Episode 69 Transcript

Adapting to Change: Agencies, AI, and the Future of Value Creation w/ Matthew Kasindorf

    Brent Trimble: Welcome to the Professional Services Pursuit, a podcast featuring expert advice and insights on the professional services industry. Again, I’m Brent, and today I’m joined by a special guest from the 4A’s, Matt Kasindorf. Matt is the SVP of Business Intelligence and Insights at the 4A’s, an organization that I’m sure everyone in our agency industry, clients, and listeners of the show are aware of and have looked to for guidance and support at some point in their career and journey within their respective agencies. Today, Matt and I will be digging into the topic of agency transformation. We’ll discuss what’s still holding many back, the cost of inaction in evolution, and much more. So, Matt, welcome.

    Matthew Kasindorf: Thank you very much. I’m very happy to be here today. I appreciate the opportunity.

    Brent Trimble: Yes, we’re excited. Before we dive into these topics, I think most of our listeners know of the 4A’s. Maybe you could just give us a little bit of an elevator synopsis of the 4A’s. But really, tell us a little bit about your journey, how you made your way to the 4A’s in your current role.

    Matthew Kasindorf: Thank you, Brent. I’ve been at the 4A’s for a little over seven years now, having spent the majority of my time before that in a dual account management and business development role at a number of different independent agencies, and then two of the global media agencies. The 4A’s, as I think everybody knows, is the American Association of Advertising Agencies. It’s been around since 1917, and we offer a number of different opportunities for our members to engage with us, to learn from us. We work with the ANA, Association of National Advertisers, to better the business in any way we can. We work with our members on areas from finance, operations, business development, account leadership, strategy, etc.

    Brent Trimble: Oh, that’s great. Thanks so much. It’s great to hear about that experience and trajectory, coming from a deep practitioner and subject matter expert, definitely having that dual hunter-farmer role in the agency world, and then seeing it all in that context. There’s much written about the agency business, and it seems everything from the trade press, of course, the advertisers themselves, and then ultimately, even in the business press. You’d be hard-pressed to find other business categories that have been under so much change and external forces, sometimes within their control, sometimes out of their control, with the fragmentation of media, the different types of service models, everything from the march of management consultancies into the agency space, and everything in between. I think most of our listeners understand those changes, and we’ve had some great guests on the show to talk about that. But at the 4A’s, when you look at the landscape of the agency business, how do you ingest and digest and think through this constant tension and change, and ultimately help service the membership and give them resources that they can use to affect their own shops through this landscape?

    Matthew Kasindorf: Well, Brent, it’s our job to stay on top of all of this. And I think we have seen such an increase in disruption to the business, certainly even this year so far. We have again, in partnership with the ANA, I think that’s actually a big deal for us and for our members because we work with them on some of the issues that we’re all facing. I know it is true that the agency business is disrupted often. If you think about television, right, at one point, the business was going to go away because television started. First, it was newspapers, then radio, then television, then, oh, this thing called cable, that will be the end of agencies. And then this thing called the internet. Well, that’s going to be the end of agencies, and then digital. So there’s been a lot of times that there have been predictions of the end of the need for agencies and agencies are flexible and agencies adapt.

    This is a similar situation that we see our members going through, and adaptation and flexibility is absolutely necessary.It is as much as the agency world is seeing disruption and things like that, the client world is too.And I think we can easily sit here and say, oh agencies are being disrupted, clients are not.But that’s not true.Clients are seeing the same level of disruption, the same fragmentation of efforts, the same attempts at demonstrating to their management, the CEOs of the world, the CFOs of the world, that advertising and marketing is a value contributor to those brands and those companies.

    I think that’s something that together, we have to work on.It’s easy for a client to say we have to cut our budget.You got to do more.The reason the budget’s been cut is because nobody’s really understanding the contribution that marketing and advertising is making towards that business.

