In this episode of The Print Report, Deborah Corn and Pat McGrew discuss the adoption, impact, and responsible use of AI in the print industry, the potential benefits and challenges associated with AI-generated content, and how to integrate AI successfully into print businesses. (Transcript and PDF download below)
Mentioned in This Episode:
AI Can Help You Ask Better Questions and Solve Bigger Problems: https://hbr.org/2023/05/ai-can-help-you-ask-better-questions-and-solve-bigger-problems
4 Tips for Transforming Your Customer Communications With AI: https://www.broadridge.com/article/4-tips-for-transforming-your-customer-communications-with-ai
Project Voice: https://www.projectvoice.ai/
Pat McGrew: https://www.linkedin.com/in/patmcgrew/
Deborah Corn: https://www.linkedin.com/in/deborahcorn/
Print Media Centr: https://printmediacentr.com
Project Peacock: https://ProjectPeacock.TV
Girls Who Print: https://girlswhoprint.net
Print Across America: https://printacrossamerica.com
TRANSCRIPT (PDF download)
[00:00:01] DC: Today on The Print Report. Artificial intelligence, the good, the bad, and the ugly.
Welcome to The Print Report with Deborah Corn and Pat McGrew. All the print that’s fit for news.
[00:00:17] DC: Hey, everybody, welcome to Podcast from the Printerverse, more specifically, The Print Report, with Deborah Corn and Pat McGrew. I am not Pat McGrew.
[00:00:27] PMG: Which means I am Pat McGrew and I am not artificial intelligence.
[00:00:32] DC: Oh, actually, I only know that because we’re on video. But who knows, in a couple of months.
[00:00:36] PMG: This is true.
[00:00:37] DC: I might be able to substitute you with artificial Intelligence. I’d rather substitute myself. Well, quite frankly, do a lot of podcasts. But okay, we are actually going to have a serious conversation about this, and I want to preface by saying that I know that a lot of people are talking about this and it can seem overwhelming. But it is really important to pay attention to this technology because it is going to affect so many things in our lives.
With that being said, the other day I received an email from Guy Kawasaki, who shared an article from May 26th on the Harvard Business Review website, and it was titled, ‘AI Can Help You Ask Better Questions and Solve Bigger Problems.’ It was authored by Hal Gregersen and Nicola Morini Bianzino. Sorry, Nicola. I wanted to read the first paragraph to get us started in this conversation.
“Just a few years ago, businesses wrestled with artificial intelligence mainly in the abstract – a ‘future of work’ problem they’d have to contend with down the line. Now? More than half the companies around the world are actively adopting AI. Although investments are particularly high in industries such as health care, data management and processing, cloud computing, and fintech, all types of organizations and functions have incorporated AI technology into their operations. And generative AI tools such as ChatGPT are forcing leaders to ask where and how AI can help their businesses.”
Patricia, I thought that that was a perfect place to start our conversation.
[00:02:38] PMG: It is because, in the end, AI has actually been with us for much longer than most people realize. We don’t always talk about it, but it’s been around for a long time. One of the things that you have to realize is that we’ve been talking about artificial intelligence since the eighties. It’s been around for a really long time.
If you go back and watch Twilight Zone episodes, if you actually go back even further, think about Star Trek. Think about the Star Trek command environment. What do they do? They talk to the computer, right? How do you talk to a computer? That is a process of artificial intelligence and machine learning. It’s its cousin.
So, it’s not surprising to learn that businesses, especially multinational large, global conglomerates, have been investigating the value proposition and the execution of artificial intelligence for a really long time. What does artificial intelligence really mean? In the end, what you’re really talking about is building a set of programs that accept input, learn what that input means, and learn how to interact with databases that have more input, which means that it can learn, right? That’s what we refer to as machine learning. It is basically writing programs that can make leaps between two data points to come up with a net new data point, right?
That’s what artificial intelligence is. It’s the process of taking a pool of data and analyzing it in such a way based on a prompt that it can return information to you, you might not have seen in that data. It does have all of those applications that the Harvard folks mentioned in their article. The thing is, remember that Harvard Business Review article, which I love, it’s a really great article, but it’s just one of just dozens and dozens that are in that AI bucket in HBR, and there are all sorts of really interesting thought processes that researchers are going through right now trying to figure out, how do we keep artificial intelligence under our control? I guess that’s part of the conversation, right?
