needed to do a little more due diligence on the front end before wasting money on me. But I still have the book. It's sitting on the coffee table right over there right now. Um, so that was interesting to me. And that you can get from like, I mean, you can get that from these episodes, right? Like I've got the Lego specs back here, I've got these Legos. So people can see I probably like Legos, I probably like plants. Like there's stuff that you can kind of get into. I mean, if somebody was like sending me Tiffany lamps, I'd I'd take it, but it's like that's just I needed a lampshade for that thing that broke and it was from Target, and I found this lampshade online, so I got it. Um, and I contrast that with like how Lego markets to me, right? So they have a new, it's the very first Star Trek Lego like um collaboration. So Star Wars stuff has been in Lego forever, but Star Trek um Paramount has never offered up that licensing for Lego ever. Um, and so the first thing is gonna be the Enterprise D. It's a big set. I saw that. Star Trek Next Generation. Yeah, so you see it on my story, but that was on my story because Lego sent that to me in email, and Lego like like I saw that everywhere immediately. I also had friends sending it to me and like all kinds of people. So I don't get content from Lego about like Lego friends, which is sort of this weird set they tried to make for girls, and everybody's like, no, Lego, like just regular Lego is good for girls or boys. It's pretty gender neutral, like you can do what you want with it. Um, so I don't get that stuff. I don't get like there's like the the Ninja Go and stuff like that. I don't get those because I don't buy those sets. Lego knows everything about me and what I've bought. Um, and so I get updates on the things I'm most likely to want. That level of personalization and experience is really, really interesting to me if you can pull that and move it to B2B.

Zac: Yeah, because they understand really hard the things that you're purchasing in like your journey as a buyer. And I think a lot of the time what B2B personalization boils down to is someone will download a white paper and then now you serve them this piece of content, which is like a demo. Just because they downloaded a white paper doesn't mean they necessarily want a demo, right? But that's a lot of the time how automation like treats that interaction. When in reality, like I feel like a lot of the time people try person they try automating too much. Like I guess what I'm trying to say is you can still personalize the delivery without personalizing the actual connection you have to the person that you're selling to, right?

Rich: Yeah, okay, so like let's unpack that a little bit because that's a big one. You can personalize the delivery without sacrificing the relationship you have with the person. I think that's what you said.

Zac: Yeah, and I think I know you've brought this up in past episodes. Understanding the actual like uh your actual like customer base is and like basically like uh creating those like ICPs is super important to tailoring their experience and personalizing their experience. Because if you're treating everybody Yeah, go ahead.

Rich: Like I say, for non-marketing people, ICP is ideal customer profile. It's like a persona, but it goes deeper. So go ahead, Zach. Sorry, I didn't mean to interrupt you.

Zac: No, no worries at all. I think uh customizing that if you're customizing the journey for j if you're just customizing one journey as opposed to customizing for those specific ideal customer profiles, you're missing out, right? Like you need to understand that specific customer to be able to to personalize your product or service to them. So pr just because you know their name and you know their job title doesn't mean you know how they're gonna ri go through the buyer's journey. Right.

Rich: And even if you know what they've done, right? So, like I viewed this website, I've read this blog post, I've downloaded this white paper. Um, those are kind of those cues that we try to use, like my purchases. Like Lego knows almost all of these. I got a couple off eBay because they were old sets, but they know almost all of these, and they know that I buy them when they come out. Like, you know, the day it's released, I buy one or two. It depends. Um, and so like getting those in front of me and letting me know they're coming is really important to them. And they've got, I'm not unique in that. They've got other people that are in that same, you know, genre. So they are hitting us kind of at mass with that. But um they get, you know, it just it just gets into this deeper behavior of my journey, and which might be different from somebody else's who buys the same things, but they buy them three months after they're released, or they only purchase at holiday, those types of things. So I think that when we try to turn our signals from our B2B like clients or prospects into like the equivalent of a product purchase, it's where you really fall down with this. Um so, and I think that that's part of what happens is when you're using AI, it can use those things, like right, it's gonna know those things already. But what's that extra depth that you give it? What are the other things in your CRM that tell it who that person is that it can use to um to talk to them and to draft things for you? Um I think because that's really where it's at.

Zac: Yeah, I think a big note there is that the automation that you're like implementing and the journey you're creating needs to be human-led and not just like everything's automated to make it as easy and you as on you as possible. There needs to be a reason behind everything.