    Brent Trimble: That’s great insight, and change, I guess, is internal. But we’re going to talk about a few specific changes. And in our role as a technology provider to quite a few agencies of all stripes, we get to hear from many leaders in the space. We attend conferences, including the 4A’s, and a lot of the challenges right now, having come post-pandemic, are economic, but there’s also technology. I’d like, before leading the witness, just to hear from you as really the industry association. If you had to narrow down to two or three, what are the primary pressures and changes that the business is confronting at the moment?

    Matthew Kasindorf: There’s a whole lot, and narrowing it down is kind of simple to a few. Agency economics is probably at the heart of most of it. What I mean by that is when times are good, it’s very easy to just let the business run. When times start to tighten up, it’s very important that agencies understand how they’re making money, where they’re making money, and how to increase their margins. This is something that is impacting every agency.

    When you think about the rise of AI, what are we doing with it ? How are we going to use it? Are we going to deploy it externally? Are we going to deploy it internally? You think about the kinds of tasks that AI can and should be appropriate for.Then you think about how the use of AI can maybe cut down on the time needed to do something or cut down on the number of person- hours that are needed to do something.Then you look at how an agency is compensated, and you look at a number of different remuneration practices and models that are centered around billable hours.AI is going to be a major disruption there.Agencies need to think about how they can work with clients in a different way.Can they work on a fixed fee basis ? Can they work on a value basis ? Can they work on a pay -for-performance basis or a combination, a hybrid of those different types ?

    But I think one of the big shake - ups, if you will, is going to be a very fast realization that relying on billable hours is not going to work moving forward.It’s something that is going to impact every area of an agency.So that, to me, is certainly the big one.

    You also have, and I kind of referenced it before, but I think you also have a fundamental issue of how do you prove the value of what you’re doing.And I’m not necessarily saying to the CMO.I’m saying to the CEO at the client side, CFO on the client side, and it’s something that marketers and agencies need to work on together.Unless the value of what we all do for a livelihood is understood, the margins, the budgets, everything are going to keep going down.Once it’s proven that advertising actually does build businesses, and we know it does, I think we’ve gotten a bit away from being able to prove it.But once we can get back to a situation where we can tie an expenditure and the work that’s done to an increase in business for the client, then clients will say, and by clients, I mean the whole organization, will get behind marketing, get behind advertising and say, “Yeah, this is a necessary part of what we do within our organization.”

    Brent Trimble: Now, those are great topics. And we’re going to dig into that a bit, particularly the fee margin, pricing value, non-hours, and AI. They’re kind of interesting, right? Because even on our podcast, we’ve had some real thought leaders in the space, who I know have relationships with the 4A’s. We’ve had Tim Williams, who’s been really preaching and teaching this for quite some time; Michael Farmer, with his two books, Madison Avenue Manslaughter and then Madison Avenue Makeover. We again, have several hundred agency and agency-like object clients who are, I’d say, experimenting with different pricing and value models.

    The interesting thing too, is, as go agency, so goes probably management consulting, too.Do you think we have reached a bit of a tipping point in the sense that AI and its ability to really accelerate tasks that maybe haven’t been strategic, although I know you’ve got some examples and I’ve heard some too of agencies really using them strategically, but you go right to creative adaptation.And agencies looking at this technology, they’re saying, this is a way to really accelerate this portion of our business.And if they have a lot of revenue tied up in that business that has a fundamental impact, is that coupled now with the rate fee pressure, really going to force change? What are you seeing in the market? Are you seeing some more momentum and some gravity and some impetus put behind this notion of, hey, let’s take a look at the model and let’s get back to maybe pricing value?

    Matthew Kasindorf: We just completed a study with the ANA on agency compensation, alternative models, and things like that. And we have seen, and I think this is the first time we’re seeing it, where fixed fees have overtaken cost-plus when it comes to our model. So we are seeing agencies focus more on a deliverable. This is the cost of the deliverable. There’s value built into this deliverable and less on “oh, it takes x many hours to do.” It’s a necessary change. It also gets to your point about creative adaptation. So yes, there are agencies who are certainly using AI in a more strategic manner. There are a few that have developed AI personas against which to test creative options, a very smart use, a very interesting use.

