In this Marketing Over Coffee:

Learn about Customer Data Platforms, Virtual Event Content, Luxury Apple Watch Bands and more!

Direct Link to File

Brought to you by our sponsors: Wix Studio and NetSuite

Customer Data Platform – a single customer record is maybe not required?

What are the 3 functions of a CDP?

Gemini showing up in Google Office and Email

Google stopping continuous scroll

7:18 – 8:47 NetSuite is the number one cloud financial system, bringing accounting, financial management, inventory, HR, into ONE platform, and ONE source of truth.

Toys R Us AI Generated Ad has come a long way since Pepperoni Hug Spot!

Lawn Mowing Death Knight

Chris’ first AI Generated single on Spotify

Wix Studio is the web platform that gives agencies and enterprises the end-to-end efficiency to design, develop and deliver exactly the way they want to!

Salesforce AI CEO on using GenAI for A/B Testing

Using virtual events to fill the content funnel

OpenAI and Reddit – Do we need more sarcasm? Are upvotes the key to better data?

The Acolyte, Speidel Watch Band, Lawrence

Vanishing Comedy Central and MTV

Gen AI Course Updates done: Special Discount on the newest Generative AI for Marketing Course! Hands on excercises to put AI to work for you! USE CODE MOC now!

Join John, Chris and Katie on threads, or on LinkedIn: Chris, John, and Katie

Sign up for the Marketing Over Coffee Newsletter to get early access!

Our theme song is Mellow G by Fonkmasters.

Machine-Generated Transcript

What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode.

John Wall – 00:00
Today’s episode is brought to you by NetSuite and Wix Studio.

Speaker 2 – 00:10
This is Marketing Over Coffee with Christopher Penn and John Wall.

John Wall – 00:18
Good morning. Welcome to Marketing Over Coffee. I’m John Wall.

Christopher Penn – 00:20
And I’m Christopher Penn.

John Wall – 00:22
Summer is here. We’re just getting ready to wrap up. Yeah, actually this is going to go out the week of 4 July. We’re recording the week before, so we’re kind of taking next week off. And I had that as a skip week, but NetSuite wanted to be in your inbox for 4 July. So here we are, and we thank them for their support of the show. Yeah, a ton of stuff going on this week. We have some fun stuff to kick around, some possibly incendiary topics that we want to go and went. Let’s jump in.

First one was an article in MarTech about CDPs. We’ve talked about these customer data platforms where you try and grab all the info. The article said, well, maybe this isn’t the right way to go. Maybe you don’t always need one record for an individual person. Like, if you have somebody, there’s people who are always having different personas. They have multiple email addresses. In fact, they have shared email addresses depending if it’s a work thing. They have their persona of them as a VP of marketing at whatever company. But on the weekends, they’re proclaiming that the Earth is flat and doing all kinds of crazy stuff. So maybe do you want to have multiple records in your CDP? And I don’t know, what’s your gut on that?

Christopher Penn – 01:27
Well, it’s interesting. So the CDP has three functions. We call them the, we literally call it AAA: authenticate, augment, activate.

Authenticate means you resolve the person’s identity so that you know John Wall on Twitter is the same as John Wall on Facebook, on LinkedIn, in your email, just that you know that’s the person’s sort of federated identity. So you authenticate. That’s who they are.

Second is augment, which is where you start bringing in extra data. So on the weekend, John goes skiing, on the weekend, John rides horses and all this stuff. Because that can help inform things like your creative and stuff if you’re going to do really good AI personalization.

The third is activate, to be able to push data to the right downstream systems like a DSP, to say, okay, this ad for our thing should go out on these channels. And here’s the unique identifier that we had that we could pass to an ad system if you have multiple records. A, that kind of defeats the point of authenticate. B, it makes augment very hard because now you’re augmenting multiple records. And C, activation’s a real pain in the butt because you’re like, okay, am I sending the right people to the right list for the right ad? I understand conceptually wanting to have a multi-dimensional profile. Like, yeah, if you have a creative that happens to have horse riding in it is like the main image, then you’d want to know that’s more likely to appeal to John Wall as than, say, Christopher Penn. But you still want that in one unified record. So the idea of multiple records kind of goes against all three main purposes of a CDP.

