Aug. 25, 2024
Advanced AI Strategies for Marketing to Drive Epic Results: Tools, Ideas, and Real-World Applications
Curious about how AI is transforming the landscape of marketing? Join us as we host Jess Gonzalez, a seasoned marketer with a decade of experience at the intersection of AI and marketing. You'll uncover the secrets behind Jess's innovative approach to content production and curation. From scraping trending topics on Reddit to leveraging a human-in-the-loop system, Jess and her team have mastered the art of creating SEO-optimized content that truly converts.
Discover how to streamline your ABM outreach and content creation with cutting-edge AI tools like HubSpot, Claude, Perplexity, and ChatGPT. Jess shares invaluable insights on automating processes to save time and cut costs, all while maintaining the quality and relevance of your marketing efforts. We also delve into the practical applications of AI in affiliate marketing—think automated image generation, review analysis, and ad creation—all integrated seamlessly into your campaign management. Tune in to learn why marketing fundamentals still matter and how AI can amplify rather than replace your existing strategies.
Jess Gonzalez on LinkedIn
Send us a Text Message, give feedback on the episode, suggest a guest or topic
Visit the Remarkable Marketing Podcast website to see all our episodes.
Visit the Remarkable Marketing Podcast on YouTube
Remarkable Marketing Podcast Highlights on Instagram
Eric Eden on LinkedIn
Chapters
00:00 - Future of Marketing and AI
12:35 - AI Tools for Marketing Efficiency
24:28 - AI in Marketing Strategy and Implementation
Transcript
WEBVTT
00:00:00.100 --> 00:00:07.825
Today we are talking about marketing and AI what is the future of marketing and AI and we have a great guest to help us talk through that.
00:00:07.825 --> 00:00:09.550
Jess, welcome to the show.
00:00:10.111 --> 00:00:10.932
Thanks for having me.
00:00:12.140 --> 00:00:19.347
Why don't we start off by you taking a minute to share a little bit about who you are, what you do, what you've done?
00:00:20.019 --> 00:00:21.224
Yeah, hi, I'm Jess.
00:00:21.224 --> 00:00:22.167
I love marketing.
00:00:22.167 --> 00:00:30.850
I've been a quota carrying marketer for almost my entire career, so I come to marketing with really a revenue and process focused lens.
00:00:30.850 --> 00:00:49.064
I've spent my career in I would say, really fortunate once of a career roles where I've worked with some of the largest media companies, publishers, some of the fastest growing SaaS brands to really grow their company more quickly by growing marketing quickly, with a focus on driving revenue.
00:00:49.064 --> 00:00:52.012
And lately I've been doing a lot with AI.
00:00:52.012 --> 00:00:53.421
I have been just a caveat.
00:00:53.421 --> 00:01:06.209
I've been using AI since before it was a hot thing and before ChatGPT was launched, so I've been in AI for about a decade, but I've been finding some really interesting ways to use it that I think not everyone's leveraging it for.
00:01:07.792 --> 00:01:10.234
Awesome, and you're also a podcaster.
00:01:10.234 --> 00:01:11.415
You have your own podcast.
00:01:12.180 --> 00:01:14.986
Yes, marketers Talking and Marketers Talking AI.
00:01:14.986 --> 00:01:25.941
Where we bring on marketers talking, we bring on marketers of all walks of life and different seniority and experience levels to share their story and what they have found that works for them in the marketing world.
00:01:25.941 --> 00:01:27.786
And then marketers talking AI.
00:01:27.786 --> 00:01:40.936
My co-founder, pavel, and I cover the recent AI news and we are talking to founders of AI Martech specifically, founders who are actually marketers, not just developers, who are building AI tools to sell to marketers.
00:01:40.936 --> 00:01:43.042
So that's an exciting series launching soon.
00:01:44.102 --> 00:01:45.665
All right, we're ready to be inspired.
00:01:45.665 --> 00:01:51.775
Tell us a story about some of the best things you've done with marketing and AI so far.
00:01:52.280 --> 00:01:52.581
Yeah.
00:01:52.581 --> 00:01:59.447
So my favorite use case and I'm going to start with content, which for a lot of people, they're probably saying yeah, content.
00:01:59.447 --> 00:02:03.402
I know there's 50 tools that will make me a million blog posts a day.
00:02:03.402 --> 00:02:04.463
It's everywhere.
00:02:04.463 --> 00:02:09.993
We took a little bit different approach, but we've been using it heavily for content production and curation.
00:02:09.993 --> 00:02:20.951
So, specifically, we had a client that was starting off fresh, brand new website, no domain authority at all, and they wanted to be able to grow and really engage in the startup community.
00:02:20.951 --> 00:02:28.342
Their ICP was founders, ideally founders who are going through incubators or some other early stage funding.
00:02:28.342 --> 00:02:36.051
So they want to get in front of this specific audience and they have nothing on their website to really make them compelling as an authority on the topic yet.
00:02:36.051 --> 00:02:50.772
So we launched a programmatic SEO effort, which means that we produce a ton of SEO optimized content, and we did it all really quickly using some different workflows, processes, tools, however you want to do it.
00:02:50.772 --> 00:03:01.163
So, instead of manually creating 200 pages, we had a system to be able to do it using Airtable, makecom, retool, zapier a couple tools similar to that.
00:03:01.746 --> 00:03:07.002
So we started off with asking ourselves what does this target audience want to engage with?
00:03:07.002 --> 00:03:12.362
And one of the spots that I've gone to for almost my entire career for inspiration is Reddit.
00:03:12.362 --> 00:03:28.866
So we actually set up a process that it would scrape some different subreddits every day and it would see what are the three or four most interesting articles that have been posted in the last 24 hours, that are trending, like what's getting the most traction.