    One of the things that, and I heard this at a conference and I really respect it, is one of the things that AI can do for an agency and for a client is to help explore more than was explorable before.And I think that gets to the idea of AI opens the door to different possibilities because it can produce things that were not able to be produced before.Using AI to help someone somewhat avoid, if you will, the blank page, I don’t care whether that’s a written page or a blank screen, but if AI can help someone get past that initial blank page and it can throw out thought starters, what that means for a client is that there may be many more creative and media opportunities to consider.And when you think about the ability of AI to bring in so much data and synthesize it, you also can be thinking about, will AI help us as an industry prove the value of advertising and marketing communications? And it may very well help us do that.

    Brent Trimble: The AI synthetic agents persona example you raise is a great one. I just came from a conference. In fact, I led a roundtable of, let’s say, medium-sized to ascendant large independent shops. Everything from Copilot to helping with rudimentary tasks, note-taking for accurate conference reports, then to very sophisticated models similar to what you described, and then, of course, creative production and content writing. Being able to harness and maybe feed that ever-present content engine in a lot of different global client engagements. Do you, with your study and your analysis on compensation, have you sensed any, and don’t think about how you want to answer this with the interaction with the ANA, but when I saw, and this is several years ago, some of the first real-life use cases of some of the adaptation tools through Adobe and through the Creative Suite, my immediate thought was, this is going to solve a kind of challenge that’s existed. Agencies have been able to do with lots of scripting, lots of headcount in different markets. But then I went to client procurement and thought, gosh, is now in every major economic global SOW review the question going to be how much of this agency that you’re proposing to complete can be done with AI because you don’t want to price some people out of it. Has there been any kind of that tenor?

    Matthew Kasindorf: I think we are hearing some of that. And I think it’s kudos to clients and to client procurement that it is not just, “Oh, I expect you to do this with AI, get rid of all these people,” because that would be a huge mistake. It’s up to agencies and clients together to have these kinds of conversations, to see what AI can be and should be doing for the agency and for the client.

    There is certainly, and I’m not saying anything anybody doesn’t know, but there’s certainly enough concern, rightfully so, over things like copyright, over things like privacy, over things like AI hallucinations, where you should not just expect AI, “Oh, go do it.We’re going to push a button and it’s all going to be taken care of.”

    I was at a conference recently and a conversation at one of the, I guess it was breakfast tables, was about human first and human last, meaning to do it well, it takes an incredible amount of training and thought to be able to prompt AIs in a way that gets you what you need.And once that happens, so that’s human first, then you’ve got AI, and then the idea of human last, which is you have to check it.You have to check it.

    I was looking at something the other day.I asked an AI for it and it gave me some information, and it said there are several case studies that basically prove this out.So I said okay, please send me these case studies or please show me the case studies.And the response was more or less, “I misspoke.There are no case studies.It’s anecdotal.”

    So you look at that and you say, okay, we are nowhere near a world, and maybe we won’t ever be, maybe we will be, where you can just say to an AI, “Do these adaptations, create a brief, create a strategy,” and it just does it and you just turn it over.And you mentioned using various AI tools for better conference reports.If you really look at the language that comes out of some of the AIs, it’s reporting what it hears and what it hears, it may have misheard.And you need to actually go through the transcripts and go through the reports to make sure that whatever you’re putting out there, be it to people within our organization, within the agency, back to a client, is truly what you want it to be.