John Wall – 02:58
Yeah. And it’s funny, my note just was failure to authenticate. So this is a vendor basically, like, trying to spin. Yeah, it’s difficult to do as well. It’s okay if some of the records are messed up.

Christopher Penn – 03:11
That makes a lot of sense. So they’re basically saying, yeah, our software can’t do the job, so just accept it.

John Wall – 03:16
Well. And I think there’s some weirdness. Like, shared accounts is a really weird corner case where. But again, that’s failure to authenticate. Like, you just don’t know which people are on that. So that’s not, you don’t throw the model out just because there’s some corner cases that don’t fit the thing. Okay, so that’s good. We’re on the same page as far as that goes. Another big one now. And I’m like crusty old man.

I’ve got Gemini showing up in the Google Suite all over the place, and I’m already annoyed because it’s like every time I open up a doc, it’s like, “Hey, how about some Gemini stuff over here?” And I’m like, “No, I’m going back to work on what I’m doing.” But no doubt, and you’ve been on the cutting edge of this stuff. How about this additional rollout of stuff? Is there stuff that you found useful in there and kind of what should be keeping our eyes out for as far as this?

Christopher Penn – 04:03
I think it certainly has some useful basics. For example, in Gmail, you can say, “Show me all the emails I sent to this client that I, that this client sent me that I haven’t responded to yet.” Unearthing, like, “Oh, I forgot to do this.” So some pretty basic clerical stuff.

The way I use Gemini within Google’s development and programming environment, so I’m using the raw model itself and not the friendly interface. It’ll be the same with Apple’s AI onboard iPhones and iOS, where they want to make life convenient for a lot of people. And these tools absolutely will. If you’re sitting in a Google Doc and you say, “Hey, Gemini, I don’t know what to write for this week’s blog post.” It will fill that purpose. It’s not how I use it, but that’s how I think how many people will. Honestly, I think that will be of benefit to people because if you’re staring at that empty page or that empty slide deck, one of the hacks of AI is have it create something that’s awful because that kicks your brain out of create mode into edit mode. You’re like, “Oh, this sucks. I can do better.” And then you’re done.

John Wall – 05:07
Yeah, yeah. You jump up. That’s like that old trick that were saying as far as going into user forums, and you put your question in there and then the first thing you do is you get another account and you answer it wrong. And then 20,000 people will come up to yell at that person who gave the wrong answer if you just answer the question or may not get answered. So yeah, that is very true. Yeah, my use case is the opposite. I always already have a bunch of stuff that I need to clean or shape. And so if you have the blank page problem, I can see that and can run with that.

Also from Search Engine Land, interesting Google stopping continuous scroll of results, which I guess doesn’t at first I was like, why would they do that? Because that seems like a great use case. But what the article said was that it’s, they only want to serve up in batches, this idea of like having to continually throw data at the page. And I would see that where in situations where somebody just starts scrolling like crazy, now you’re having to do ten or 12 calls, and they would rather put the brakes on that and manually do that. So I’m seeing that as they just don’t want to be serving as much data. They’re trying to trim bandwidth. But I don’t, what is your, what’s your take on that?

Christopher Penn – 06:14
I wonder how much the AI answers factors into that. Like the more that loads in results said, I wonder if it’s making subsequent calls to their model, which would dramatically increase the cost of running that internally. So if they’re, if they do it in batches, like, okay, here’s the first ten results, and the AI answers will generate an AI answer for that batch of ten and not have to have a streaming like, “Oh, here’s a hundred, 2131, 140 results just keeps coming in.” And the machine’s like, “All right, I’ll keep recompiling this.” And I could see that being very computationally expensive.

John Wall – 06:46
Yeah, right. The server just lighting on fire as people are just sitting there scrolling away like mad. So that makes sense.

Christopher Penn – 06:51
The other thing I wonder is, here’s the thing with Google in particular, it always has to come back to ads somehow. Maybe the continuous scroll isn’t serving up enough ads.