[00:05:05] DC: It is part of the conversation. I think without being Debbie Doom here, it is a little alarming to me that the people who let this loose in the world are now running around begging governments to regulate it and pay more closer attention, signing petitions, begging, literally saying that, “We could have a really big problem here if the artificial intelligence starts getting their ‘intelligence’ from unreliable places.”
[00:05:42] PMG: That already is. I mean, frankly, that’s not a going to happen. That is a, it’s happening today.
[00:05:49] DC: Yes. But I mean, let’s look at all the bots that are out on social media. What if those bots were instead creating websites with the same propaganda on it. I just wanted to point out that it’s –
[00:06:00] PMG: But they are.
[00:06:01] DC: They are. But I’m just saying it hasn’t gotten to the point where it’s overriding the system yet. But all the people, this is what the technologists are really afraid of, that when you start to let’s say look up the history of the United States, it’s going to change.
[00:06:20] PMG: Yes, and it already has in places. I think the thing to realize is that we have a tendency to look at these kinds of technologies through the lens of the world we live in, right? We live in the print industry, we talk a lot about print. We know that the print industry is a funny place because we’re not always the technology leaders. A good part of the print industry still lives in the eighties, as a lot of it lives in the nineties, and some of it lives in the early 2000s. There’s really only a small part of the overall global print industry that is actually living in the 2020s.
[00:06:57] DC: And beyond. Just a couple of weeks, I had a couple of folks –
[00:07:01] PMG: Yes. Some of them are actually looking way forward and are actually embracing a lot of technologies. We live in an industry where we find we constantly have to explain technology, and we have to explain the ramifications of it. We owe it to the market that we try to educate to help them understand where the value proposition is in a new technology, and where the potential pitfalls are in a new technology. Not just AI. I mean, it’s all the different technologies that we work with.
We kind of box AI for a second, and we talk about it. You mentioned in the lead in the Harvard article, generative prompt technology, right? The idea that you ask a question and answers come back. That’s your ChatGPT, that you’re Bard. That’s a lot of – there’s Copy.ai. There are just tons and tons of them out there.
So, the way those technologies work is that you as the asker, you go to Mount Olympus, and you ask the Oracle, “Oh, mighty Oracle, what are the best ways for me to lift my company in the marketplace?” And I do signs and banners, right? Based on how you formulate your question to the Oracle, the Oracle goes and looks at a whole bunch of data pools and says, “Wow, I found a cool thing here and a cool thing there. Okay.” And then he goes, “Uh, questioner, I have an answer for you. Here are the things that you should do.” And it can be really, really helpful technology, because if you’ve got a block, if you just can’t – I think I’ve written the same blog 800 times, or I just need something fresh for my website. This technology can be really, really useful. It can help you get over it. Same thing for images, right?
[00:09:02] DC: Well, okay. But let’s keep images to the side for a second. Let’s just – because I have a whole thing about the images. But let’s just go back to what you said, because I see it a little differently than you do, which is that printing is part of the communications industry. We’re not just a self-contained unit when we require the input from other people in order to create print, right? We need stuff, like files, like content, like things to print. In that way, it’s important for printers to understand what’s going on the other side of this as well about things that might be coming to them, like images that may or may not be printable depending upon how they were generated. But going back to how ChatGPT which I am using as just a generic term, like a Band-Aid at this point for was it generative? What is it called? What’s the official –
[00:10:07] PMG: Prompt technology.
[00:10:09] DC: When I say that, that’s what I mean, although I know that there are different things. But I subscribe to all these newsletters now just to learn about the prompts, which you mentioned, and I think it’s really important that people understand how they work. But also understand what the potentials are.
So, for example, I just downloaded something that a printer could use, or a technology company, any vendor, go to ChatGPT and say, “Analyze the strengths and weaknesses of my top three competitors based on this information”, and then put the information in there. “Suggest a pricing strategy for my product considering these factors. Help me create a compelling brand story based on these company values. What are the current market trends in ‘insert industry here.’” Think about the ways that you can use that information to generate business, to generate content, and to create a symbiotic relationship with the people who have to give us content in order to get it out in the world.
[00:11:22] DC: Printspiration is streaming across the printerverse on the Project Peacock Network, and our mission to provide education and resources for print customers, students, and printers around the world has never been more accessible. Watch what you want, when you want, where you want. It’s free. Visit projectpeacock.tv to access original programming, and replays from our online events. Learn about the Peacock partners and companies featured in our shows. Join our mailing list to learn about new episode premieres, and series launches, and create a free account to make watch lists. Ready for your close-up? Get your Peacock show on air by visiting projectpeacock.tv. and request your partnership proposal today. Peacock long and prosper.