Rich: Yeah. And we'd love to just like flip on that personalize button and have everything that goes out be hyper-personalized, like down to the individual, right? And and we'll get there. Like we'll get there to the point where that's like 90% pretty good. But the number one thing that it's gonna depend on is your data. If your data is garbage, the AI is going to spew all kinds of garbage to people, and you're gonna lose them so fast, it's not even gonna be funny.

Zac: Yeah. You need to understand that data really well. Things like the customers' pain points, things like basically you need to have is give them, give it as much good data as possible.

Rich: Yep.

Rich: And you can aggregate that, right? Like you can group people, like we talked about the ICPs, ideal customer profiles. Have four to six of those, and most of your people will fall into those. But, you know, do you have some sort of an area where you've got a how this person deviates from the ICP or something unique about them? You know, do each of your people in your database have a tone associated with them? Like if your tone like ours is like friendly and loose and comfortable and like we're just crazy. Um, but if we get somebody who's like, like we do stuff in finance and insurance, and they might want something a little more formal and buttoned up, um, or they want more facts and details, like we would need to note that somewhere. Um, and I know HubSpot's working on kind of there's updates. I've seen a ton of updates in the company profiles and in the contact profiles for things that they're putting in. Um, but that's where that human side really comes in is what do they need to know? What does the AI need to know to personalize this? And then the bigger one for me is how do you structure that communication? So you've got certain pieces that just have to be what they are, but you've given, you've given the AI enough flexibility to personalize it and attempt to relate to that person through what you know, but you've also put rails on that so that it can't go like completely crazy.

Zac: It's complicated.

Rich: It is. I'm waiting for some some retail to use AI and have it start coming out with like 80% off offers, 100% off off, like really weird stuff just to try because somebody forgot to put a rail on it that said, you know, do not ever offer more than 30% discount or something like that. And so it's just going to keep trying to get that person to buy. And I mean, if you give it to me for free, I'll probably buy it. Like that sounds great.

Zac: And the more you automate, the more you risk losing your brand voice. And like you'll probably struggle with things like tone, timing, and empathy, right? Because those are hard to replicate, like with AI. And I think it's a fine balance, right? Like, how do you find that balance of things that are automated by AI that makes sense? Will you still keep that like human connection where it feels like I'm receiving a message from a real human being, not a bot that is replying that is that saw my meta description and wants to sell me something, or that saw me download something, right?

Rich: Yeah, I mean, I have so I have an AI uh agent in my inbox, uh, my email inbox. Um, and most of what it does is it sorts things, it triages for me and puts them where it needs to go. But some certain things, like it it assumes that if I'm CC'd and not in the to field, that it's just an FYI and it puts it in an FOYI column. Well, humans don't use the CC field like that. And when you reply, whatever email program you're using adjusts who's in the CC and who's in the two. And so it may be off on that. Um, it hasn't gotten to where if I'm mentioned, like if my name shows up in that email but I'm only CC'd, that it actually kicks it over to the, you know, for review or for response or response needed because I'm in there. I think it will. But what's funny for me is it always drafts like these replies for anything it thinks needs a reply. And I read them and the information is decent most of the time. Um, but the way that it's saying it, it's just so off on my tone. Like nobody would believe. Like I actually used it with somebody, it's a person who's a client and a friend, because I wanted to see. I just took what the AI gave me. It was accurate, it was right, and I fired it back to him. Three seconds, I get a text message. Um, did your email get hacked? And I'm like, no. He's like, is AI writing responses for you? And I'm like, wow. And he's like, yeah, he's like, you just wouldn't say that. And I'm like, yeah, I know. I said that's why I sent it to you. I wanted to see. And he's like, he's like, super obvious, dude. Like, don't like, and I'm like, no, I usually delete them or rewrite them or whatever. Um, and it's it's not learning either. So I need to really talk to that company and figure out why. So those are the things where you don't know, even HubSpot says, like, turn on some of the AI features where it's you know writing these emails for you or pulling the other personalization. Don't let it send anything. Go put human eyes on that thing. You know, humans at the front giving it data, programming it, figuring out how to prompt it and how to make sure it's got what it needs, humans at the end making sure that it's accurate and giving it some kind of that, you know, bullshit test and seeing if it works. Um, and then, you know, there's plenty of things that AI can do in the middle. Um, I think that that's fine. Um so it's this one's gonna be so hard. So hard.