John Wall – 07:04
Yeah, right. No, and that would just making sure you’re really forcing those ads every time at the top. And maybe, like you said, more processing time to retarget them, too. Yeah, yeah. That’s always in the mix for them.

All right. We just want to take a second, though. We do want to thank NetSuite by Oracle for their support of Marketing Over Coffee.

You know, for all of our clients, there comes a point where they get large enough and they’re managing so many systems that you’re just caught up in the bureaucracy of it all. You’re actually spending more maintaining all this complexity. Smart businesses reduce costs and headaches when they get large enough by graduating to NetSuite by Oracle.

NetSuite is the number one cloud financial system, bringing accounting, financial management, inventory, HR into one platform and one source of truth. With NetSuite, you reduce IT costs because NetSuite lives in the cloud with no hardware required, accessed from anywhere. You cut the cost of maintaining multiple systems because you’ve got one unified business management suite. You improve efficiency by bringing all your major business processes into one platform, slashing manual tasks and errors. Over 37,000 companies have already made the move. So do the math, see how you’ll profit with NetSuite.

Again, we’ve seen it firsthand for our clients. Instead of building all these integrations or running batch reports so that you can get inventory and the financials in order along with the marketing and sales stuff, just get it all one platform. And of course, having it in the cloud makes a whole slew of headaches go away. By popular demand, NetSuite has extended its one-of-a-kind flexible financing program for a few more weeks. Head to netsuite.com/coffee. That’s netsuite.com/coffee. Again, netsuite.com/coffee. And we thank NetSuite by Oracle for their support of the show.

We have a bunch of stuff coming back from Cannes. All the ad folks are talking about that. Again, I don’t know. This is one thing we have not been able to work. We need to get the full-on boondoggle going over there, so we get to hang out in France for a week and, like, wander around. And I don’t know, hang out on yachts and stuff or whatever. I don’t understand why we’re missing the boat on that. Well, I know why we’re not missing the boat because we’re not, like, making TV ads, which is where all that stuff goes.

Christopher Penn – 09:11
Toys “R” Us had its big OpenAI debut at Cannes. They debuted a one minute fully generated ad made entirely with OpenAI Sora. And you can tell it’s still definitely AI generated. There’s still little artifact stuff, but it looks much better than Pepperoni Hug Spot did a year ago.

John Wall – 09:32
Pepperoni Hug Spot is still classic, and we will include a link to that because if you have not watched that, it is still fantastic. Yeah, it was that, I don’t know. That was just so weird to me because, I don’t know. The Toys “R” Us brand has just, like, been through so much insane. In fact, they got Red Lobstered. If you don’t know the Toys “R” Us story, they were earlier, Red Lobster was their deal. They just got, like, carved up and sent to side. Yeah, it definitely proves what’s going on with that and what they’re with Macy’s now. I guess they’re in stores as far as their thing. But yeah, it definitely shows that these things can work and can, like you said, when you look at the gain made in a year, it’s insane.

Christopher Penn – 10:07
Exactly. I posted that on Threads, and boy, did that generate a lot of very angry comments, mostly from creative folks saying, “Oh, this is yet another way that private equity is destroying the world. And Toys “R” Us couldn’t find a budget to make this so they had a machine do it instead.” Ignoring the fact that the agency they used to make it probably charged just as much.

But to their point, when you build an ad like that, there’s no set, there’s no props, there’s no costume, there’s none of the folks that work on a set like that. It’s just a bunch of geeks in a server room putting something together. So there definitely is that angle. But the reality is, it made headlines. And now, in a year’s time, it may be, in another year’s time, it may be to the point where you’re like, yeah, you will have to look real hard at a video to tell whether it is machine made or not.

John Wall – 11:00
All right. And. Well, so we have another piece, too, talking about creative jobs disappearing. You had an interesting piece. The CTO of OpenAI was talking about creative jobs, and what’s the story there?

Christopher Penn – 11:12
So on June 19, Mira Murati, the CTO of OpenAI was at the Dartmouth College’s School of Engineering, and a Dartmouth alum asked about creative jobs and AI’s impact and creative jobs. And Murati said, “Yeah, of course, AI can do these things.” And she said, “AI tools will lower the barrier for anyone to think of themselves as creative.” And of course, the alums like, “Well, what about the creative industry, the jaws people that are in the employee?” And she goes, “Well, some creative jobs maybe will go away, but maybe they shouldn’t have been there in the first place.”