[00:12:13] PMG: So, you want to know the danger with all of those prompts.
[00:12:16] DC: Sure, Pat.
[00:12:19] PMG: I’m not a fan of the newsletters that send you all these prompts. I’ve mentioned to you before, I get all these newsletters, I get all these offers to sell me successful prompts, winning prompts, and y’all, I’m just not a fan of them. Let’s just take the one that you just talked about. What are the current market trends in an industry?
I’ve seen that one probably in a dozen different newsletters. If you put that one into ChatGPT, you know what happens, right? So, the first thing it says to you is, “Well, my knowledge cut-off is September 2021.” Anything that happened after that, it doesn’t know. We were in the middle of COVID in 2021, September 2021. My problem is that when I look at the trends that it comes back with, I don’t find anything particularly thrilling there. I mean, and we’re going to have to agree to disagree –
[00:13:20] DC: “Generate blog posts ideas that would interest my target audience. Insert target audience here. Help me create a positioning statement for my product.” I’m just saying that these are – I want to acknowledge what you’re saying. Heard. 2021 is allegedly the cutoff point except for – Bard is the one that is up to date. I believe you said.
[00:13:20] PMG: Bard is more modern because it borrows from the same data pool as an average Google search.
[00:13:54] DC: Okay. So, generating five blog post ideas that would interest my target audience and I put in designers as my target audience, I’m still going to be way ahead of the game than I would be if I had to come up with those myself. Then, I’m not saying replace humans. I’m saying that then a human should review it and say, “You know what, I’ve got another one that’s not listed here.” Or, “This is a great one, but it’s not up to date. I can tweak this a bit.” I don’t think throwing the baby out with the bathwater is –
[00:14:31] PMG: Not suggesting that.
[00:14:32] DC: Okay, so then what do you want to agree to disagree on? Let’s get on. What I think we’re disagreeing about is the validity of the information if it only ends in 2021, and I would agree with that. I still don’t think you should ask it something in post it and share it. You need to address it.
[00:14:52] PMG: I think that we can agree that the value proposition is to break blocks, right? The value proposition is – and these tools have been around for 20 years. Writers have used them for decades.
[00:15:06] DC: I’ve never heard of any of them. I’ve never used them. ChatGPT was like a miracle to me.
[00:15:12] PMG: I get it. I think for a lot of people, that’s probably the perception. Professional writers, scriptwriters, comedy writers, and long-form writers have used these tools for decades. So, they’ve been around for a long time. They haven’t typically been in the mass market, because they require a level of expertise to understand what’s coming back at you. With ChatGPT and the follow on GPTs have done is democratize the accessibility, right? Because anybody can go to OpenAI, and sign up for an account. The next thing you know, you’ve got all manner of tools available to you. Then, if you are a professional corporation, and you need something a little more robust, they have their paid plans, right? It’s the same thing with Discord and Midjourney. You can do the exact same thing. It’s the same thing, Bard, all these things, Copy.ai, same thing.
For me, the issue is that I’m not excited at the prospect of the result of a lot of people using these tools who don’t understand how they work. Who are just throwing, “Hey, I got a newsletter with 15 prompts in it that it says is going to make me my business shine. So, I’m just going to start copying those prompts into ChatGPT, and whatever it tells me to do, that’s what I’m going to do next.” Because as much as you and I wouldn’t ever do that, we like these tools, they’re great for blocks, they’re great for building frameworks, and they’re great for listing some ideas on an arc. They’re great for all those things. As much as you and I would use them that way and do use them that way, we would never copy something out of the response, and literally post it on LinkedIn, which I have seen happen too many times.
[00:17:06] DC: Yeah, no, I agree with you. I mean, ChatGPT doesn’t turn somebody into a writer or research expert, or an analyst the same way Canva doesn’t turn someone into a designer. You have some tools, and you can play with them, but at the end of the day – I heard something really interesting, which is that AI is not going to replace your job. Somebody who understands AI is going to replace you.
[00:17:33] PMG: I’ve seen that a lot. I’ve seen that repeated a lot and I think that’s fair. I think that’s a really great way to think about it.
[00:17:38] DC: I do too.