00:03:28.866 --> 00:03:32.862
And so we use Reddit to curate the topic pool.
00:03:32.862 --> 00:03:38.741
And then we took those topics, put them into an engine to get some different headlines out of it.
00:03:38.741 --> 00:03:51.942
So we would take, let's say, 10 topics that are doing really well, let's distill that down into three that we care about, and then for each of those topics let's generate 10 different titles, and then we would have a human in the loop.
00:03:51.942 --> 00:04:06.260
Human in the loop is a term that not a lot of people may be familiar with, but it means that you have an actual person who's taking your AI output and they're looking at it to refine it before putting it back into AI to get to that next step.
00:04:06.260 --> 00:04:14.472
So our human in the loop would go in and they would look and see those different titles and they would say, okay, these eight, I think, are really compelling.
00:04:14.472 --> 00:04:17.002
I like these two, but they're not quite great.
00:04:17.002 --> 00:04:25.791
Let me go regenerate them or make some changes, and then we would take those titles and we would use those to produce content, but here's where it gets different.
00:04:25.791 --> 00:04:29.728
You can't just put that title into ChatGPT.
00:04:29.728 --> 00:04:30.550
You can.
00:04:30.550 --> 00:04:43.889
You're going to get something, though, that really sounds AI generated, and so what we did is we really thought about what is the unique point of view that our client has, what is their opinion and ideas.
00:04:43.928 --> 00:04:49.007
We want to make sure is included, and are there different areas that we know that they play really well.
00:04:49.007 --> 00:05:04.545
So, for example, if you were a company that does, let's say, software development, you'd want to make sure that topic tied into software development somewhere, and it might not always be super obvious, but AI is really good at making stuff up, and so we would tell it.
00:05:04.545 --> 00:05:05.365
Here's our topic.
00:05:05.365 --> 00:05:23.464
We want to make sure that the tone is optimistic, and here's our super long explanation of the brand voice so we can make sure the content actually sounds like it was written by a specific person, and we want to make sure that it ties into the future of software development, and these are tech stars funded companies.
00:05:23.464 --> 00:05:41.790
We want to mention tech stars and maybe find a couple references to these other publications, so we're giving it a really robust prompt that literally took weeks to develop and months of fine tuning, and then it puts out a piece of content that then goes to a person to look at Quick glance.
00:05:41.790 --> 00:05:43.803
Does it look like it's in the right direction, yes or no?
00:05:44.447 --> 00:06:03.685
And if it is, it'll go back into a system where we actually have different AI models talking to each other to provide feedback on the content to each other, and so it's like having a team of three or four copy editors that are coming in reviewing and providing feedback, but it all happens in like 20 seconds and it goes back and forth.
00:06:03.685 --> 00:06:08.353
And then it would spit out this article that sounds pretty good.
00:06:08.353 --> 00:06:16.747
It's 95% of the way there, and we actually eventually added in a step where it would also provide feedback of what it thinks should be tweaked in the article.
00:06:16.747 --> 00:06:26.242
And then that goes to another human in the loop who's an actual copy editor, actual writer, who looks through, makes some changes, make sure that it feels human.
00:06:26.242 --> 00:06:29.548
It sounds human Also, that all the data is correct in there.
00:06:29.548 --> 00:06:31.273
Ai does hallucinate.
00:06:31.379 --> 00:06:39.367
We try and get rid of that as much as possible, but sometimes it will make up its own facts and even put in a citation for it that doesn't exist.
00:06:39.367 --> 00:06:43.601
So they give it a once look over and then it's ready for publishing.
00:06:43.601 --> 00:07:07.141
And so by doing that, we're able to make a system where it's automatically finding trending topics on Reddit and producing 10 to 20 blog posts in one or two days, which means in a month you can put out hundreds of pieces of content that are all unique, all designed to speak to specific keywords and following topics that are resonating with your audience.
00:07:07.141 --> 00:07:17.370
So it's not just using it for that first content production piece, but it's thinking about how can AI give me leverage across what should be an entire content production process?
00:07:17.370 --> 00:07:18.680
And we've taken that.
00:07:18.721 --> 00:07:26.500
We built it for a client, we tweaked it, repeated it, built it for another client, tweaked it, repeated it, built it for ourselves too, tweaked it, repeated it, built it for ourselves too, so we could put out more content.
00:07:26.500 --> 00:07:32.452
And we found that by really putting out this content, it's helping with domain authority.
00:07:32.452 --> 00:07:39.934
So one of our clients went from zero to two in a day and then they've gone up to about 12 within a week and now they're close to 30.
00:07:39.934 --> 00:07:45.528
But more importantly, it's attracting the right audience to them.
00:07:45.528 --> 00:07:54.612
That's turning into inbound requests, that's turning into deals and opportunities, and that's ultimately what we care about is building marketing that drives revenue.
00:07:55.321 --> 00:08:12.105
It's really almost fueling your inbound engine that wasn't existing before the website was built was built and is it working that it's generating this volume of content with this machine and it's getting indexed, it's getting ranked?
00:08:12.105 --> 00:08:13.649
Is that piece of it working?
00:08:13.990 --> 00:08:16.401
Yeah, I will say the bottleneck on all of it.
00:08:16.401 --> 00:08:17.324
There's a couple spots.
00:08:17.324 --> 00:08:23.444
One is we insist on human editors in the process because we want to make sure it's good quality content.
00:08:23.444 --> 00:08:31.370
That can bog it down a little bit depending on the length of the content, and Google doesn't just auto magically index every single thing you give to it.