John Wall – 11:42
So, all right, how to make friends and influence people there.

Christopher Penn – 11:47
I can see both sides of it because we’ve both been in companies where the creative team is cranking out yet another PowerPoint deck with, the same 58 slides. “Here’s our awards we’ve won. Here’s the lake in front of our building.” And yeah, that definitely is not the height of creativity. Our machine probably should do that because I remember we’re back at the old agency, our creative team hated putting together PowerPoint decks. They’re like, “This is the most useless use of our skills possible.”

But the other side is like, yeah, there’s some stuff that a company that’s focused on profitability above all else, they’re going to pick the machines. That’s kind of a given.

John Wall – 12:30
Yeah. Right now, if you’re making, unique stuff, if that’s part of your strategy or your brand or whatever, that’s one thing. But yeah, for so many brands, it’s like, well, we just need some kind of image.

I did, I have to give an award, too, though, for Threads. I should have saved the link. One fantastic proven use case for generative AI images is for playlist images. And so the guy had his Death Metal for Lawn Mowing playlist, and he had a perfect image of, like, the Death Knight using the push mower. And I was like, oh yeah, all right, this is. Yeah, I give that thumbs up. That’s the right way to play that.

Christopher Penn – 13:09
Speaking of which, my first AI-generated

John Wall – 13:12
singles on Spotify, the Ann Handley song. Is that the.

Christopher Penn – 13:17
No, no, that one’s. That one’s in process right now. But I’ve actually got a catalog of stuff that I’ve been processing and using AI to generate and then if it sounds decent trying it out and stuff. So I’m doing this as an experiment to see what does it look like to publish this stuff. Because, and this is a part where there’s a lot of confusion, a lot of people think that because AI content can’t be copyrighted, you can’t put it up for sale. You absolutely can put it up for sale. What you can’t do is enforce your sole right to sell it.

So, for example, if you take the Mona Lisa, you put the Mona Lisa on a t-shirt, you sell that t-shirt, you can sell that legally. That’s totally fine. But if your next-door neighbor sells the identical t-shirt, you can’t do anything about it.

So if you make a great song via AI and you put it up on Spotify and it gets a million hits or whatever, and then a record label’s like, “Oh, that sounds great. Let’s grab it.” And it’s AI generated, you can’t stop them from doing that because it’s public domain. But you can absolutely monetize public domain content. People do all the time. There’s a gazillion and a half copies of books on Amazon that are, “Socrates and Plato” and stuff that’s thousands of years old, and people are, charge $0.99 for it.

John Wall – 14:28
Yeah, right. They just show that.

It’s funny, I’ve had, because Spotify is like, “Yeah, we want these AI-generated songs because they don’t have to pay for those as part of the stream.” And I’ve had playlists where it starts inserting songs, when you get to the end of your own list, it’ll just randomly match stuff up, and every once in a while I hear our song, I’m like, “Oh my God, that has to be generative AI, because that song is so horrible. Those lyrics are so rotten.” And then I go and dig and look and it’s like, “Oh no, this is really a band. People thought this was a good song.” That’s really unfortunate. So yeah, I haven’t, maybe I am already being fooled by better stuff, but it is just funny how I’m kind of like, “Oh, that’s got to be generated because it’s so horrible.” But no, there are humans who can make horrible stuff.

Christopher Penn – 15:12
I’m working on, like, one release a day because it takes about 14 days for it to go through the process of the system, get up there. So I hope by the end of summer that we have a couple of albums.

John Wall – 15:24
I have a few tracks, get ready to go.

Some other stuff. I had to laugh at this. Salesforce actually has an AI CEO who was doing a keynote, and she was talking about using generative AI for A/B testing and ads. And I was like, “Man, you should just be listening to Marketing Over Coffee. We’ve talked about that for like two years already.”

Christopher Penn – 15:41
So we’ve talked about that for way longer than that, too. Yeah, I talked about that back in 2020.