[00:17:39] PMG: Because the people who become facile with these tools, do you remember when Photoshop first hit the market, right? Oh, my god, the hue and cry, “Oh, my God. It’s going to take the art out of art.”
[00:17:52] DC: It’s the end of photography, as we know it.
[00:17:55] PMG: God, it’s going to be terrible because of this and that and the other thing.
[00:17:58] DC: Photographers.
[00:17:59] PMG: Then, as photographic libraries started to grow, you know, like Shutterstock, and Getty, and all those. They were like, “Oh, my God, photographers will be out of business. No one will ever hire a photographer again.” I don’t know about you. But neither one of those things happened.
[00:18:14] DC: No. But what did happen, retouching got really expensive when Photoshop came along.
[00:18:20] PMG: But the thing that also happened was that it democratized access to a bunch of tools, it made it possible for people who couldn’t afford the $300 an hour photographer or the $700 an hour professional researcher, it made it possible for them to be able to do the work that – and get what they wanted from their artistic expression. We’ve had these hues and cries before. But the other thing that it did is it made it possible for people to create really, really bad photography, and do really, really bad retouching, and really, really bad color correction.
[00:18:55] DC: Just the same where they could do really bad design on Canva and all that stuff.
[00:18:59] PMG: Really bad design in InDesign or in Cork. I mean, just because you have a tool that can help you fast path some things, does not mean that that tool imbues you with the powers of the educated designer or the educated photographer. I think that we’re navigating to is that we’re agreeing that the value proposition here is that it’s a tool in the toolbox, but it is not the end result. And that you as a professional, as a professional sign shop proprietor, a professional direct mail shop provider, a commercial printer, an agency, the freelancer who’s working with all these people. The burden is on you to understand what’s possible with these tools and also to understand what you shouldn’t trust about these tools. Because one of the things that has been really concerning me about AI. It’s one of the reasons I pinged you about wanting to do this print report is that I’m starting to see the litigation and I’m starting to see the really interesting ways that people are trying to game systems with AI.
So, the one I mentioned to you the other day, was the judge that threw, well, a legal brief, because God bless them, some lawyer thought it was okay to put a prompt into a GPT and ask for a brief for a case he was taking to court. It was a he. I’m not just generalizing that it was a he, who took it to court, and turned in the brief, where not a single one of the cases cited in the case law was real. Clearly, none of the links to any of the case law worked. The judge called the litigants into court and said, “Hi, no. Absolutely not and we’re throwing this one out. What made you think you could do this?” That’s one case that I’ve been reading about and following, but it’s happening a lot in the legal industry. It is happening a lot in pure blog writing.
One of my other favorite ones, is a lawsuit that involves suing someone who posted a blog, claiming that they wrote it. What they had actually done was put a prompt into a GPT, and had the GPT write it, don’t know which one, don’t care, and published it under their name. The problem was that the GPT, based on the libraries that it was looking at, literally picked up whole paragraphs, and did not reorganize them, did not analyze them. Literally just plopped somebody else’s paragraphs into it. There was no way to know that. If you were the one who put the prompt into the GPT, there was nothing there that would have told you, “Hey, this whole paragraph has appeared under somebody else’s name some way.”
[00:22:08] DC: Yes, it doesn’t give you sources or anything.
[00:22:11] PMG: It doesn’t give – well, it can get you sources if you ask it correctly. But at the end of the day, for things like long-form text, like a blog, or something like that, customer communication messages, the danger is unless you take the next step, and you, as the person who put the prompt in, now, take what you think you want to use, and you put it into a plagiarism checker. You can find yourself in court. Because it only takes two or three sentences to be guilty of plagiarism.
[00:22:46] DC: Yes. I agree. In our Print Report, when we went over events, and I had been a project voice, and this was a major focus at this conference, and they have a charter going now about protecting people’s intellectual property.
[00:23:02] PMG: Nice proactive thing to do.
[00:23:04] DC: Yes. I mean, they’re not saying stop technology. Be afraid of it. They’re saying, “Hold on a second, we need to be able to show people the sources.” Then from there, they have to make their own decisions, which is nothing different than a Facebook post. At this point now, you have to be a forensic detective, to figure out, “Is it real? Is it not real?” I don’t just mean from AI, I just mean, the information because we have this Earth one, Earth two thing going here.
Okay. Let’s put content to the side for a moment, which we’ve agreed to disagree, although I’m not really sure what. I think we still really agree on everything.
