Episode 447

podcast photo thumbnail



Clients increasingly want to know exactly what they’re getting in return for their dollars spent with agencies. This data can be tricky to compile, let alone interpret and visualize in a way the client actually understands. So this week, we’re diving deeper into the ongoing AI conversation to dial in on one thing — how AI tools can prove an agency’s ROI to clients.

In the episode, our guest, Jonathon Reilly, gives a great overview of how AI can revolutionize this task. The big idea is that AI tools can crunch huge amounts of data and surface valuable insights way faster than a human data analyst could. It allows you to be more responsive to clients, show the impact you’re delivering, and ultimately retain clients longer by demonstrating your value more clearly.

But it’s not just about reporting — Jonathon thinks AI tools will help agencies be more efficient across the board, from creative ideation to repetitive tasks. The goal isn’t to replace human effort, but to use AI as an assistant that makes us faster and better at our jobs (more ROI for clients).

Tune in to learn more about how AI tools will level the playing field for small- to mid-sized agencies to bring more value to our clients in a tangible, understandable way.

A big thank you to our podcast’s presenting sponsor, White Label IQ. They’re an amazing resource for agencies who want to outsource their design, dev, or PPC work at wholesale prices. Check out their special offer (10 free hours!) for podcast listeners here.

AI tools

What You Will Learn in This Episode:

  • How AI intersects with data to help agencies prove their worth to clients
  • Helping clients understand the data and numbers
  • Will AI tools remove the need to hire data analysts in the future?
  • Being data savvy to prove ROI
  • Leveraging AI tools to maximize agency efficiency
  • How AI will change the billable hour
  • Starting the conversation with your employees about encouraging the use of AI tools
  • Embrace AI, but with boundaries and structure
  • Getting more educated in the AI landscape
  • How AI tools are leveling the playing field for small- to mid-sized agencies
  • How Akkio works to maximize agency efficiency

“We are strong believers that AI will enable a shared analytics communication surface between agencies and the clients.” - Jonathon Reilly Click To Tweet
“The more you can demonstrate your impact to clients, the better. You'll retain them and be able to grow your business.” - Jonathon Reilly Click To Tweet
“Make sure that the output you make at the end of the day still meets your expectations as an agency.” - Jonathon Reilly Click To Tweet
“Smaller agencies will be able to bring the same level of data sophistication just in an easier-to-use toolset.” - Jonathon Reilly Click To Tweet
“The set of people who can understand what's happening, why it's happening, and what you can do about it from data has expanded dramatically.” - Jonathon Reilly Click To Tweet

Ways to contact Jonathon:


Hey, everybody. Drew here. You know, we are always looking for more ways to be helpful and meet you wherever you’re at to help you grow your agency. It’s one of the reasons why we’ve produced this podcast for so long, and I’m super grateful that you listen as often as you do. However, there are some topics that are better suited for quick hyper-focused answers in under 10 minutes. That’s where our YouTube channel really comes in. For quick doses of inspiration, best practices, tips and tricks, head over to youtube.com/the at sign Agency Management institute. Again, that’s youtube.com/the at sign or symbol.

And then Agency Management Institute, all one word. Subscribe and search the existing video database for all sorts of actionable topics that you can implement in your shop today. Alright, let’s get to the show.

Running an agency can be a lonely proposition, But. it doesn’t have to be. We can learn how to be better faster if we learn together. Welcome to Agency Management Institute’s Build, a Better Agency Podcast, presented by White Label IQ. Tune in every week for insights on how small to mid-size agencies are surviving and thriving in today’s market with 25 plus years of experience. As both an agency owner and agency consultant. Please welcome your host, Drew McLellan.

Hey everybody. Drew McLellan here with another episode of Build a Better Agency. Glad to be back with you. Thanks for hanging out with us again today. I think you’re gonna love this episode. It is, as you might imagine, a topic that many people are talking about. We’re gonna talk about ai, but we’re gonna talk about it in terms of sort of how we can be leveraging it to better the conversations we have with clients. So I think you’re gonna enjoy that and I think you’re gonna find our guest super informative. So I’m excited to introduce you to him in just a minute. Before we do that, I wanna remind you that the Build a Better Agency Summit is right around the corner. May 21st and 22nd, May 20th if you’re a member, is Member day.

But one of the speakers that I’m super excited about is a gentleman named Jeff Saltzstein. And Jeff focuses, he’s a former professional tennis player, former ranked top hundred world ranked tennis player, and he now is an elite success coach and he works on performance and health optimization for leaders and for business owners. And he’s developed this thing called the Zone of Excellence, which is a sort of a three-step methodology for managing stress, cultivating resilience, and really tapping into our full potential.