00:08:31.370 --> 00:08:37.052
So we've had a couple instances where it's taken a couple weeks to get all of the content indexed.
00:08:37.052 --> 00:08:42.894
Typically, we'll see a good portion of it indexed up front at least three or four new pages a day.
00:08:42.894 --> 00:08:46.081
At the low end we had one client that we were surprised.
00:08:46.081 --> 00:08:48.027
Day one we put it out.
00:08:48.027 --> 00:08:59.168
It was a pretty large piece that was really more of a pillar page of content, and the next day they had an opportunity to come in from it, which is astonishing, because sometimes Google does take a bit to publish.
00:08:59.168 --> 00:09:13.432
What I will say is we haven't seen any instances where Google's penalizing using AI generated content, because I honestly think Google can't tell that it's AI generated, because it's not bad content.
00:09:15.200 --> 00:09:30.731
Yeah, I think we'll get to a point one day when there'll be really good detectors all around that'll be able to tell which content was generated with AI and perhaps what percentage of it was generated with AI, but I think the only one who perhaps has that tool?
00:09:30.731 --> 00:09:38.374
There was this scandalous story about how OpenAI has had this detector internally but hasn't released it.
00:09:38.374 --> 00:09:40.404
But because they haven't released it, no one really knows.
00:09:40.404 --> 00:09:42.610
They say that they had this internal tool.
00:09:42.610 --> 00:09:51.683
They could tell for 99%, but probably Google doesn't know, like you're saying, and it does probably take still just like if you were just creating it all manually.
00:09:51.683 --> 00:10:01.134
It does take a little bit of time Weeks sometimes can be 30, 60 days, even if you're producing a good volume of content for it to get indexed.
00:10:01.134 --> 00:10:03.953
That's what I've seen across many projects over time.
00:10:03.953 --> 00:10:08.033
It's never instantaneous, so that probably is still there, like you're saying.
00:10:08.205 --> 00:10:54.673
But I think that's fascinating because I worked for one company just to give the manual analogy next to that, where we did generate a massive amount of business this is five years ago from creating a thousand blog posts that were each like 2000 words long, written by freelancers, and I can't tell you how much brain damage came from that process of humans writing 2000 word blog posts that didn't weren't keyword stuff, that were actually decent, and then trying to get three or five of them a week and eventually, over time, over a period of two years, we got to enough where we were getting like hundreds of thousands of visitors a month to our website that were going into our software free trial from the blog.
00:10:54.673 --> 00:11:05.566
But we spent years of manual labor and a lot of money paying freelancers to do that and it was good, but this is a lot less painful way to do something like that.
00:11:05.826 --> 00:11:07.370
When you think about the cost.
00:11:07.370 --> 00:11:15.154
So, related to this project, we had a client that was doing programmatic landing pages specifically for ABM.
00:11:15.154 --> 00:11:31.832
So they had 40 accounts that they wanted to do landing pages for and the landing pages were pretty uniform, except we're just going to change the company listed on the top, we're going to change the URL and maybe five or six pieces of content we're going to slightly tweak, but we're not trying to index them.
00:11:31.832 --> 00:11:35.732
So we don't care if it's duplicative content because they're really designed for ABM outreach.
00:11:35.732 --> 00:11:36.956
They had a quote.
00:11:37.017 --> 00:11:40.004
If it's duplicative content because they're really designed for ABM outreach, they had a quote.
00:11:40.004 --> 00:11:45.131
It was $30,000 for 50 pages and I was like, are you kidding me?
00:11:45.131 --> 00:11:47.339
For someone to go and build them all and that's had them getting like 80% of the content.
00:11:47.339 --> 00:11:48.403
I was like that's crazy.
00:11:48.403 --> 00:11:50.066
And so they were on HubSpot.
00:11:50.066 --> 00:11:56.274
We rolled out HubSpot DB, made a spreadsheet and we did have to set up the landing page template and the variables.
00:11:56.274 --> 00:12:03.912
But it took an hour and a half to do to build all the pages when the quote was $30,000 before, which just seems ridiculous.
00:12:03.912 --> 00:12:19.589
And if you look at how much sometimes a blog post especially on the B2B side costs, you're spending $1,000, $1,500 per blog post coming out, so I would imagine that's a massive cost savings for a lot of companies to be able to build these programmatically too.
00:12:20.772 --> 00:12:23.668
Yeah, and this belt tightening era, just a lot of people.
00:12:23.668 --> 00:12:25.096
They just are like we can't do it.
00:12:25.096 --> 00:12:27.787
That's where a lot of companies have gone with it.
00:12:27.787 --> 00:12:39.777
So I think personalization is the use case that you're mentioning, that a lot of people are starting to experiment with, in the sense of creating a landing page for each target account.
00:12:39.777 --> 00:12:43.193
To do it manually is, like, really expensive.
00:12:43.193 --> 00:12:53.284
As you laid out To do it automated, you probably still want to check over the work, but it's a lot faster than trying to do it all manually and that's an awesome process.
00:12:53.666 --> 00:13:01.293
I think what AI is really great at that is underutilized by people is giving you options, giving you something to react to.
00:13:01.293 --> 00:13:13.191
So if super boring example you need to figure out meals for a week, it's a lot easier for us as humans to look at our options and say, okay, monday night you can have one of these four things what do you want?
00:13:13.191 --> 00:13:16.025
Easier to pick than to say, okay, what do you want.
00:13:16.025 --> 00:13:26.157
Monday night it's so much easier to make a decision and so using AI or even some automated systems to help you figure out what your options are and then be able to make that decision more rapidly.
00:13:26.157 --> 00:13:30.015
I think that's one of the most underutilized areas for AI right now for marketing.