[00:23:45] PMG: We kind of agree on it. It’s just that I’m annoyed, I think is maybe the best way of thinking about it.
[00:23:50] DC: That’s fair because you are a professional writer. It has a different ping to you.
[00:23:57] PMG: Yes, it does.
[00:23:58] DC: For me, as somebody who is consistently trying to put ideas out into the world, I’ve found that something like saying what are the latest and greatest achievements in whatever it might be, really helps me, because all it does is gives me a list to start my research on. To your most excellent point, no, I would never take anything that it gave me and just stick it somewhere. Although, and I did mention this to you yesterday, ChatGPT wrote a killer freaking about us for Print Media Centr, and I didn’t tweak it, but I used it. Because in this case, it was something else speaking about Print Media Centr, not me, which took all of my personal hang-ups about, let’s say –
[00:24:51] PMG: How to phrase things.
[00:24:51] DC: – being braggy or sounding full of myself, whatever. I’m like, “Oh, well, this is what the Internet is saying about Print Media Centr so it’s fair.
Okay, at Project Voice, they had divided up tracks into healthcare, which the article also said is a bastion of AI technology investigation and usage, financial technology all into it, and the other – automotive is going crazy with it. The other area is what they call customer connection, but it’s everything from service to sales and everything else like that. I believe that that was the closest to the printing industry as far as if we look at printers, or we look at everybody, we all have customers, we all have customer communications. So, let’s focus on this for the next 5, 10 minutes, just about that aspect of it, because that is something that can actually help printers out there.
[00:25:59] PMG: If what you’re doing isn’t helping you grow, let McGrewGroup help you fix that. Better sales talk tracks, more compelling print samples, and winning workflow strategies can be yours. With decades of experience in transaction, direct mail, and commercial print, as well as years of marketing expertise, we can help with business and production strategies. CCM advice and develop your content. McGrewGroup group is ready to help you grow, expand, optimize, and thrive. Drop us a note on LinkedIn, or at our website, mcgrewgroup.com.
[00:26:34] PMG: Where do you want to start?
[00:26:35] DC: Customer communication. Chatbots. I mean, there’s a zillion – and I almost feel bad calling them chatbots because they’re not bots anymore. They have a little more intelligence behind them.
[00:26:49] PMG: Okay. So, I’ve been tracking AI for a long time. It’s been in the list of things that I kind of traipse through on a weekend morning with coffee, right? AI newsletters, articles in HBR, and other stuff. I took a little bit of time yesterday afternoon, to look at what I actually had in my AI bucket. What’s in my AI bucket? What if I captured and stuck into a folder because I didn’t want to lose it? So, imagine the giggling that happened when I tripped over, ‘4 Tips for Transforming Your Customer Communications With AI’ You want to guess the date?
[00:27:33] DC: 2014?
[00:27:36] PMG: Your close it was in the teens. It was in the teens. We’ll put it there. It is from a company called Broadridge.
[00:27:42] DC: Oh, we know them.
[00:27:44] PMG: We do know them.
[00:27:46] DC: Does Matt Swaine still work there?
[00:27:47] PMG: He does. This was not the beginning. This was not the first article that they ever published on AI. But it was their – it was the beginning of a campaign that they were working on, to talk – to basically socialize the idea of what AI could bring to a customer communications environment. They had already made the investment at Broadridge. They had gone and hired some really smart people. They had looked at what tools were available in academia, which is where most of the AI tools lived, and they started looking at what would you do with AI? Would you just use it to make processing more efficient? Would you just use to make things faster? Or would you actually use AI to start to analyze the context of an inbound message to try and identify, is this person mad at us? Are they thrilled with us? Do they just have a question that we can answer very easily? Or do they have a more complex question so that we can route it appropriately? Could we use AI to do that? Could we help our clients build smarter outbound communication because they know their history with every one of their clients?
But a customer service rep does not know the history of every single client, right? If you’ve got hundreds of thousands of clients, no single CSR knows everything that’s going on in every one of those accounts. You end up putting them into these big broad buckets, which means that you’re probably miscommunicating to most of them, right? So, could we use AI to help us be smarter and help our customers be smarter about how they communicate with your clients?
The article that I found was done by Rob Krugman, who was, at that time, the Chief Digital Officer. He’s a really, really smart guy and is always looking for an angle. I think he – I mean, he hit on it. He knew that that was something that Broadridge had the infrastructure to do, because they had the investment capital to look at it very broadly, and they had the vision to play a long game.
