Joining AdQuick

Alex Kracov: So I love billboards, and I found them to be super impactful in growing Lattice. So I'm super excited to chat with you today, Chris, about all things billboards. I'd love to start today's call with talking about the origins of AdQuick. Can you talk a little bit about how you ended up working in AdQuick?

Chris Gadek: Yeah, I guess before we - thanks for having me, first and foremost. I love what you're doing here with the pod. I think there's a need for this podcast, actually. Not a lot of B2B thought leaders coming out and sharing their wisdom. So kudos to you on this, Alex. But yeah, so I guess to get started, AdQuick is an out-of-home advertising marketplace. We make it easy to purchase, plan and measure the medium. We plug into a lot of analytics solutions and omni channel solutions, so you can then understand the impact of out-of-home.

But I guess the story of AdQuick is very closely linked to the story of Instacart, the grocery delivery business. And so our founding team came out of Instacart. Our CEO was in charge of launching new markets. At the time, having somebody go do your grocery shopping for you was somewhat of a crazy idea. I'm sure you remember that. Very much like Uber. And so one of the things that they had learned on the expansion team, of which my CEO was a part of. I think he was number 15 at Instacart. So very early days. One of the things that they learned is, they had the chicken-and-the-egg problem. They had a couple sides of their marketplace that they had to of get up and running. In all their experiments across digital channels, TV, and various other formats, they found that out-of-home worked really well. And for Instacart, at that time, they had to acquire new - the supply side partners. Basically, the stores that the shoppers would basically be operating in. Then they had to find new customers, people willing to participate in this experiment. And so they found that it worked. In launching I think 20, 25 markets, our CEO, Matt, had the realization that, hey, there's a need for the ability to buy this stuff easily on the internet. In the long term, it should be measurable and we should understand its impact. And so that was the nexus of the creation of AdQuick.

In 2017, we had gotten our first initial customers, notably Peloton and Squarespace. That was enough to garner the interest of initialized capital. We raised our seed from Garry Tan and Alexis Ohanian. Garry is now President of Y Combinator. Alexis Ohanian is one of the loveliest men in the world and also Founder of Reddit. But yeah, so they raised money. They started thinking about how we're going to get new customers. And that required somebody with my skill set. At the time, AdQuick was working on getting advertising inventory on Lime scooters. So everybody's favorite transportation mode from the late 2010s. We just hired a head of sales. There was a demand generation, lead generation issue that most startups encounter.

And so the GM at Lime just happened to work with somebody that I was close with, and they've put us in touch. Like, hey, this guy has these shops. He went to Reforge. He checks all the boxes - worked at marketplaces, worked in SaaS, worked in enterprise B2B. Fehim, the technical co-founder, flew up to San Francisco. They left San Francisco and started the company in Venice Beach. He flew up. They showed me the product at Phil's on Market Street. And what was supposed to be 30 minutes turned into a few hours of nerding out and thinking about how we can grow the initial customer base. I basically told them, like, I'll see you guys in two weeks. I picked up my life in San Francisco and moved down to Venice Beach. I got a place right on the ocean, and the rest is history.

Alex Kracov: It's amazing. And so what was the company like when you joined? How many people were there? There was a working product. But do you remember what your first couple of weeks or months look like? Did you just throw up a bunch of billboards to try and grow this thing, or did you have other tactics? How did you kind of approach that growth challenge in the early days?

Chris Gadek: Sure. So early days, it was very engineering heavy. I think I was employee 10. And so we only had two salespeople, a head of sales, and most were developers scattered across countries. We did remote before most of the world went remote, so that was kind of cool. But in terms of early traction, we had those early wins. I think within the first two months, we had lost our largest customer to an agency, which was fucking terrible. Then subsequently, we lost our second customer to another agency using the same tactic. And so, really, I honestly had to kind of start from scratch. I joined in 2018. The preceding year's cohort kind of fizzled out in terms of revenue retention. And so we had to pick it back up. And so not coming from the advertising world, I had to learn about the market as fast as possible. Then obviously, Alex, you've been based in San Francisco, so you know what the footprint of out-of-home in the Bay Area is. It's all startup stuff. But that's not universal. That's not the case in other markets. When you live in your little tech bubble, you're like, "Oh, people advertise other different types of products in different cities." That became so apparent when I moved to Los Angeles. Then I realized that there were no software billboards, and it was only entertainment billboards and ambulance chasers. Right? So it was kind of eye opening.