00:13:31.485 --> 00:13:31.826
I agree.
00:13:31.826 --> 00:13:37.120
One of my favorite use cases along those lines is in using AI as a co-pilot.
00:13:37.120 --> 00:13:43.176
When I'm doing messaging architecture projects, I have it take like the about us for a company.
00:13:43.176 --> 00:13:56.000
I'm like, take this and rewrite the about us for the company in 10 different styles and I will come back with a conversational one, a formal one, an inspirational one in 10 different ways.
00:13:56.000 --> 00:14:07.456
And then it's very interesting to share those options with executives and see which one resonates with them the most, based on their company culture, because it's the same fact set as like their core about us that they had.
00:14:07.456 --> 00:14:13.706
But people are like I didn't think about it that way, and so I think it is really great, as a co-pilot, of giving options of different routes to go.
00:14:13.706 --> 00:14:16.068
Like I said, I think that's a great tip of giving options of different routes to go Like I said.
00:14:16.089 --> 00:14:17.129
I think that's a great tip.
00:14:17.730 --> 00:14:19.511
My co-founder, pavel, and I.
00:14:19.511 --> 00:14:22.274
We use AI as a co-pilot in almost every meeting.
00:14:22.274 --> 00:14:42.528
We have every brainstorming session and sometimes what we find is that it will in the notes, it will pick out things that we didn't realize we talked about, or it'll put down action items where it'll say based on these other action items, you might want to also consider these couple things and it's been really helpful for that, where it's almost like having another brain in the room.
00:14:42.528 --> 00:14:52.857
And on the feedback side, similar to what you're talking about with the messaging, take that messaging, put it in AI and ask AI to tell you who the target audience is.
00:14:53.725 --> 00:15:00.010
I found that I've written some stuff that I thought was like really fantastic and it did not convert and I put it in.
00:15:00.010 --> 00:15:01.375
I was like what's going on here?
00:15:01.375 --> 00:15:08.667
And AI told me that my copy was for a totally different audience and I was like, oh shit, I guess I missed a note here somewhere.
00:15:08.667 --> 00:15:20.471
But I wouldn't have realized it without AI being that kind of totally I would say objective person that's able to come in and give you feedback, not knowing what's in your head for the context behind it.
00:15:21.913 --> 00:15:25.727
So you're doing some really advanced things here in August 2024.
00:15:25.727 --> 00:15:27.910
What's your marketing?
00:15:27.910 --> 00:15:29.232
Ai tech stack.
00:15:30.575 --> 00:15:33.599
Honestly, I love Claude and Perplexity.
00:15:33.599 --> 00:15:35.370
Chatgpt, of course, is in their 4.0.
00:15:35.370 --> 00:15:40.671
But I think Claude and Perplexity ChatGPT, of course, is in their 4.0, but I think Claude and Perplexity are two of the best tools that people can use.
00:15:40.671 --> 00:15:44.097
Personally, I'm a fan of Claude more for writing.
00:15:44.097 --> 00:15:51.697
I think that it produces more human feeling content than ChatGPT, unless you have.
00:15:51.697 --> 00:16:02.739
If you take the time to make a super robust prompt, chatgpt can get you there too, but I feel like Claude, out of the box, will give you a better writing product than ChatGPT will.
00:16:02.739 --> 00:16:09.577
So I'm a fan of Claude for writing and then sending that Claude output over to ChatGPT to give it feedback on.
00:16:10.164 --> 00:16:11.768
I love perplexity for search.
00:16:11.768 --> 00:16:25.395
I feel like it's giving me answers to things that I asked the most, like I asked it the most random question that's not super specific and it somehow knows exactly what I want, which feels magical.
00:16:25.395 --> 00:16:36.076
And then, when we look between the tools, where I think they really become powerful is how we're automating between them and so, using Zapier, retool, makecom to try and find ways to pull it in.
00:16:36.076 --> 00:16:39.174
I'll give an example For my podcast.
00:16:39.174 --> 00:16:56.506
I have an entire workflow between ClickUp and HubSpot and then I'll actually take the topic and I'll take the transcript from the podcast and I'll send it into Summarize AI, which is my favorite tool for video transcriptions and summaries, and it'll pop out a couple of different title options.
00:16:56.506 --> 00:17:04.976
I can take those and send those back to Claude to give me other options if I don't like it and finding ways to really get those tools to work together.
00:17:04.976 --> 00:17:07.869
So Claude Can't Recommend Enough Perplexity.
00:17:07.869 --> 00:17:11.396
Summarize are my kind of three favorite, I would say.
00:17:12.077 --> 00:17:12.739
Three go-to.
00:17:12.739 --> 00:17:17.076
And what about for images or soon-to-be video?
00:17:17.076 --> 00:17:18.404
Any favorites in those areas?
00:17:18.444 --> 00:17:26.621
I'm going to tell you I so I'm going to age myself a little bit here I don't use Photoshop.
00:17:26.621 --> 00:17:30.938
I use Fireworks still, which Adobe sunset in like 2012.
00:17:30.938 --> 00:17:33.787
But I never took the time to use Photoshop.
00:17:33.787 --> 00:17:41.132
But I've actually been using the generative AI in Photoshop recently because it's so easy to use, so that's been my go-to.
00:17:41.132 --> 00:17:43.079
I've played around with a couple different tools.
00:17:43.079 --> 00:17:57.358
There's one that I've gone to when I need to generate something from scratch, where I'm starting from absolutely nothing, but I'm typically starting from an image that I'm trying to edit, and so I'll go into Photoshop for it, or I'll actually use Canva, especially like headshots.