Zac: It's hard because I think a lot of people are struggling with that balance, right? It's like you said, like it's easy to tell when like a message isn't authentic and it's just like AI. I mean, I've gotten emails that have hi first name. Like that's like the worst it gets, right?

Rich: That's bad database. Like, that's a bad database. Or, you know, hi Zachary Hazen. It's like, okay, nobody calls me that.

Zac: Like, you know, yeah, I get some really I mean if you really want to like catch my attention, you'll say, Hey Zach, because I mean I don't even I don't know there's a lot of people that'll call me Zachary without knowing me, but like if you like go on my like social profiles and stuff, usually it's Zach, not Zachary.

Rich: So same. I get like when I get something from a company and it's like, you know, dear Richard, it's like okay, like you have you know nothing. Because there's peace, so people I went to like elementary school with and high school call me Richard because I didn't really like shorten it until I got to college. But even ever since then, and even then, most of those people that I'm still connected to have adapted and shortened it to just rich. But it's one of those that's just a really key element. Um, and that's where like if you're doing B2B marketing, you've got to have like first name, last name that might have come with their credit card or their invoices or whatever, but you've got to have like preferred name or nickname. And then you've got to teach the AI to go for the preferred name or when to go for it. Like if it's a contract, use their formal name. If it's, you know, just an email checking in, use their preferred name. And that's where we talk about like AI prompting is going to be a huge job industry moving forward. Like, um, you know, yes, AI will probably take jobs. You hear big companies like Salesforce and Amazon and Microsoft and Facebook talking about how they're getting rid of tens of thousands of people to, you know, and replacing them with AI. Salesforce was the most blatant about it. Um and yeah, that's gonna happen. Like it happens with every technology, right? But um it's it's going to, it's still gonna take that human on the front end. And then there's all like like quality control engineers. Like it becomes builds this whole industry around AI that people can adapt their skills to, but that's a whole other episode.

Zac: I was gonna say, for now, if you're going to use AI, I think you still need to check everything, right? I I think we said that a lot. You can automate things like sequences, but maybe like personally actually write the outreach messages. There's probably some messages you could automate that are like that would would already have been automated, but if you're outreaching to like a warm lead or somebody that you really want to convert, then personalize it yourself. Like you can use a starting point that AI gives you, kind of like you were saying with your uh automated like agent inside of your inbox. If you're actively checking those things and like reworking them to be more personalized and human, I think that's okay. But uh you can also like automate things like reporting, but like actually manual reviewing it so that it makes sense to like what you're doing, and like the data is like good and makes sense, and it's not just like useless.

Rich: But uh yeah, it's it's hard because I mean so so, and part of what we're talking about is that loop marketing that HubSpot rolled out, right? Because it part of that is hyper personalization in there. So this is fairly new trying to get this done. Um, we're in the personalization beta inside of HubSpot. We have not like actually executed anything with it. We're still playing around. Um, so there'll be more to come on this, but I think the the biggest watch out here is if you think you can just turn on a personalization button with AI in your marketing marketing um automation tools, it's probably gonna backfire on you. So be careful. Um, the other one is you know, the number one thing you can do right now to be ready for hyperpersonalization, clean data, clean data, clean data. Get your CRM database like super cleaned up, get as much in there as possible. If you're not pulling in people's LinkedIn profiles or, you know, public, you know, social media that they've got out there, start gathering those because again, AI can go to a LinkedIn profile and see their latest update. Um, those types of things. Like you can use that later, but like you've got to have the info to be able to do it. And it's not creepy because you're just pulling public stuff. Um the the other one that I think is pretty good is using it to pull like um background information. So like I feel fairly confident in the sources we have that I can pull in technologies a company's using and I can see if they're using HubSpot and WordPress, I could draft a communication to go to those people. And what I could use AI for is anytime a new data, a new company gets enriched and has that in there, enroll them in this and send them these communications and pull this specific information in. So that's a pretty good use.

Zac: Yeah. And I think kind of what we I think a lot of people think about personalization as who, right? Like the demographics of who that person is. But I think what it's really moving towards is why, the intent behind the person and what what they're doing, right? So 100%. And that that all comes with better data, right? Like if you have better data, you can understand the why. It's one thing to know who, it's another thing to know why.

Rich: So yeah, and I think I go, I always go further than why. I go to so what. Yep. So this is what you can do. This you know who I am, this is what you can do for me. This is why you think I need it. So what? Why should I choose you? Why do you matter? Um, and I think that like that depth of why and so what and why do I care, that piece of it is gonna get that's where the magic happens because that's where you're hitting a pain point, you're hopefully solving that pain point, those types of things.