So, from the teens to 2023, it’s not really been that long. One of the things that they’ve been able to do is quietly, bury AI tools in their infrastructure. It’s not all over the news. It’s not all over the blogs. It’s just something that very quietly –
[00:30:45] DC: By the way, FedEx, you could very well be talking to a robot if you call FedEx.
[00:30:51] PMG: Absolutely. Because they’ve got the voice.
[00:30:54] DC: Exactly. I mentioned this on our last podcast, I heard an entire, a bot, troubleshoot somebody’s passport delivery issues at a concierge, at a hotel. I mean, that is a very specific problem that a robot handled, not what are your hours and today we are – and had an English accent. It was crazy.
[00:31:15] PMG: It can do that because the protocols are known. They didn’t have to invent the solution. The protocols are known. But the vast majority of customer communication does involve known protocols.
[00:31:34] DC: Absolutely. So, what I also saw examples of, and I just want to put this out there so that the printers understand that there could be a way that they can incorporate this into their lives, especially with workforce development issues. But again, you need the right person to really help you with this. But I saw an example of a person goes to a website, the artificial intelligence figures out where they’re coming from, understands what they’re looking at, on the website, compiles a, whatever it might be, an email, a newsletter, a printed piece, if that’s what they want to send, that either highlights what they were looking at, or goes further, or is some sales pitch about it, and then communicates whether it’s sending it to the printer, with a mailing list, and it goes out. If you have set up that – well, you are a printer, so whether it’s your own thing and you send it out. Or if it goes electronically, it was then sending that contact information to their Salesforce, which was the example that was used. And then Salesforce took over the relationship, let’s say the electronic relationship at that point.
There was rarely a human involved unless the problem went beyond what the AI was programmed – I don’t really want to say programmed but was given the knowledge base to understand. I think that’s probably a better way of saying it, right?
[00:33:13] PMG: Yes, that’s better. Yes.
[00:33:16] DC: If it went beyond that, it knew that it needed to say, “Let me get you to somebody to handle this.” Just like if you call the airport, you call the airlines or something, and the human can help you, they send you to the technical support, whatever it be. So, it does know its own limitations in some sense. But the point is that, if that customer didn’t require a human, that entire relationship is now being managed by artificial intelligence, and if they’re purchasing things, that’s why they’re doing it.
I don’t know how close we are to that being accessible to everybody in an affordable manner. I was told that at minimum, a system that could do what I just said, I’m not saying I have no idea about Salesforce or the other APIs that it could call into. But a system like I just described, minimal starts at $50,000 a year, which by the way, how much do you pay people for that? I’m all for people, but not if a technology like this could resolve a lot of the problems that stop work from going in and out of a print shop.
[00:34:41] PMG: I think you’ve got the right path in the thought process there because, look, we know that in our industry, every company is people challenged, right? It is really hard to hire new people. It takes time to train them. There are certain repetitive tasks that people don’t want to do, right? I mean, they just don’t want to sit there and answer the exact same question over and over and over again. They don’t want to this file from place A to place B over and over and over again. There are a huge number of really valid reasons that embedding AI into your environment is a really valuable thing to do. You’re right about the pricing. I mean, a full-time person, hell, even a part-time person, let’s be honest, burn up 50 grand before the year is out.
But the other thing that you have to do is you have to understand that in order to take advantage of these great new technologies, you don’t just call up your Salesforce rep and go, “Hey, hey, he, yeah, you got some AI for me? Can you have that here tomorrow? And what, I’ll be up by the afternoon, right? I mean, because it’s AI, it’ll install itself and it’ll be fine, right?” Because I have heard people believe that that’s what happens, that when you bring AI into your business environment, because it’s AI itself installs, and it self-maintains, and it just knows what my business needs. Nothing could be further from the truth. Just like any software, and this is software –
[00:36:14] DC: And it’s an integration. It is not a –
[00:36:18] PMG: It is an integration and it is workflow. If you do not know what is going on in your shop, you do not know the kinds of questions that are coming to your team members, AI isn’t going to help you. Spend a year building your SOPs. Get your standard operating procedures documented. Get your processes documented. Consider adding as much automation as you can to your production workflow, so that you’re freeing up some people.
Now, that you’ve got some people freed up, take those people and have them start documenting, what are the things that usually go wrong? What are the questions people are asking? How do we resolve those questions? Because a print shop is not FedEx, and a print shop is not a manufacturing plant. It is not a bioscience or a healthcare, or an insurance agency.