And so when thinking about growth, I put myself in many digitally-native marketer's shoes, which is understanding what are the questions I have about this medium. The early days of growth were building out engineering as marketing tools to answer the most common questions really quickly. It's important to note that the industry itself is very sales-driven. And so not a whole lot of content out there for people to do self-discovery, self-education historically has been gated behind somebody that you need to talk to via email or phone call. And so we did a bunch of research, scraping Quora, scraping Google, and we kind of quantified the amount of questions based on the topic. We ran some clustering algorithms over that. Then that basically informed our strategy of what tools we should build for people like ourselves who go and self-educate and then try to find the answers themselves online. It turns out that 40% of most queries around out-of-home and billboards in general pertained to cost and pricing. And so we built, in the early days, one of the first things we did was build a calculator. That ended up swallowing a majority of the searches on Google. That's how we kicked it off, in addition to running some ad programs in the early days.

Alex Kracov: That's awesome. It reminds me a little bit of HubSpot's playbook with the website grader, if you remember that thing, where you can put in your website URL and then it spit out a bunch of information. Then it's like all your website is crappy. Your website is great. Then it's a bunch of just lead gen for HubSpot. It's super smart that you can detect that same playbook for the billboard side of things.

Chris Gadek: Yeah. And to your audience, especially the younger folks in B2B, if you haven't read Traction by the founder of DuckDuckGo, Gabriel Weinberg, and Justin Mares who's got Kettle & Fire, incredible book. It's a good way to think about which levers to pull in the early days and how to quantify their impact. It's a great book.

Digitizing Out-of-Home Campaigns

Alex Kracov: I'd love to spend a little bit time talking about how AdQuick sort of built out this out-of-home network. Because historically, out-of-home had this very fragmented group of providers. Then it seems like AdQuick sort of aggregates it all into one place in this beautiful digital format. There's like 1,500 plus premium providers. How did the companies sort of go about building out this network? How did you convince people? I mean, was it just a giant sort of biz dev sales effort? Can you kind of tell that story a little bit?

Chris Gadek: Yeah, sure. Having worked at multiple marketplaces, it would be remiss not to tell you that most great marketplaces start with CSV uploads. In the beginning, we built inventory ingestion systems. And so every media owner in the space, large or small, had a custom Excel grid that they used with various attribute and value pairs. Basically, we just built an ingestion engine to take all these unique grids and then standardize them and build it into a model on our back end. But those were the early days. As we grew, media owners were also building their internal systems. So in the landscape of the supply side of our marketplace, the media owner companies, there's about three or four large players. They're publicly-traded. They do billions in revenue. There's a smaller cohort. So that group does about 40% of most of the inventory in the space. Then there's this middle tier, this mid-market tier, where maybe they have like 100 to 200, to 1,000 units in multiple markets. They typically just run a good sales process. Then there's your mom and pops where we have just a couple pieces of inventory. So it was important for us to delineate and understand the differences in those supply side participants.

And so we started with working with the people with the most inventory. And so they built out the capabilities internally to give us - at first, we started with lat long, lat longitude, and then the name and the owner. Eventually, that grew into having impression data, having availability, having pricing data. A lot of these developments over time helped us streamline the data exchange. Because one of the biggest pain points, if you ask an advertiser or agency, the biggest challenge or the biggest pain point is the constant back and forth of like, hey, is this available? Are you looking at this billboard in this spreadsheet, or in this PowerPoint deck? It was very disheveled and disorganized and fragmented. And so we just made it really easy to put it in one place. We also noticed that - so that was one subset. Then there's the mid-tier of the market, which they didn't really have in-house software capabilities. And so from previous marketplaces, we had learned that it's just best to build out tools for the supply side to use. We built those tools out so we can get easy access into their pricing and availability. We saw that as a competitive advantage. Because at the time - I think this is still pretty much true - most agencies, if they're doing things the old-fashioned way, they're spending two, three months on a billboard campaign in a couple of cities. And so we built the inventory management piece, the request for proposal piece, the sales tools for the supply side of our marketplace. We gave that away for free. It was a great way of getting early traction. Then the remainder of the longtail, we pretty much just scraped, actually.