Zac: And I think, yeah, I was gonna say focusing on like industry pain points, like trends is a good way of like showing you understand the customer, especially if it's timely. Like if something happens with like a new like HubSpot software, sending out an email to like our existing clients with like, hey, this is really important. Like we're sending this email out so you know all the changes that are happening within HubSpot. Boom, that's like a great example of personalization.

Rich: And I think part of that is knowing like what are they using in HubSpot, right? Because we have clients who use Sales Hub only or sales service and marketing or marketing only or marketing and content. And saying, sending them a um, you know, if I'm using Sales Hub only and you're sending me something about marketing studio, uh, you'd you've lost me. You'd I don't care because it has nothing to do with it. So um it's actually a place where we could take our HubSpot tips that we send out, um, I think it's every month, right? That goes out. Um and that personalization, just a very simple one would be look at what they've got in their in their hub profile on which tools they're using in HubSpot, and then send them information on those. But then on the on the inverse, if you're going to send a Sales Hub person something about Marketing Studio, you've really got to get into the why now, right? Because we know you're not using it. Acknowledge, like, you know, we realize you're only using sales today. Here's how Marketing Studio can help fuel the front end of your sales efforts with Marketing Hub. And it becomes this kind of upsell cross-sell.

Zac: I think it's and I think that's important to like understand too, is like you're not exploiting the data that you have, you're anticipating the needs of your clients and your potential clients and leads. You're using that personalization to be empathetic to their needs, right? Like us sending out like a HubSpot tips every like month to the people that we know are using this specific site of HubSpot are because we want to help them, not because, you know, not because we like are like trying to be creepy with it.

Rich: Yeah, we're not selling anything. We're just like we go through and read all that, and there's a ton of stuff that's coming, and we've got a bunch of people who look at it and we share things in Slack and whatnot. We're just trying to be helpful by digesting it because they may not have seen this feature that they really want. You know, and to that point, there is that piece where the thing that Caitlin talked to a client about that they really want, but HubSpot doesn't do yet, comes across in a beta. It's hard for AI to pull that together. Is it possible? Probably. But for her to be able to make that connection and just shoot that over to them and say, hey, I enrolled you in this beta. Here's why, here's what it does. You know, let's meet and talk about it. Here's my calendar link, that kind of thing. Could AI do that? Probably we could get there. Um, but it's gonna be right now faster and easier for Caitlin to fire that over and make sure that it's right. So I mean. It's a lot. Personalization and hyperpersonalization, it can go south really fast. And if you've got bad data, you're screwed.

Zac: Like that's just I mean, I think the biggest point you can take away from all of this is don't be creepy. Understand understanding your audience with good, clean data is really important. And the why and intent behind their purchasing decisions. And also just don't lose the human element to anything you're doing. It's easy just to set up a workflow and fully automate everything, but if you're not personally curating those messages, and like again, adding that human connection, it's going to be bad. And it's it's not going to be a good reflection of who you are and your brand voice.

Rich: Yeah. And I think, you know, as we kind of wind up, um, if anybody like completely disagrees with us, if you think we're full of crap, um, we'd love to hear from you. Like, we love debating and talking about this stuff. That's how things move forward and push forward. Um, we're, you know, cautiously optimistic about how AI can help in this area and personalization. Like I said, we're exploring it. Um, but a lot of that isn't it's not a magic bullet. It's not just ready to be turned on and you know, let it run away with it. So um definitely uh Zach, I'll wrap up and let you know how to get in touch with us and definitely do.

Zac: Yeah, you can find our agency at antidote71.com and all of our socials are there as well. If you have a question you'd like to send our way and debate us, we're always open for debate, like Rich said. Head to CTAPodcast.live to shoot us an email. Even better, if you want your voice to be on an episode of a podcast, you can hit our hotline at 402-718-9971. Your question will feature into a future episode of the podcast. We'll also take uh, you know, like I don't want to say aggressive comments, but if you d disagree with us strongly, we'll definitely like be willing to hear you out and talk to you.

Rich: So yeah. All right. Sounds good. I think I mean that's that's kind of it. Like, we'll be curious to see where this is going. I'm sure we'll revisit this in a few months. Um, and after we've you know played with it a little bit more and we'll we'll see what's going on. But uh until next time, good to see you, Zach. Good to see ya.