    Brent Trimble: I think there’s an undercurrent of shops embracing tools for utility work and using various agents at their disposal for tasks like scheduling. We had an agency example where they were digesting lots of data, not super sophisticated, but you think of a new business role. For many years, the pre-pitch briefing on a category or a client involved getting smart quickly about their business. These enterprising planning teams would produce massive books full of valuable information. You’d read them on airplanes. They contained everything from earnings statements to lists of acquisitions to previous brand work. One shop was saying they’ve developed ways, though not perfect, to take the budget that would have been for these analysts and apply it to different types of hires and invest in different practices. That’s an example of a positive impact of headcount and people. Rather than enlisting an army of folks to go through, digest, and collate all this information, let’s take that payroll and maybe the same people, frankly, and put them at a different type of practice or something tied to effectiveness.

    You mentioned research around fixed price.That is, I think, a nice interim step to everything we’ve talked about, a move away from pricing just the billable hour.With the 4A’s, are you doing any paired research with the ANA or anything upcoming from the 4A’s that listeners would want to be aware of? Trends around agency usage of AI, maybe best practice? You mentioned copyright.That’s a huge issue.Agencies are the stewards of a tremendous amount of client data, IP, copy, trade secrets.We want to avoid well - meaning interns suddenly uploading reams of this to help train an AI agent, so there has to be care there.But any interesting research that exists or maybe that is in your pipeline that would be of note ?

    Matthew Kasindorf: Yes, we are close to finishing a research study on the use of AI within agencies. We have the alternative compensation report available for purchase on our site. You raised an interesting question about how you think about using AI. We look at it, perhaps oversimplified, as whether an agency can use AI to improve its internal processes, workflow efficiency, etc., and then whether you are using it externally for something that will be put out into the marketplace.

    There are concerns on both sides.One is, are you putting client information or your own information into a system that’s training the world ? Or have you built your own walled garden, in which case, what we do stays within us.Let’s say, from a new business perspective, could we upload all of our case studies in there and then have that as a system that, based on a marketing problem that comes in from a new business prospect, the AI could help us identify which case studies are the most relevant.

    Someone made a comment to me a month or so ago.AI is a machine; it doesn’t forget.When you think about someone who is working on, say, three RFPs at the same time and has to build the case studies, that person may forget that there was one we did three years ago that would have been perfect for this particular RFP.They overlook it.AI would not overlook it.It doesn’t forget things.

    On the external use point, that’s really where you get into a lot of the copyright questions.There’s a lot of discussion about whether something produced by AI can be copyrighted.I don’t think it can yet.I don’t follow the laws that closely, but it’s a great question.How do you know and can you verify that what is in an AI - produced piece of content isn’t infringing on someone else’s copyrighted material ? Those are important questions, things to figure out before everyone just jumps in and has stuff created.

    The other side of it, Brent, is the consumers, the audiences, the people who are exposed to AI.There was some concern that consumers have over viewing what they perceive as AI - created content.As an industry, we also have to be very careful if consumers are questioning whether they trust AI - created work.We’ve got to be careful because the advertising business has never been one that people just trust wholeheartedly.Anything that could damage the level of trust we do have, we’ve got to be careful with.So it’s complex.We all have to be somewhat measured in how fast we rush to adopt things like AI, and in truth, any technology.We should be a part of it.Agencies should be a part of it.Clients should be a part of it.But agencies and clients should do it together with the understanding that it may not work everywhere.It will work someplace.It’ll be fantastic in some places.That’s something that needs to be discussed and understood within the client - agency relationship

    Brent Trimble: That’s great advice, and it certainly sounds like you and your role in the 4A’s have a really good handle on that. When we conclude the episode, I’ll give listeners the website domain, so they can go and take a look at the abstract of that research and then purchase if they like.