00:17:57.358 --> 00:18:01.931
Sometimes clients have photos that are a bit older and you need to sharpen them up.
00:18:01.931 --> 00:18:06.155
I've been really impressed with Canva's AI options for editing photos.
00:18:07.978 --> 00:18:10.800
Yeah, so is the Adobe tool.
00:18:10.800 --> 00:18:12.883
Is that Fireflies?
00:18:13.663 --> 00:18:15.045
Yes, I think so yeah.
00:18:16.386 --> 00:18:17.694
Have you found that to be decent?
00:18:17.694 --> 00:18:18.638
I haven't tested that one.
00:18:18.819 --> 00:18:21.590
Oh, so I actually made my wedding invitations with it.
00:18:21.590 --> 00:18:35.358
I had a Call of Duty themed wedding, which, for those who don't know which is probably everyone I met my husband on Call of Duty, like all single people in our 30s, and we wanted a Call of Duty themed wedding because a lot of our gaming friends were coming.
00:18:35.358 --> 00:18:57.035
If you search Call of Duty themed party stuff, it's all for, like little kids and it's all army themed, and we wanted something that I would describe as a little more up level, a little more upscale, and I had some different ideas but I couldn't quite get it to work and I spent four or five hours trying to blend images together and just going in and zooming in and erasing pixels.
00:18:57.035 --> 00:19:01.733
Then I was like, let me just give this a try, and I think the video is actually still up on my LinkedIn.
00:19:01.733 --> 00:19:12.613
It took me less than 10 minutes to make our wedding invitations and they ended up being totally custom.
00:19:12.634 --> 00:19:19.077
I wanted to blend a couple images it's two people parachuting in and it says dropping in together and so I wanted to alter them so they look a little more like the operators that we play in the game.
00:19:19.077 --> 00:19:21.185
And it was the best process ever.
00:19:21.185 --> 00:19:22.413
It was the first time I used it.
00:19:22.413 --> 00:19:32.452
It was so easy to use and, as someone again who did not take the time to learn Photoshop because I'm still using the 2010 tools I just didn't want to learn it.
00:19:32.452 --> 00:19:35.622
I found it easy to use, so I think that's saying a bit about it.
00:19:36.451 --> 00:19:41.750
Now I have to know wedding venue, reception venue was it also Call of Duty themed or just the invites?
00:19:42.031 --> 00:19:47.403
Yes, so we turned into a LAN party, which is also a common aging myself.
00:19:47.403 --> 00:19:52.941
But yeah, we had it at our house and then we had the reception at a what do we call it?
00:19:52.941 --> 00:19:57.125
It's a place that has a ton of computers that are networked together to have a LAN party.
00:19:57.125 --> 00:19:58.570
It's like a nerd thing.
00:19:58.570 --> 00:20:00.756
So, yeah, everything was really Call of Duty theme.
00:20:00.756 --> 00:20:04.625
A bunch of our gaming friends came in, so it was a good time.
00:20:05.507 --> 00:20:06.169
Congratulations.
00:20:06.169 --> 00:20:06.690
That's awesome.
00:20:06.690 --> 00:20:20.142
So I actually have been testing mid-journey against Dolly to see which ones I can get better images out of, and I think for a lot of the marketing I do like images for blog posts right.
00:20:20.142 --> 00:20:33.361
I don't need to spend hundreds of dollars for an image, an abstract image, for a blog post with one of the stock photo companies, but I think people can tell that it's AI in a lot of cases.
00:20:33.361 --> 00:20:36.309
But I've actually gotten differing results.
00:20:36.309 --> 00:20:43.423
But with ChatGPT Dolly, you can actually upload an image and have it change it, which is interesting.
00:20:43.423 --> 00:20:45.813
Not quite Photoshop, but you can say take this image and do this with it, and a lot of times it'll it, which is interesting.
00:20:45.813 --> 00:20:54.065
Not quite Photoshop, but you can say take this image and do this with it, and a lot of times it'll do some pretty cool things and I've gotten some pretty good quality stuff out of Midjourney.
00:20:54.125 --> 00:20:54.828
I like to travel.
00:20:54.828 --> 00:21:06.398
I've been to 63 countries and so all the places that I've been to, I had it generate images of all the places I've been to see how accurate it could make them.
00:21:06.398 --> 00:21:13.891
Like does it feel like when I was there, and what I did was to create these really complex prompts.
00:21:13.891 --> 00:21:48.032
For mid journey I uploaded a guide to chat GPT for oh, of here's how to create prompts in mid journey, and then I basically had chat GPT create the prompts for MidJourney that are like really detailed, with like camera settings and all of the details are like four or five paragraphs long prompts, and I created an Instagram account and put all the pictures there and I'm like you know what it looks, like these pictures are from all those places, like it's pretty good If you were doing a travel blog, for example, like it is pretty good for a lot of use cases, I think for marketers and I think it's going to get better.
00:21:48.634 --> 00:21:49.698
I love that use case.
00:21:49.698 --> 00:21:54.175
Here's my question Are you a subscriber to MidJourney's print magazine?
00:21:55.577 --> 00:21:56.019
I'm not.
00:21:56.680 --> 00:22:06.598
It's $4 a month and it's a collection of, I think people can submit their images to it and it'll have the image and then it'll also have the prompt that was used to produce it.
00:22:06.598 --> 00:22:11.942
And sometimes I will see a beautiful image and the prompt is like half the page.
00:22:11.942 --> 00:22:17.578
And there's other times where I'm like that image is awesome and the prompt is give me what a hot dog dreams of.
00:22:17.578 --> 00:22:28.214
And I'm like I never get anything like that when I put random shit in, like random stuff, random shit in like random stuff you have to use.