The funny thing is, since billboards are a real estate asset, they have to be registered with the Department of Transportation. So one hacky thing that we did in the early days was scrape all these permits. But I think to kind of summarize where this was all headed is, you can buy digital inventory for your marketing investments in a split second. And very much like Expedia did for the airline industry, what we wanted to get, the place that we wanted to get to was, hey, this is available. I know how much it costs, and I know that my audience is a great fit for where this placement is. I made it sound super, super clean but it was super messy. I think we had a biz dev team headed up by Katie Burwell who's an incredible woman who has deep expertise and knowledge of the out-of-home landscape. She had been at previous media owners. And so she led those initiatives and one by one doing the things that don't scale, bringing the people online, teaching them the value of the product. Ultimately, that's how we built it up. Now we look for bigger acquisition chunks. We're now in 90 countries. So we'll typically try to plug into other networks. And so we can have the most broadest reach in set of inventory in the globe.

Alex Kracov: It's a super interesting story. It really is a great job of articulating how you actually stand up a marketplace. I love it. It's like, alright. Is this CSV uploaded? Because it's like you got to get on the radar of all these different, the mom-and-pop providers. But then I assume the big players, the clear channels out front, those types of folks. I assume the pitch to them is, hey, we're going to drive you leads in business. Right? And so we're kind of this interface, and then we're going to send you a bunch of business. I assume that's kind of why they're willing to partner with AdQuick.

Chris Gadek: Yeah, and they've been incredible partners to date, the big three. We now have direct APIs. So our systems talk to one another. That was something that we had dreamed of in the early days. All this would be so much easier if they were a software company. But they're not. Some are family-run businesses that have been around a hundred years. Some are spin offs from large conglomerates also that are very old. But they've been working with us, and it's been a great, great - we've been very happy with the progress that we've made in that department.

Alex Kracov: Yeah, it's funny as you're telling the story of how you digitized all this. Because I remember when I bought billboards at Lattice, I was like a beautiful mind. I had my series of spreadsheets and my PDFs to figure out the placement. Then I'm going on Google Maps and street view to be, like, where is this location? Then I even remember me and the two co-founders of Lattice driving down the 101 with a laptop and meeting. Like, slow down. Slow down. That's the billboard we're going to get. Slow down. So we could actually go and see the actual placements of where we wanted to buy. It was fun, but it was quite the process. So it's really cool how AdQuick add makes that all digital. It's just like at your fingertips, as opposed to doing all of that work that I had to do back in the day.

Chris Gadek: Yeah, and we sat down with those users very closely, both on the advertiser side and the agency side, to understand their process, what was important to them. We just took those activities and built them into tech, into features. At first, AdQuick was literally a map with inventories and pictures on it and hooked up to a contracting system. Then we started adding in planning features that made that process a lot faster. Our name is AdQuick. That was the initial value proposition. So something that takes you a couple of months, you can now execute within the span of a week and a half. But we layered on the ability to do demographic targeting, zip code-based targeting, points of interests. So, hey, find me every billboard near Starbucks. So we plugged in a bunch of APIs. That powered the location intelligence there. They made it super easy to do, like, hey, I want all the billboards within two miles of a Carl's Jr, or something, or a Taco Bell. We're able to build out those functionalities. Then most recently, I guess the final piece is audiences. The digital audiences that you use in your day to day for your marketing investments are now readily transferable to use for planning in the real world, which is kind of neat.

Measuring the Impact of Billboard Campaigns

Alex Kracov: Really cool. Can you talk a little bit about how attribution fits all into this? Because I think AdQuick has made some really good strides in actually how do you measure the success of billboards. And you're telling me before this. I think COVID in 2020 and all of that was kind of a good kick in the butt for AdQuick to start building all this attribution stuff. Can you share that story?