So given the stress of the last couple years as agency owners and leaders, I think Jeff’s lessons and framework are really gonna be a valuable way to spend 45 minutes at the summit. So I highly recommend that you come to the summit, join us, but that you check out Jeff’s presentation and take away from that some ways that you can feel better, perform better, have more resilience, reduce that stress a little bit, and feel good about what you are doing, and then be able to take that back and share it with your team. So very excited about that. So join us May 21st and 22nd, build a better agency summit this year in Denver.

We’d love to see you. Tickets are on sale now. You can just head over to the Agency Management Institute website, so agency management institute.com, find the Build a Better Agency, summit navigation link, and click on registration and you’re all set. Grab your tickets before we sell out. We will be glad to have you. So, alright with that, lemme tell you a little bit about John Riley. So John Re is the CEO and founder of a company called ao. And AO is an easy to use AI platform that really helps users kind of build and deploy AI solutions in a matter of minutes. So they have a sort of do it yourself model.

You can have them help you build it out. They’ve got team members who are ready to hop online with you and help you think through both the strategy and the technology of the build out. So we’re gonna talk a little bit with John about how we can leverage AI tools, not just specifically his, but how we can talk, how we can leverage those tools to connect the dots for clients that will help us, help us define and explain and support the ROI of the spend that they have with us. So without any further ado, let’s welcome John to the show. John, welcome to the podcast. Thanks for joining us.

Thanks for having me. Glad to be here.

So give everybody a sense of kind of your background and your expertise before, before we dig into the conversation.

Yeah, I mean my, my deep background is I was a electrical engineer originally designing televisions for Sony Electronics, but I moved my way through like Sonos the wireless audio company as a product manager, you know, from when they were small to when they grew kind of large and then did a 3D printing startup. And now I’m the co-founder of a company called ao. And we make analytics software, AI enabled analytics software for agencies. And it’s a, you know, it’s an interesting time in that AI analytics space and, you know, our, our kind of firm belief is that, you know, if you can make things easier to use, more people will use them.

And, and my background from consumer side, you know, that’s what I’m trying to bring what we collectively try to bring to the, to the agency spaces, like tools that are easy enough for everybody to use type of thing. So that’s my sort of deep background product and engineering and the but but now building, building a company.

So you’re gonna have to keep it super simple for me. Don’t have an engineering brain. So yeah, you’ll have to, I’ll, you’ll have to dumb it down for me a little. So let’s talk a little bit about, I mean, everybody on the planet is talking about AI right now. It, it’s like, even though it’s been around forever, it’s like all of a sudden they think about a year ago everyone was like, oh, AI’s a thing. And I think it was like the whole world discovered Chad GPT on the same day and all of a sudden that sort of triggered it. But I think agencies have been trying to figure out how to, how to think about AI in terms of their business model for a long time. And I think particularly around data. So, you know, one of the pain points for agencies, boy for the last decade has been how do we show our clients what’s working and that when every, every time they give us a dollar, they get $2 worth of leads or visits or sales or whatever the metric is.

So let’s talk a little bit about how you envision AI kind of intersecting with data to help agencies prove their worth to clients better.

Yeah, I mean, that, that’s kind of the really interesting one. One of the really interesting components of sort of where language models have enabled, like content creation for the creatives brainstorming on copy, they’ve, they’ve also enabled the ability to start to interact with data in new ways. And you know, like that can be as simple as allowing someone to ask a question of a dataset and get the answer automatically without having to know SQL or code in order to extract that from the actual dataset, right? To like much more complex operations. You could do data transformations and pipelining and stuff like that. And so, you know, like the, the value prop of any agency to a client, and I I’ve worked with many of them on the like client side as well, is that they’re gonna help your incremental ad dollar go further, right?

They’re gonna help you drive a better return on your advertising investment and as a client, of course, like you’re monitoring that result. And, and that’s the thing that you’re after in the agency engagement. And so all these circumstances start to arise where, you know, executives at the client side suddenly have questions like, what’s the latest, you know, performance, what’s my return on advertising spend? Which ads are doing the best, which ads are doing the worst? And as an agency, these, these questions often get run as interrupts. You know, I’ve, I’ve been guilty of doing this to my agencies too, like, you know, it’s, it’s Friday afternoon and you know, we have a board meeting coming up and I’ve got a couple of questions about how the performance is going.

And, you know, some, somebody on the agency side has to scramble and, you know, has to, to maybe get into Excel and put together some pivot tables. The old way of doing this is actually quite hard, you know, generate some charts and bring those back into some sort of PowerPoint format or email format where the, where the client can actually see the impact of their services, you know, that that’s gonna change. And I, I, I’m, we are strong believers that AI is gonna enable like a shared analytics communication surface between agencies and the clients. Where now you can do things like just give a chat endpoint to your client and they can simply ask those questions that they have directly and self-serve for something like 95% of the answers.