[00:37:18] DC: Or a winery. I told you I heard a bot sell cases of wine.
[00:37:21] PMG: Or a winery. What it is, is a print manufacturing environment that has a lot of people-based touchpoints. Because of that, you’ve got to document it all.
[00:37:37] DC: Yes, but this is exactly where I want it to get to, so thank you for getting there. Because this is the whole point. Now, is the time to do. Everything Pat just said, this is not going away. You cannot, we cannot, the printing industry cannot ostrich themselves about this. I realize some people out there don’t have responsive websites. We’re about to get to the point now where you need to skip over all of that, and really just move towards this, or with all due respect, start considering your mergers, acquisitions, and all that other stuff. You know what, get out of the way, because otherwise, we’re going to be bulldozed by this. I always think that that’s one of the worst aspects of the PR of the printing industry.
I always say it’s not the medium of print that is antiquated. It’s the damn processes in the print shops that make anyone under 40 think, “Oh, my God, what is – are there gerbils in there on hamster wheels running things?” You know what I’m saying? They don’t even – what do you mean you don’t – I don’t get a notification when my job is ready.” They don’t even understand that.
So, the reason why I wanted to have this conversation with you is that this is coming and not putting the content to the side which is already there, and we should probably have another print report about the images because I think there’s a whole business opportunity for printers about the images.
[00:39:21] PMG: And business danger and risk mitigation requirements, yes.
[00:39:23] DC: Well, there you go. I mean, you always have to be – you’re actually Debbie Downer. We have to –
[00:39:30] PMG: I know, but that’s because I’m a – it’s not because I’m not enthusiastic about the technology dip. It is because I’m watching people jumping into it who don’t understand the risks that they’re taking if they haven’t laid their infrastructure.
[00:39:45] DC: News from the Printerverse delivers topical sales and marketing insight, along with plenty of printspiration one time a month to inboxes everywhere. Our contributors cover the industry and the future of print media and marketing with strategy first strengthening your customer relationships, better targeting of your prospects, and practical advice for helping your business grow. Printspiration is just a click away. Subscribe to News from The Printerverse at printmediacentr.com. Print long and prosper.
[00:40:22] DC: Okay, that I 1,000% agree with. Yes, let me just plug this in and see what happens. Probably a bad idea.
[00:40:29] PMG: It’s not the best strategy.
[00:40:31] DC: No, by the way, and I’ll give you an example. I did that. I plugged my computer in, in Switzerland, and all of a sudden it was like, “Eh, eh, eh.” I had no idea what was going on. And then I realized, “Oh, if you actually plug the wrong plug into a socket, it gets very upset.” So, you shouldn’t just say, “Oh, there’s a socket, I’ve got a plug, everything should work out just fine.” But do we agree without disagreeing that whatever anyone listening to this podcast does about this, understanding the uses for it is super important to then start understanding if this is something you can incorporate into your business, how it would save you time, help you create better processes, better relationships, make more money, whatever it is. Then, to your most excellent point, find the right company.
Because to another point that you said, and even just go into the project voice thing, there’s a roomful of vendors, room full of companies there, that focus on one thing. AI for shopping carts. You have a shopping cart, you need AI, you need it to hook up to all these other things. You need a company to do that, or you can try with a “consultant”, but I wouldn’t trust anybody with that. But a company that has been immersed in this for a long time and understands the legalities, as well as everything else.
[00:42:05] PMG: I think that I would add one more thing to that. In the print industry, we have a farm full of vendors and some very, very smart vendors who are – they’ve got their foot forward on technology. They’ve been – people like Broadridge, who have been looking at AI for a decade. There are a lot of companies that have been looking at and trying to gently figure out what the value propositions are for our industry around AI. It is sometimes harder to talk to someone who is not from our industry, about how to integrate their solution into our industry, because it’s so strange, and it’s so diverse, right? Because printers are different from wide format. Printers from commercial printers.
But the vendors that you are already working with, for me, they’re your first stop. Have a conversation and you can disagree with that. But the reality is, no matter who you were talking to with, no matter who your workflow provider is, who your hardware provider is, your point tool providers are, no matter who those people are, they all have an AI program in their business today. They all have people who are looking at it. They may not have your AI-based answer, they may not have a solution that’s already AI empowered, but they may have partners who have already worked with them and have worked with the smart people in the AI industry and they can build that integration – remember what you said about integration? This is an integration play. You don’t just plant AI and it sits there all by itself, right? It’s got to be integrated into your business. For that to happen, it only makes sense to talk to the vendors who were helping support the way your business runs today, and using their power to make sure that when you add AI empowerment to your business systems, that you’re doing it in a way that isn’t going to break your business systems and shut you down.