00:22:28.236 --> 00:22:43.082
The weird thing about mid journey is that you have to use discord to put in the prompts, which is really weird because it's like a gamer chat app and it's a very different sort of experience because when you're putting in prompts, other people are putting in prompts too and you're in a stream watching what other people are doing.
00:22:43.082 --> 00:22:51.045
And sometimes if I'm really bored, I'll just go in there and not put in prompts and watch what people are creating and the prompts that they're using.
00:22:51.045 --> 00:22:55.961
And it's very fascinating because some people are really sophisticated.
00:22:55.961 --> 00:23:08.833
They put in like camera settings and like the background should look like this and they're doing like product photography and I'm seeing like restaurants clearly creating menu, like photos, and I'm like that looks good enough to eat.
00:23:08.833 --> 00:23:09.976
I'd eat the whole thing.
00:23:09.976 --> 00:23:12.442
I'm like out of one to 10, that's 100.
00:23:13.351 --> 00:23:23.417
I'm just like voyeuristically watching Midjourney images, so that, and or the magazine, I think is a great way to get inspired on how to do great stuff.
00:23:23.478 --> 00:23:29.801
I think Midjourney is trying to make more of a community around it and hoping that's the exact use case that you see it, you get inspired.
00:23:30.349 --> 00:23:35.041
We actually built a scraper so that we could do an API call to generate an image and return it.
00:23:35.041 --> 00:24:01.634
We built a little affiliate marketing tool, so essentially it would take trending, it would take high commission products off of a couple of different sites that give you affiliate links and then it would use Google listing for that product to pull in reviews and feedback on it and they would use those reviews and feedbacks with ChatGPT to write the prompt that would go to mid journey to generate the images and the headlines for the ads and then it would automatically push it into Airtable and then push it into Facebook and run it as an actual campaign.
00:24:01.634 --> 00:24:10.182
And we in that process, like we can't have someone like manually sitting here in discord doing this, and so there are ways to make it automated.
00:24:10.182 --> 00:24:17.623
It's not easy and you do lose access to seeing what other people are doing with it, but there are a couple ways to get it there.
00:24:19.550 --> 00:24:31.951
So where do you think this is all going with AI, specifically for marketers and companies using AI in their sales, marketing and their customer journeys?
00:24:31.951 --> 00:24:34.038
Where do you think this is going?
00:24:34.038 --> 00:24:36.653
I'm fascinated to see how this story goes.
00:24:36.953 --> 00:24:39.703
Yeah, so I'm going to give you the most boring answer ever.
00:24:39.703 --> 00:24:42.232
It goes back to marketing fundamentals.
00:24:42.232 --> 00:24:44.945
What I'm noticing is there's two things.
00:24:44.945 --> 00:24:50.991
There's a lot of developers that are building AI tools for marketers who have never done marketing, who don't know what marketers need.
00:24:50.991 --> 00:25:01.195
They're the ones you see that say, you can make a hundred blog posts a day, but they're not giving you anything to make sure that it's the right topic, or they're not giving you tools to figure out your content pillars.
00:25:01.195 --> 00:25:02.780
They're just giving you volume.
00:25:02.780 --> 00:25:12.015
They're like marketers love blog posts, let me give them more blog posts.
00:25:12.015 --> 00:25:12.637
There's no strategy behind it.
00:25:12.637 --> 00:25:15.548
And then, on the other side, you have people that are just wanting to consume everything AI, but they don't actually have a system.
00:25:15.567 --> 00:25:21.671
Let's say that you decide to bring on a tool to help you do outbounding so you can do more cold outbounding with AI.
00:25:21.671 --> 00:25:23.534
What happens when someone responds?
00:25:23.534 --> 00:25:26.920
Does it go into an actual process where your rep is alerted?
00:25:26.920 --> 00:25:28.541
Does it do any scoring?
00:25:28.541 --> 00:25:30.184
Are you qualifying anyone?
00:25:30.184 --> 00:25:37.476
If you're missing marketing fundamentals, all you're doing is amplifying tactics and they're only going to get you so far.
00:25:37.476 --> 00:25:55.916
The people who really win in the age of AI are the marketers who understand how to do marketing well and use AI to help amplify what they're currently doing, not the marketers who just use it as a way to get more stuff done and just put sometimes complete trash out there with it.
00:25:55.916 --> 00:25:59.163
So that's my take Got to know the fundamentals.
00:26:00.611 --> 00:26:01.534
I agree a hundred percent.
00:26:01.534 --> 00:26:07.809
I think that AI is a great tool, but it can't just replace everything you're doing.
00:26:07.869 --> 00:26:10.557
You have to give it at this stage the direction.
00:26:10.557 --> 00:26:17.059
It can be a really great research buddy or a co-pilot, but it may not be ready to be a full agent.
00:26:17.059 --> 00:26:27.442
And I think of the movie the Matrix and the agents just popping in and they take care of the task completely without direction, just popping in and they take care of the task completely without direction.
00:26:27.442 --> 00:26:29.184
I don't think we're definitely not there right now.
00:26:29.184 --> 00:26:42.796
The question is how long will it take for us to go from a stage of AI is all co-pilot to there's AI agents that can do certain use cases and they can do it all the way through or almost all the way through.
00:26:42.796 --> 00:26:49.183
It just gets a human approval like book a trip according to your specifications, book a meeting according to your specifications.
00:26:49.931 --> 00:26:58.643
It seems like we'll get to this point where there'll be agents and there'll be some pain of getting there where it doesn't right, but eventually there'll be some agent processes.