And then, you know, of course you can also use that same like shared surface to create live dashboards that automatically update so that like at any point in time the value you’re delivering is, is like extra clear and the clickdown questions are easy to get to. So there’s like, no, there’s very little confusion or or lag between delivering performance and the client’s ability to, to see that impact. And, and the more you can demonstrate your impact to clients, like the better, obviously, right? ’cause then you’ll retain them and, and be able to grow your business. But But it also, you know, it enables stuff beyond just showing your impact in an analytics surface.

It’s also getting to the point where a lot of the things where you’d previously need a data scientist to do right, extracting the patterns from the data that are driving your key outcomes are getting easier and easier for anyone to do. And as that happens, you can also start to leverage AI and machine learning to actually optimize that ad spend. So you know, the, the wins are out of the data twofold. One, you know, making your value prop more obvious to the, to the client and, and more accessible, but two, actually like delivering a larger impact directly to them in more cases by being able to, you know, both surface the patterns and the data and then leverage those into better decision making on how to spend the ad dollars.

Yeah, I think, I think part of the concern, like I’m sure when you said self-serve, let the client get in there, click around whatever, I’m sure the agent, the people listening had two reactions. Number one, we can’t even get ’em to look at the report. We send them that we, that we basically spoonfeed to them to understand the data. So number one, I can’t get ’em to look at the data now, why will an interface make it easier or better? Or they’ll do it. Number two, they don’t understand it. So how, how are the tools or what do I have to do as the agency to help the, you know, having a client click around and see a bunch of numbers that they don’t understand just causes more headache and angst for an agency.

So how do you envi how do you envision the tools helping with that part of the problem?

It, I mean, a actually it’s pretty fascinating. So the work you would do as an agency to describe what the numbers mean and why they matter, you can automate that bit too. So, you know, if, if you create a chart showing, I don’t know, click through rate or, or cost per click over time you can, you can actually use AI to not only generate the chart, but to explain what’s interesting about the chart, what the trends are, the chart in the chart are, and how to think about the chart. And you know, honestly, like some charts are pretty confusing. Like, you know, simple wine graph over time is pretty easy, but there’s lots of charts that take a lot of thinking in order to really understand what they’re saying.

You can actually start to use AI and large language models in particular to parse those charts and actually describe what they say and what the takeaway should be. So, you know, it, it goes beyond, just give me an answer to how we’re doing into explain how I should think about that answer and why. And, and so that’s where you really start to see the benefit I think is, is beyond just the, you know, here’s a bunch of data to look at, right? But, but here’s how to think about it. And, and we have people even writing like summaries that you could just read that say how, how’s it going, why and how to think about it in addition to like, here’s the like more detail of the actual performance you can look at.

And, and of course client side, there’s a range of like capabilities, right? You, you, you maybe have your, your marketing team and, and somebody on there who’s in demand gen, who’s, because of their need fairly technically sophisticated at looking at analytics and charts and stuff. And then you have an executive who’s just wondering like, what, what is my ROI on the spend? Is this agency still worth it or not? Type of thing. Yeah. And, and you can really start to give audience appropriate explanations at each level in a more automated fashion, which is, which is quite impactful, right? Because it lets you tailor that message But, it lets you do it almost as easy as like g PT lets you write copy, right?


And that’s where it’s going is that, you know, understanding the data is going to be easy regardless of your technical skills. It’s beyond just showing you a chart for sure.

So do you envision a time, so I I think one of the concerns agencies have had over the last five years is they don’t have anyone on the team who can not only crunch the numbers in a time efficient way, but more importantly under a, a, understand which numbers to crunch B, crunch the numbers in a time specific way, and c analyze the data. So a lot of agencies have been thinking, God, do I really have to, do I have to really budget for a data analyst on my team of 25? So do you see AI removing that concern for agencies that we won’t need that level of sophistication on staff because there will be tools out there that, that deliver that

I’m not sure I’d buy removing data competency as the outcome here. Right? I think, you know, I I think part of the way that you make the next advertising dollar do better than your last advertising dollar is, is by having some people who are at least a little bit able to think through the value of the data or how to leverage it, right? I certainly believe that the people who do that deeply today will be able to do 10 times as much work in the same amount of time. And an audience of people who like really haven’t dug in as deeply in the past will be enabled to, to do a lot more, a lot more easily.