Chris Gadek: Sure. Yeah, this was a wild time. We had raised the series A February 2020. March 11th, the world shuts down. Everybody's locked in their homes, and nobody knows what the price there where is it at. Because everybody's locked in their house. There's only a subset of the initial audience that was seeing these ads in the real world going forward. And so in the earliest days of the pandemic, one of the things that we wanted to answer for the supply side of our marketplace and we wanted to be helpful is to understand what percent of traffic is still around post-lockdown. And so we started developing technologies in which we would pump in mobile device IDs into the platform and then start to track exposures to certain placements. When we rolled all that stuff up, we were then able to go to each one of our media partners and be like, "Hey, your inventory is in this town. We're seeing 60% of normal of pre-COVID levels in terms of mobile device behavior. So that was something that we can track. Basically, the suppliers didn't know what the price or where it is at given the changes. Once we developed the exposure modeling, then we just needed to map that to the conversion. And so in the early days, there's a whole suite of ways that we can measure now. But in the early days, it was map the exposure to the location and then have a pixel that catches it on the other end and can basically inform any success event on a website, on a mobile app, or in-store.

In thinking about out-of-home measurement, just like everything else. I know that grow at all costs is kind of gone. But growth marketing should still be part of everybody's repertoire. And so we usually tell our clients start test measure scale. Given that out-of-home is a reach channel, any campaign is inherently going to reach audiences with varying degrees of intensity. Some people may have not heard of you. So therefore, they're learning about you for the first time. Some people might know who you are, and then they'll click through. Right? They'll see your billboard. They'll take action. They'll go do something on your website or your mobile app. But for us, it was super important to build a full suite of tools to basically understand how this impacts the entire funnel. And so we do that through a variety of ways, brand lift surveys. We do organic search lift. In the early days, we started off by quantifying out-of-home's impact on your digital channels. So if you are running a treatment in a market and you're also doing paid digital programs, those will perform a lot better in a market in which you are not running out of home. That's just an accepted truth. But we've evolved into now doing statistical modeling. We plug into most MMMs, MTA models. And so basically, wherever you operate, you can effectively measure your campaign.

Alex Kracov: Yeah, I think there are so many just benefits to billboards. And so many things you said there resonate with me. When we did that at Lattice, we didn't use any of the fancy attribution stuff that you're describing. But what we did, the one thing we did was just track the lift to our website and our brand traffic. And we could tell, okay, something is definitely happening here. Then we would place it in different markets and be like, okay, why is web traffic going up in Seattle, or New York, or whatever? It's like, oh, because we put the billboards there obviously. Right? We were able to track that. And so we did a bad job of connecting it to the lead and the account. But then you would have it overall and get a sense. The way I always thought about it was, like, it just makes all of our marketing better - the outbound emails we're sending, the direct mail we're sending, the digital ads we're sending. There was such a qualitative benefit that you could just feel. People would get on a call and be like, "I saw your billboard," or, "The employees got all excited." There was just this energy that you could feel anytime we did a billboard campaign where it just felt like it was working. I know that's terrible to say in this world of marketers who are obsessed with attribution, but it really did work. You could just feel the momentum that it drove in the business. I think there's so much intangible stuff of just feeling bigger than you are, building a big brand seeing legit. We did our first billboards at Lattice, I think, just after series A. It was like, oh, wow. This is like a real company. That was the impact we wanted to have, and it did that.

Chris Gadek: You know, that's a perfect time to start thinking about making that big splash. I went to school in Northern California. One of my professors from my MBA - I did an MBA with the data science. Not very relevant to startup life but relevant for running a company. One of my professors, Chuck Byers, he was the Head of Comms for TSMC, Taiwanese Semiconductor Manufacturing Company. He showed us this old ad from McGraw Hill from the '50s. It's like some cranky, old man in a chair. He's like, I don't know who you are. I don't know your company. I don't know your product. I don't know what your company stands for. I don't know your customers, your record. What is it you want to sell me? I think that when you have a well-placed billboard, you can shortcut that a whole lot. If you want to have a successful B2B business, you have to be able to answer those questions confidently. And I think if you have a well-placed billboard and a well-designed website, you can really shortcut those conversations and shrink your sales cycle.