00:00:00.100 --> 00:00:07.825
00:00:07.825 --> 00:00:09.550
00:00:10.111 --> 00:00:10.932
00:00:12.140 --> 00:00:19.347
00:00:20.019 --> 00:00:21.224
00:00:21.224 --> 00:00:22.167
00:00:22.167 --> 00:00:30.850
00:00:30.850 --> 00:00:49.064
00:00:49.064 --> 00:00:52.012
00:00:52.012 --> 00:00:53.421
00:00:53.421 --> 00:01:06.209
00:01:07.792 --> 00:01:10.234
00:01:10.234 --> 00:01:11.415
00:01:12.180 --> 00:01:14.986
00:01:14.986 --> 00:01:25.941
00:01:25.941 --> 00:01:27.786
00:01:27.786 --> 00:01:40.936
00:01:40.936 --> 00:01:43.042
00:01:44.102 --> 00:01:45.665
00:01:45.665 --> 00:01:51.775
00:01:52.280 --> 00:01:52.581
00:01:52.581 --> 00:01:59.447
00:01:59.447 --> 00:02:03.402
00:02:03.402 --> 00:02:04.463
00:02:04.463 --> 00:02:09.993
00:02:09.993 --> 00:02:20.951
00:02:20.951 --> 00:02:28.342
00:02:28.342 --> 00:02:36.051
00:02:36.051 --> 00:02:50.772
00:02:50.772 --> 00:03:01.163
00:03:01.746 --> 00:03:07.002
00:03:07.002 --> 00:03:12.362
00:03:12.362 --> 00:03:28.866
00:03:28.866 --> 00:03:32.862
00:03:32.862 --> 00:03:38.741
00:03:38.741 --> 00:03:51.942
00:03:51.942 --> 00:04:06.260
00:04:06.260 --> 00:04:14.472
00:04:14.472 --> 00:04:17.002
00:04:17.002 --> 00:04:25.791
00:04:25.791 --> 00:04:29.728
00:04:29.728 --> 00:04:30.550
00:04:30.550 --> 00:04:43.889
00:04:43.928 --> 00:04:49.007
00:04:49.007 --> 00:05:04.545
00:05:04.545 --> 00:05:05.365
00:05:05.365 --> 00:05:23.464
00:05:23.464 --> 00:05:41.790
00:05:41.790 --> 00:05:43.803
00:05:44.447 --> 00:06:03.685
00:06:03.685 --> 00:06:08.353
00:06:08.353 --> 00:06:16.747
00:06:16.747 --> 00:06:26.242
00:06:26.242 --> 00:06:29.548
00:06:29.548 --> 00:06:31.273
00:06:31.379 --> 00:06:39.367
00:06:39.367 --> 00:06:43.601
00:06:43.601 --> 00:07:07.141
00:07:07.141 --> 00:07:17.370
00:07:17.370 --> 00:07:18.680
00:07:18.721 --> 00:07:26.500
00:07:26.500 --> 00:07:32.452
00:07:32.452 --> 00:07:39.934
00:07:39.934 --> 00:07:45.528
00:07:45.528 --> 00:07:54.612
00:07:55.321 --> 00:08:12.105
00:08:12.105 --> 00:08:13.649
00:08:13.990 --> 00:08:16.401
00:08:16.401 --> 00:08:17.324
00:08:17.324 --> 00:08:23.444
00:08:23.444 --> 00:08:31.370
00:08:31.370 --> 00:08:37.052
00:08:37.052 --> 00:08:42.894
00:08:42.894 --> 00:08:46.081
00:08:46.081 --> 00:08:48.027
00:08:48.027 --> 00:08:59.168
00:08:59.168 --> 00:09:13.432
00:09:15.200 --> 00:09:30.731
00:09:30.731 --> 00:09:38.374
00:09:38.374 --> 00:09:40.404
00:09:40.404 --> 00:09:42.610
00:09:42.610 --> 00:09:51.683
00:09:51.683 --> 00:10:01.134
00:10:01.134 --> 00:10:03.953
00:10:03.953 --> 00:10:08.033
00:10:08.205 --> 00:10:54.673
00:10:54.673 --> 00:11:05.566
00:11:05.826 --> 00:11:07.370
00:11:07.370 --> 00:11:15.154
00:11:15.154 --> 00:11:31.832
00:11:31.832 --> 00:11:35.732
00:11:35.732 --> 00:11:36.956
00:11:37.017 --> 00:11:40.004
00:11:40.004 --> 00:11:45.131
00:11:45.131 --> 00:11:47.339
00:11:47.339 --> 00:11:48.403
00:11:48.403 --> 00:11:50.066
00:11:50.066 --> 00:11:56.274
00:11:56.274 --> 00:12:03.912
00:12:03.912 --> 00:12:19.589
00:12:20.772 --> 00:12:23.668
00:12:23.668 --> 00:12:25.096
00:12:25.096 --> 00:12:27.787
00:12:27.787 --> 00:12:39.777
00:12:39.777 --> 00:12:43.193
00:12:43.193 --> 00:12:53.284
00:12:53.666 --> 00:13:01.293
00:13:01.293 --> 00:13:13.191
00:13:13.191 --> 00:13:16.025
00:13:16.025 --> 00:13:26.157
00:13:26.157 --> 00:13:30.015
00:13:31.485 --> 00:13:31.826
00:13:31.826 --> 00:13:37.120
00:13:37.120 --> 00:13:43.176
00:13:43.176 --> 00:13:56.000
00:13:56.000 --> 00:14:07.456
00:14:07.456 --> 00:14:13.706
00:14:13.706 --> 00:14:16.068
00:14:16.089 --> 00:14:17.129