And so you might find a world where your, your ability to really, really leverage data to make decisions and communicate with customers will be, you know, an order of magnitude better than it has been in the past 10 x or something. But, you know, I I like to think about like, my, my analogy is like when Excel came along, right? And, you know, back, back when Excel came along, it made working with data much, much easier for a lot of people. Did it get rid of the need for people who are data savvy or competent? And actually it generated a need for more of them because suddenly a lot more data could, could be leveraged to make good decisions. I suspect we’ll see something like that here where more people will be able to get more value from data and, and the competitive environment will require you to do so.

So, you know, so I think I suspect that much of your differentiated value you’ll, you’ll offer in terms of like optimizing advertising spend, you know, half of that’s like fantastic creative and half of that’s making good data driven decisions, figuring out like the right audiences, the right channels, the right, you know, calls to action. Those are both gonna be key, key skillsets going forward in my opinion. So yeah, I I didn’t think get rid of the data people.

So you’re saying we’re still gonna need someone with that level

Of expertise? Yeah.

So what’s, what, how does, does the agency of 10 do that?

Yeah, you know, I’ve seen it a couple ways. Like, you know, sometimes they have as a member of the team, like a, a data savvy person, sometimes that’s a co-founder. You know, I think, I think as you scale up to about the 10 person level, you typically have a data savvy person. Sometimes they leverage other agencies for the data analytics piece. I’ve, I’ve seen that happen too, where, you know, there’s an agency who’s value prop is entirely to other agencies of doing the data analytics for them sort of on demand. So there, there’s a variety of approaches that are possible there. But mean, i I suspect most 10 person agencies have at least some data competency already inside of them.

And, and that person’s gonna be wildly more enabled or, or the group of people who are like, you know, doing the early task, right? Like even even at the pitch side, you know, you get access to like the Google ads account or something. Your first job is to like put together a report on ideas of what you could do and how you’ve been doing and where it’s going wrong. That that’s a data-driven task. Somebody’s doing that already, even if you’re 10 people.

Yeah. So your assumption is every agency is doing a lot of digital work, which I don’t think is the case, but I do think every agency’s gonna have to understand data and ROI if they’re a brand shop, they’re a PR shop. So maybe different, maybe a different topic altogether, but fair enough,

If they’re in the, in the advertising marketing dollar spend space

Yeah, right. Then you definitely, then you’re right. Then somebody’s gonna understand how to crunch the numbers. But for a lot of agencies that’s tangential to their core deliverables and so they’re really struggling with it.

Yeah, yeah. And you know, for them it’ll also get much easier and Right. You know, to the extent that they, that they demonstrate value to their clients through showing performance metrics, the ability to put those performance stories together, it, it’s gonna be 10 times faster than it was before. Yeah. It took you a day, it’ll take you an hour. Yeah. Take you 15 minutes in some cases.

So as we think about sort of how, where all of this is going, where do you see the leverage for agencies? Where, where is this? So we talked about the fact that they will be able to look at data in a deeper way and be able to sort of talk about ROI better, we’ve also talked about the fact that they will have a more sophisticated data analyst probably in the tools than they would on the team team But it. This also has to be a time savings, right?

Yeah, yeah. So, you know, one, one of the things we do in, in our company is sort of require or mandate that everywhere you possibly can, you leverage AI tooling to do your job because it makes you more efficient to doing your job. So, you know, our software engineers use AI to help them write code copilot and it makes ’em write code, you know, almost twice as fast for the good ones. They’re able to, they’re able to do like twice as much work. Our, you know, our team that’s doing creative work uses it to help brainstorm creative assets. Yeah. I I don’t know that it’s quite at the place where it’s replacing creative work.

I I, I don’t really think it is great, but as, as far as brain, which

Of course has everybody freaked out.

Yeah. But, but really if you look at what’s generated by it today, you know, you could kind of tell it’s, it’s, it’s only as good as your interactions with it to assist you. But it’s a great brainstorming tool, right? It, it’s a, it’s a tool that helps you do a lot of the busy work so that the, that the, you know, the, the way I would think about it is it’s like an assistant really, and it helps you do work that you would have to do, but isn’t the key value creative or digital analytics stuff. And then on top of that, you know, you bring in sort of your expertise and that’s where the value’s created. So, so everywhere up and down the stack though, I mean, even if it’s just checking the grammar and, and like something Yeah.

Gets, gets better faster. And so that means you can do more with less resources and, and you know, that creates a bunch of interesting outcomes if you think about it. And, and I think, you know, that these models are just gonna get better and better

Of course. Yeah.

So, so we’ll see where it all ends. But you know, it, it does, it does suggest some questions about the billable hour as a concept, right? Because, you know, if you get 10 times as efficient at delivering val