Alex Kracov: I think why they're so impactful for B2B, I think most people think, alright, they're really good for consumer because everyone can buy the product if it's consumer. So huge reach on a billboard. I think it's so good for B2B, because every B2B company is so focused on their persona. Like at Dock, we're selling into sales and customer success. Lattice was HR. But that's only one important persona. There are so many other people at the company: the CEO, the CFO, the employees, all these people who might not really know about you. But if a billboard goes up, they probably saw it. Then when your order form goes to the CFO for approval, they're going to be way more likely to be like, "Okay, this seems like legit. I know of this company for some reason." They don't even know why they're thinking that. But I think things like just big brand advertising in general can be so impactful there. So I don't know. I'm a big fan. I can't wait to do it at Dock.

Chris Gadek: We're now starting to see the last batch of series A companies start thinking about splashes in the market in May and June. So a lot new billboard campaigns coming to the Bay Area soon.

Tips to Run Your First Billboard Campaign

Alex Kracov: Okay. So series A is a good time to start thinking about it. Is there like a minimum spend or size of budget you recommend? Maybe it's like if my marketing budget is a million dollars, should I start thinking about doing a billboard? I don't know. Is there any rules of thumb for marketers out there? How do you think about it?

Chris Gadek: Yeah, so it's important to note that out-of-home, first, a real estate asset. So real estate varies from market to market in terms of its pricing, in terms of its reach, so on and so forth. And so the nice thing about having all the data about all the inventory in a given market is, we can quantify the amount of available impressions in the market. And so the way we think about how much should we spend in the market is actually a percentage of share of voice achieved over a four-week period. Most billboard campaigns are running for a week increments. And so we typically have recommendations for here's an entry level 10% share of voice. So basically, you're getting 10% of the available impressions in that market for a given period of time. There's a saturation which is 25% and then domination, which is you're getting 50% of all the impressions in the market.

Just to give you a sense, I have this cheat sheet pulled up right here to make it super easy so I'm not bullshitting you. Obviously, New York is going to have the largest entry point. If you don't have a half a million dollars to spend in New York and you want to do a broad reach in campaign, I probably advise spending that elsewhere. Because you can enter a market like Philly for 150k. Some cities, you can literally take over for a couple $100,000 for an entire month. I think the way to start thinking about it is which markets have my audience there first. Then you can arbitrage. Okay, well, this has a high concentration of the audience that I want to reach, and it's a cheap market. And so when you start to work with us, we typically in the first couple campaigns, you run with us. We operate kind of like a concierge white glove. We know that a lot of folks have bought either billboards the old-fashioned way, or they haven't bought them at all. And so we take this into account. We handhold when you're coming into the market. But basically, what we tell our customers is like, you should know who you're trying to reach. We can help you find them in the real world. But you should know your personas, and you should probably have at least some budget carved out to hit that entry point. So for most markets, that entry point is like 50k. There are options, lower price options, for pop ups and experiential type marketing programs that can be had for less than 10k if you're very early days. But usually, we recommend if you're doing stuff like that, we probably recommend that you just do it for a conference or something smaller like that instead of trying to run a market-level campaign.

Alex Kracov: Yeah, it makes a ton of sense. So start with the city where you think your personas are living and then kind of go from there. Then obviously, New York is going to be very expensive versus something like Seattle or Austin. It might be a little bit cheaper. So it makes a ton of sense. When I've done billboards, there was like two parts of my brain going. There's the placements, right? Like, okay, I've selected my city. But where am I going to put these billboards? Is it going to be an actual billboard? Is it going to be a bus shelter, an airport transit? There's all those things. Then there's the creative side. Where do you recommend people start, or is it just all got to happen together, and you're figuring it out together? How do you think about that?