[00:44:10] DC: Yes. But it’s also super important. One of the things that I particularly like to ask the vendors is what is their vision of the future, because people are making investments in that company, as well as making investments in their own company, if they decide to work with them. As we know, there are giant printers out there, but there are smaller printers who might have to mortgage their house in order to buy a piece of equipment. Understanding, do I need to replace this in three years? Can I add on to it? Understanding what the vision is, is super important. That’s what I would say about what you just said. If you’re in the market for technology, make sure one of the questions that you’re asking is, “What is your vision of the future of AI for my business, for your business, for the printing industry, for how this software can, or workflow system, or whatever it might be, my customer communication system, how is this all going to work together?”
If they look at you like a deer in headlights, probably not the company that you want to make an investment with. If they’re on a road to it and can give you a roadmap, and at least you know that they’re discussing it. To your point, there might be some companies that are further along, who have made a conscious decision to integrate somehow with the printing industry. But right now, the only thing I’ve seen is the integration with the customer relationship, customer support, part of it, sales, and content too. Not the actual running of the print shop, but I’m sure that’s coming. Yes?
[00:45:57] PMG: Yes. I mean, inevitably, what happens is that as more and more workflow automation enters the print shop, a lot of those automation tools start to use AI and machine learning to learn how work flows through your shop and create more optimized routes.
[00:46:16] DC: And it’s scheduled? Can it decide – we can put these five jobs on the press, do the imposition? I mean, does it –
[00:46:22] PMG: It’s already happening. So, let’s think, the most popular imposition tools in the printing market today all have AI engines underneath them to help decide – here’s the fun thing about a print shop, right? “Joe’s been doing our imposition for 50 years. He knows more about how our equipment needs to be fed. He knows our customers’ jobs. He knows the perfect way to impose this job, so it just goes smoothly.” A funny thing happens, if you hand that job to an AI engine, it’ll come back with 100 ways that you could impose it, and 20 of them would save you money because they eliminate more waste. Joe didn’t know any of them because Joe has been doing it the same way for well over 50 years.
So AI is already in our industry, and it’s already doing things in our industry, but it hasn’t been lifted as AI. It’s been just, “Oh, we’re faster than anybody else.”
[00:47:19] DC: Okay, I get you. It is like the secret sauce.
[00:47:19] PMG: It is. It’s really an integral part of a lot of the workflow automation. I had job scheduling of a lot of the job scheduling engines have an AI infrastructure under them, because if you can tell the job scheduler, what your equipment is, what your finishing is, and what the run lengths are, for a block of time, it can then use its learning to understand what the best order of execution is, for the jobs which jobs to batch together, which ones to separate, so that it minimizes substrate changes, minimizes printer setup, minimal changes, finishing machine changes. It can also be taught to understand, like for cutters, how long between blade changes, so that I don’t wind up with dull blades in the middle of a long-run job, right?
All of those things become something that Joe used to know. But now, the machine, the scheduling system can eliminate the problem by saying, “I’m not going to schedule to the point where that blade is going to be dull. Because I know the substrates and I know what I’m trying to do.”
[00:48:32] DC: Yes. I mean, it’s completely fascinating. Okay, obviously, we can go on and on and on about this. I’d love to do a part two for the next Print Report. But for now, let’s leave this here. A lot for people to think about.
Hey, out there, listening to The Print Report, please leave us some stars and a review. We’re doing very well on Apple podcasts. But a couple more moves up the charts would be awesome and that only happens through stars and reviews. That is their SEO. That is how it works over there. If you enjoyed this episode or any episode of The Print Report or the other episodes on Podcasts from the Printerverse, please do me this favor and just scroll down to the bottom, leave some stars, leave some reviews. Really appreciate it.
Until next time, everybody. Print report long and prosper.
[00:49:28] DC: Thanks for listening to Podcasts from the Printerverse. Please subscribe, click some stars, and leave us a review. Connect with us through printmediacentr.com. We’d love to hear your feedback on our shows and topics that are of interest for future broadcasts. Until next time, thanks for joining us, print long and prosper.