    One final topic as we go through this discussion is technology - related.We, at Kantata, participate in a webinar with TSIA, an industry association for technology and services.There was this notion of technical debt.They did some research around this notion that systems, processes, technologies, some antiquated, some deprecated, some just not well integrated, potentially hold them back and effectively are a tax on an organization.It can deliver bad employee experiences, reduce client satisfaction, spread out the all - important compensation process by erroneous invoicing, everything we’ve all experienced, I think, in either agencies and consultancies because we’re always in that cobbler’s children’s shoes type of proposition.There’s certainly AI potential within technologies to run agency operations better, but what are you seeing, and how does it relate to agencies in this pressured environment ? We’re being bombarded by this complete explosion and acceleration of AI, and then conversely, knowing that some of the systems, the processes, the rigor, maybe trends leading to a new pricing model, the technology might be holding us back.Where do you see the knock effect that has on the agency business ? What’s your counsel to them when you think about things like operational rigor, which hasn’t always been a hallmark of consulting as well as the agency space ?

    Matthew Kasindorf: Depending on who you talk to, agencies are either well-run machines, controlled chaos, or just total chaos. There’s a school of thought that chaos can’t be controlled, shouldn’t be controlled, and it should just be left to percolate and run that way. Interestingly, when times are flush, if you will, maybe it works that way and the need for operational rigor, the need for a complete, holistic view of your business is not as great as it may be today. What I mean by that is you need to be looking at everything from your recruitment, your retention of talent, to your recruitment and retention of your clients, and of course, new revenue, new business. Putting in place whatever systems and operationalizing processes that are going to make it better internally to be able to deliver what you need to deliver for your clients is critical.

    You don’t want your teams, your staff getting frustrated because the internal systems aren’t there to deliver what needs to be done quickly.Clients used to be good, fast, cheap.Pick two of the three.Now it’s good, fast, cheap, and a lot, all four, absolutely necessary.So as an agency, you have to think about how do we deliver that ? Are we set up to deliver that ? What do we need to do to improve operations to be able to deliver that, and at the same time to make a margin that’s acceptable and keeps us as a healthy organization ? Because the reality is, and honestly, Brent, good clients do know this, that the agency needs to be a healthy, ongoing concern for it to deliver great product, great service to that client.You need to be able to pay your people appropriately.You need to be able to invest in the resources that you need to make it all work.

    Brent Trimble: That's a great summary. I hadn't heard it put that way, but it sounds like what you're saying is good operations in experience really build the conditions for success.

    Matthew Kasindorf: I could not agree more.

    Brent Trimble: Well, this has been fascinating. I want to ensure that our listeners know how to reach the 4A’s. So, for those who are listening and are interested in what the 4A’s offer, as well as what membership might entail, and to access the research that Matt referenced, the website is the 4A’s, aaaa.org. Matt, do you have any insight on where they should go to peruse some of the research that’s available for purchase or download once they arrive?

    Matthew Kasindorf: Sure. So, on the homepage, the alternative compensation paper is right up there. Just click on it, and you can go to a link to take a look and purchase it. As we finish up this piece of research that we’re doing on AI, that will also end up on our homepage, right at the top. But certainly, folks can reach out if there are any specific things they want to talk about, different types of research we may have done, or talk about membership. Feel free.

    Brent Trimble: That’s excellent. Thanks so much for joining us. It’s been a great discussion. I think this theme of transformation, changing the compensation model, the explosion of AI, and how they’re intertwined has been really compelling. I believe it will provide lots of value and some tangible ideas to our listeners. So, thanks so much for joining. And for our listeners, again, our guest has been Matthew Kasindorf. He’s the SVP of Business Intelligence and Insights at the 4A’s. I’m Brent Trimble, one of the co-hosts. And as always, reach out with questions, ideas, comments at podcast@kantata.com, again, podcast@kantata.com.

    Matt, thanks again and appreciate your time today.

    Matthew Kasindorf: Thank you for having me, Brent.I enjoyed it.

    Brent Trimble: If you enjoyed this podcast, let us know by giving the show a five - star review on your favorite podcast platform and leaving a comment.If you haven't already subscribed to the show, you could do so anywhere you get podcasts on any podcast app. To learn more about the power of Kantata’s purpose-built technology, go to kantata.com. Thanks again for listening.