Chris Gadek: When we're speaking with customer or prospects, we usually tell them that your audience dictates your placements, and your placements dictate your creative. What do I mean by that? So your audience is dictating your placements. Let's say your audience is HR, an HR professional. They over index in Austin, right? Maybe there's certain types of inventory types that are very close to where that subset hangs out or lifts. You might have that audience not exist in other markets or be less representative. First, we select. We decide where the audience is. Then we understand what is the makeup of the placements around there. Then there are rules of thumb for different formats. Obviously, out-of-home is more than just billboards. It could be train ads, ads at bus stops, ads on buses, ads in taxis, ads wrapped cars. You name it. It could be planes in the sky. It could be a whole lot of things.

Ultimately, how a user interacts with that ad - let's call them users for this purpose. There are people out in the real world. How they interact with an ad will be contingent upon where they are relative to the placement. So for example, an example of a placement dictating creative would be like you wouldn't put a couple sentences on a highway billboard when people are speeding by at 65 miles an hour. If you find that your audience is exposed to that placement and that placement happens to be a 14 x 48 billboard, your creative is going to be - it's going to be important that you're very succinct with your creative. So probably like four or five words max. You're going to want to have a lot of high contrasting colors, right? So you want the message to pop. And ultimately, you don't want to confuse. You want one message or one takeaway. And so that changes when, let's say, somebody's waiting on a bus at a bus shelter, it will have more time on their hands. And so you've seen brands like Oatly write paragraphs. The nice thing about out-of-home is, it becomes your canvas once you start thinking about where the person is throughout their day, what they can engage with, so on and so forth. Those are probably some good rules of thumb for thinking about how creative should be closely intertwined with what placements you've selected.

Alex Kracov: When you come up with the creative and the messaging for these billboards, how cute and quirky do you need to get? Oatly is a good example. Very different and out there sort of campaign and weird messaging to stand out, right? But then, I feel like there's other brands. Intercom, I think, has a billboard now in SF where it's just like, "We're the best customer success platform. We're doing awesome in AI." It's like literally just saying what they are. How do you think about that balance between being sort of quirky and just being on point with your messaging and just say here's who we are? I don't know. How do you think about about that?

Chris Gadek: You know, I think about it the same way as like a sales pitch. You don't want to be rambling too much on the ad. There's a couple ways B2B companies could think about out-of-home placements. One is the flex, right? "Hey, we're the best," or, "We're number one in the industry. We have the largest share." You see companies like Salesforce and Oracle been doing that in print, in TV for decades, right? That's one tactic. Another tactic is, "Hey, we solve this problem." And I think B2B companies often forget that they solve a problem, and they forget to include the problem in the creative. So one of the most recent ads - I mean, this was a TV ad, but it was done by a B2B company. I think it was CrowdStrike. It's like, "We stop cyber attacks." It's like three words. It tells you what they do. You know the brand name. You can go and learn more. But ultimately, you got to keep it simple. I think communicating what pain point you solve if you don't have that mass market awareness is probably a good way to start thinking about it.

Alex Kracov: Do you recommend people test creative ahead of time, like run a bunch of digital ads with the same billboards and kind of see what - I mean, same messaging that'll go in the billboards and see what works? Or is it like you want to keep this private and special and keep it in the billboard? How do you think about that?

Chris Gadek: Oh, no, no. We definitely encourage testing your messaging. The only caveat is we got to make sure that the audiences are the same and the only thing that changes. The design is the same but the only thing that changes are the wording. And so when you run a proper experiment on digital by pumping a bunch of ads at a landing page and then seeing which one had the higher click through, that's usually a good way to start. Being like, okay, this is going to resonate more. So we do have customers that have different personas or different value propositions for different markets or different cities. Carvana is one of our largest clients. We'll run a 200-city billboard campaign. They have a very sophisticated quantitative marketing team. They'll have the messages that land with the best people or the people in that market. And so a lot of the testing basically flows into that.

Alex Kracov: Let's say a marketer is listening to this call, and I want to go convince my exec team to go run a billboard, the CEO, CFO type of persona. How should I go about convincing them that this is a good idea?