John Wall – 15:47
But I guess what else are you going to throw out there as the easy one for the ad crowd? It was my take on that. But so yeah, and AI CEO is a position, so there’s something you can shoot for if that’s what you want to go with.

Another interesting one, MarTech had an article about virtual events. And we see these where a company is like, okay, on X day, we’re going to have seven webinars running throughout the whole day. But it was interesting. The take on that was that you use that to fuel your content calendar. It’s not like, “Hey, let’s throw an event.” It’s more like, “Hey, we want to have seven recordings that we can roll over the next seven months. Why don’t we just do them all on a single day and so we can make an event out of it?”

But then the money quote of that was, “We would never, ever run an event without a post-event distribution strategy. It’s all about how we can use the content that we just created.” And that’s just that. I love that because we’ve talked tons of times about there’s all these shows where they go and create all this great content and then the videos just kind of never see the light of day. It’s like, why the heck aren’t you rolling that six months later? I mean, it’s half out of date, but it’s still fantastic stuff that people could get for free if they couldn’t go for the show. Yeah, I wanted to give a quick plug for that.

Another one here, a heavy discussion topic, OpenAI scraping all of Reddit. My first thought with that was like, yeah, we don’t need generative AI to get more sarcastic. Like, that’s not what we’re looking for. But the flip side was this article did mention upvoting as being able to validate and get correct data. I was like, okay, yeah, as much as it’s a rotten neighborhood, upvoting is for real. Like, that is, I could see how people could use that as data quality. So I don’t know, is it worth the trade for the sarcasm?

Christopher Penn – 17:27
I guess it depends. Reddit is a series of communities. It’s not one entity. So there’s, yeah, there are some communities that just crap. They’re, flat Earth and all that stuff. There are some communities where there’s decent content, and it depends on that community’s individual moderators as to how strictly they enforce rules and things.

So the other day, I was getting ready for a webinar I’m going to be doing for real estate agents, and I went to the Realtors community, which is very heavily moderated. There are folks in there who do a good job keeping it clean. So pulling all that data out, it became very straightforward to say, like, these are the things that realtors, actual realtors care about, real estate agents. Okay, the market’s shifting, the pricing of people being priced out of a market, all this stuff. So it really depends on the community.

And I think the bigger lesson there for everyone is that you have to have some source of data. If you want tools like generative AI to do really well, you need to have some source of data, ideally something that difficult to access or is not available to the general public. What do you have? So, for example, we have the Analytics for Marketers Slack group. As an administrator, I can go in there and export the entire thing. No one else except an administrator can do that. But I could then take that to train a model or to build content from that based on all the responses that people have given over the last five years in there. Think about Marketing Over Coffee. Marketing Over Coffee has more than 800 episodes. Yes, it’s all in public, but you would still have to take an enormous amount of time to download them all, transcribe them all, and then process them all to turn them into a data set that you could use. So if we did that, we had that data set, it would be an advantage over someone who didn’t. Like, you could write a book using just the contents of Marketing Over Coffee as a way to have your own data store. So you’ve got to have some form of data that a machine can use because these tools are good at making stuff, but they’re great at synthesizing and reorganizing stuff you already have.

John Wall – 19:31
That’s interesting of choosing the targets because, yeah, like I said, some communities are absolute circuses, but then there’s also some really high-quality neighborhoods that you can get decent stuff out of.

It’s funny, a couple of weeks ago was talking about apps I use for running. I’m always tinkering, trying to find the right magic app for that. And like, the big one for me is running and getting heart rate audio cues. “Hey, you’re slacking. You gotta pick it up a notch.” And to get that in my ear so that I can go. And over in Thread, somebody was like, “Hey, you listed all these apps, you’re actually not testing the Apple Fitness watch app, the one that comes with it.” And the guy was right. The thing is, it’s actually come a long way. It’s actually very good. So that actually could be the magic one.

I also wanted to give a shout-out for the Speidel watch band. I got a brand new Apple Watch band. They’re out of Providence, Rhode Island, a company that’s been around for over 100 years. So that’s kind of cool. I have one of these old-school watch bands that I picked up, and I’ve been enjoying that.