00:14:17.730 --> 00:14:19.511
00:14:19.511 --> 00:14:22.274
00:14:22.274 --> 00:14:42.528
00:14:42.528 --> 00:14:52.857
00:14:53.725 --> 00:15:00.010
00:15:00.010 --> 00:15:01.375
00:15:01.375 --> 00:15:08.667
00:15:08.667 --> 00:15:20.471
00:15:21.913 --> 00:15:25.727
00:15:25.727 --> 00:15:27.910
00:15:27.910 --> 00:15:29.232
00:15:30.575 --> 00:15:33.599
00:15:33.599 --> 00:15:35.370
00:15:35.370 --> 00:15:40.671
00:15:40.671 --> 00:15:44.097
00:15:44.097 --> 00:15:51.697
00:15:51.697 --> 00:16:02.739
00:16:02.739 --> 00:16:09.577
00:16:10.164 --> 00:16:11.768
00:16:11.768 --> 00:16:25.395
00:16:25.395 --> 00:16:36.076
00:16:36.076 --> 00:16:39.174
00:16:39.174 --> 00:16:56.506
00:16:56.506 --> 00:17:04.976
00:17:04.976 --> 00:17:07.869
00:17:07.869 --> 00:17:11.396
00:17:12.077 --> 00:17:12.739
00:17:12.739 --> 00:17:17.076
00:17:17.076 --> 00:17:18.404
00:17:18.444 --> 00:17:26.621
00:17:26.621 --> 00:17:30.938
00:17:30.938 --> 00:17:33.787
00:17:33.787 --> 00:17:41.132
00:17:41.132 --> 00:17:43.079
00:17:43.079 --> 00:17:57.358
00:17:57.358 --> 00:18:01.931
00:18:01.931 --> 00:18:06.155
00:18:07.978 --> 00:18:10.800
00:18:10.800 --> 00:18:12.883
00:18:13.663 --> 00:18:15.045
00:18:16.386 --> 00:18:17.694
00:18:17.694 --> 00:18:18.638
00:18:18.819 --> 00:18:21.590
00:18:21.590 --> 00:18:35.358
00:18:35.358 --> 00:18:57.035
00:18:57.035 --> 00:19:01.733
00:19:01.733 --> 00:19:12.613
00:19:12.634 --> 00:19:19.077
00:19:19.077 --> 00:19:21.185
00:19:21.185 --> 00:19:22.413
00:19:22.413 --> 00:19:32.452
00:19:32.452 --> 00:19:35.622
00:19:36.451 --> 00:19:41.750
00:19:42.031 --> 00:19:47.403
00:19:47.403 --> 00:19:52.941
00:19:52.941 --> 00:19:57.125
00:19:57.125 --> 00:19:58.570
00:19:58.570 --> 00:20:00.756
00:20:00.756 --> 00:20:04.625
00:20:05.507 --> 00:20:06.169
00:20:06.169 --> 00:20:06.690
00:20:06.690 --> 00:20:20.142
00:20:20.142 --> 00:20:33.361
00:20:33.361 --> 00:20:36.309
00:20:36.309 --> 00:20:43.423
00:20:43.423 --> 00:20:45.813
00:20:45.813 --> 00:20:54.065
00:20:54.125 --> 00:20:54.828
00:20:54.828 --> 00:21:06.398
00:21:06.398 --> 00:21:13.891
00:21:13.891 --> 00:21:48.032
00:21:48.634 --> 00:21:49.698
00:21:49.698 --> 00:21:54.175
00:21:55.577 --> 00:21:56.019
00:21:56.680 --> 00:22:06.598
00:22:06.598 --> 00:22:11.942
00:22:11.942 --> 00:22:17.578
00:22:17.578 --> 00:22:28.214
00:22:28.236 --> 00:22:43.082
00:22:43.082 --> 00:22:51.045
00:22:51.045 --> 00:22:55.961
00:22:55.961 --> 00:23:08.833
00:23:08.833 --> 00:23:09.976
00:23:09.976 --> 00:23:12.442
00:23:13.351 --> 00:23:23.417
00:23:23.478 --> 00:23:29.801
00:23:30.349 --> 00:23:35.041
00:23:35.041 --> 00:24:01.634
00:24:01.634 --> 00:24:10.182
00:24:10.182 --> 00:24:17.623
00:24:19.550 --> 00:24:31.951
00:24:31.951 --> 00:24:34.038
00:24:34.038 --> 00:24:36.653
00:24:36.953 --> 00:24:39.703
00:24:39.703 --> 00:24:42.232
00:24:42.232 --> 00:24:44.945
00:24:44.945 --> 00:24:50.991
00:24:50.991 --> 00:25:01.195
00:25:01.195 --> 00:25:02.780
00:25:02.780 --> 00:25:12.015
00:25:12.015 --> 00:25:12.637
00:25:12.637 --> 00:25:15.548
00:25:15.567 --> 00:25:21.671
00:25:21.671 --> 00:25:23.534
00:25:23.534 --> 00:25:26.920
00:25:26.920 --> 00:25:28.541
00:25:28.541 --> 00:25:30.184
00:25:30.184 --> 00:25:37.476
00:25:37.476 --> 00:25:55.916
00:25:55.916 --> 00:25:59.163
00:26:00.611 --> 00:26:01.534
00:26:01.534 --> 00:26:07.809
00:26:07.869 --> 00:26:10.557
00:26:10.557 --> 00:26:17.059
00:26:17.059 --> 00:26:27.442
00:26:27.442 --> 00:26:29.184
00:26:29.184 --> 00:26:42.796
00:26:42.796 --> 00:26:49.183
00:26:49.931 --> 00:26:58.643