And then going out the door to media recommendations, I have to give a plug for The Acolyte. If you’re a Star Wars fan, if you have not watched that yet, that is very cool. And it’s kind of like a, it’s got a detective slant to it. It’s kind of a whodunit story, which has been great on the media front.

Christopher Penn – 20:45
This is on the flip side. So Paramount, which owns a bunch of media properties, has been removing stuff for a variety of reasons. They removed the entirety of the Comedy Central backlogs, like 20 or 30 years worth of The Daily Show and all that stuff is just gone now. And they removed all of the old MTV back catalog. So everything from our misguided youth is just gone now.

John Wall – 21:12
Really? Do they talk about why that’s all? Because I have heard a lot of people saying just cost-cutting is killing the back catalog because it doesn’t have traffic.

Christopher Penn – 21:21
But anything else, it’s not that it’s traffic, it’s royalties.

John Wall – 21:24
You have less options out there to make payouts on.

Christopher Penn – 21:28
Exactly. Because if you’ve got Kurt Loder’s whatever and people are watching the back catalog of Madonna’s concerts from the ’80s, they still got to pay royalties on that because it’s a performance rights thing. So if it’s gone, then you don’t pay royalties on it anymore.

John Wall – 21:43
Yeah. I don’t know. That’s a tough call. Is it? Because I mean, it’s long-tail stuff. So you’re going to be paying not what you’re paying on the big stuff. And now you’re forcing, I mean, the viewers are either going to go to different content on the same thing or they’re going to leave, which is worse.

Christopher Penn – 21:57
Right.

John Wall – 21:57
So I don’t know. That seems to be a bad idea to me now.

Christopher Penn – 22:00
Well, it’s a bad idea. There is some aspect of cultural significance to that content. Having the early seasons of The Daily Show, Jon Stewart’s first time around, there’s value to that content. But they’ve decided it’s not worth paying for that. But you’re right because if people don’t have access to that content and they don’t find anything that’s compelling in your new content, because I think one of the worries is it cannibalizes viewership from new content, and new content is how streamers judge the profitability of their services. Are they attracting new subscribers, and is the new content getting views? And if you’ve got 50 years of old content, people like, “I’m going to go watch the Tiffany concert from 1989 instead,” your new stuff may not perform as well, which means that you’re going to have a harder time justifying the investment in it.

John Wall – 22:48
Yeah, no, that’s interesting because that does sound like a roll of the dice if somebody’s got three other options to get the new stuff.

All right, well, that’s interesting. I’m going to give a plug for Lawrence, the band. I had found them through Jacob Collier. I’ll put a link to them in the show notes. They had a new album come out Friday, which was really some good stuff.

Yeah, we’ve got 4 July, so I’ve got traveling on the road, but nothing for business stuff. How about for you? Have you got stuff going on the next couple of weeks?

Christopher Penn – 23:16
Mostly just holiday stuff right now. July, I think, for the moment, is relatively quiet, although I did just send out a promotion to my mailing list for keynote talks and stuff like that. And I can already see that you’ve got some stuff on your calendar, so I may not be all that quiet anymore.

John Wall – 23:35
Yeah, and just for anybody doing events, whether virtual or live, your AI deck has been through a ton of revs, and it’s just rock solid. And we’ve already been booking up a bunch of stuff. So if you have a fall event that you want to have somebody show up and give us a ping, but get it done because everybody’s like, “Oh, July’s wide open,” but September, not wide open. September is mostly locked down. And yeah, we got a bunch of shows coming up in the fall, but we’re going to deal with that after the 4 July here. So that’s going to do it for this week. So until next week, enjoy the coffee.

Christopher Penn – 24:09
Enjoy the coffee.

Speaker 2 – 24:11
You’ve been listening to Marketing Over Coffee. Christopher Penn blogs at christopherspenn.com. Read more from John J. Wall at jw5150.com. The Marketing Over Coffee theme song is called Melo G by Funk Masters. You can find it at Musicale from Mevio or follow the link in our show notes.

The post CDP Functions, AI Creatives, Reddit and The Acolyte! appeared first on Marketing Over Coffee Marketing Podcast.