Chris Gadek: Well, first and foremost, you could tell them that we can measure it now. That wasn't always the case, the type of thing where it's like, oh, I'm spending all my money on advertising. And I don't know which half works or whatever the saying was. But aside from it being more measurable if you compare it against other channels, out-of-home has the highest recall, most memorable. It's hard to remember the ads that you saw on Instagram and all the ones that show up in feeds. But for some reason, people are way better at remembering that they saw an ad in real life. Then the second piece is that it has the highest percentage of users that take an action after seeing the ad. Most of the time, that's going to be a Google search. And for all the noobs, people that are new to out-of-home or thinking about out-of-home for the first time, just let me save you the exercises or the experiments. Vanity URLs don't really work. QR codes don't really work. But you know what does work? Having a compelling message and then having your brand well represented on the billboard. Then they're going to search for that on Google. That tends to work pretty well.

The other selling point is, we now have - a lot of the concerns in the early days was, I'm wasting a lot of money on people that aren't going to receive my message or won't act on my message. And to that, we've built out audiences. Basically, you could find behavioral - I can't speak today. It's Friday. But behaviors exhibited in the real world online can now be used as audiences in AdQuick. So if you have an audience provider through LiveRamp, TransUnion, Experian, wherever you get your audiences from first party, you could upload it and be like, this is where my people hang out in the real world. Then usually, you can use a slider to optimize for that audience against a set of inventory. And it'll basically score them and then start removing the ones that don't fit. And so we're getting smarter about how we think of this mass reach channel, and we are making it performance driven and then we could prove the incrementality of it, too.

Alex Kracov: I was trying to think about how I convinced Jack and Jay Zachary, the CEO and COO of Lattice, to do billboards. And Jack was the CEO. He thought it was fun and exciting, I think. Even in my first week of working there, he's like, where do we get to put up the billboard? He thought it was fun and exciting. But I still had to make a case. I think what I did was trying to bring them along through the process. So there was like as I was starting to explore, showing them the placements, building decks to show how it's going to work, how much it's going to cost, here's what the creative is going to look like and try and do as much work before I pulled the trigger on the spend. We also started small. I think in the first year we did it, we started with just two billboards in SF. Then the second year, we did a bigger buyout with transit, the bus shelters all around the city. Then we could see that working. Then we had a giant campaign actually planned for COVID that never ran unfortunately. But it was like this slow incremental build and then also bringing along with the process and then showing it in context of the overall marketing mix of stuff we were doing. It was like, hey, we're doing all these other things in marketing, and billboards accounts for 3% or 5% of our overall spend and stuff. So this is just one of the many ways we're getting people. And I would always describe it as - I always use war as an analogy. It's like billboards in brand advertising is sort of like the bombers who are going out and doing cover for the people on the ground, the sales team, the digital ads, all those things. So that's kind of how I thought about it at the time.

Chris Gadek: That's a great way to think about it. I guess another way that you can also try to convince your C-level team or your executive team of getting some budget for billboards is to pull in a heavy hitter who's had success with them. Where were you in 2015? Were you in San Francisco?

Alex Kracov: Yeah, in those days, it was like Uber for everything. Everybody was building a logistics startup that was doing something - delivering massages, laundry, whatever, packages. I happen to be a part of Doorman which was a startup that did on-demand package delivery. We ended up going B2B. Then my first experience of trying to buy billboards was, we had Andrew Chen who was, at the time, leading supply side growth for Uber. He was advising our CEO on solving that chicken and the egg problem. Because I had been brought on to launch two new markets, Chicago and New York. Andrew was like, "Hey, these work really well." Then I went through that whole process. And so we didn't end up buying because it was just - the experience was like pulling teeth. Then it turned out that it was way more expensive than we had originally anticipated. Because going back to that earlier part in the conversation, there wasn't really a whole lot of information out there to self-educate. And so one good way was to be like, "Hey, they're successful. They're seeing success with this. Not really quantifiable, but they said it worked." So that's another way to think about it.

AdQuick for Agencies

Alex Kracov: Switching gears a little bit. You've mentioned agencies a few times in this conversation. And I'm curious how agencies play a role in AdQuick's own growth. Are you partnering with them? Do you have sort of a big effort to just work with agencies and kind of sell through them, or do you focus more on working with companies as a mix of both? How do you sort of think about that? Because I imagine it's very different ICPs that you have to eventually target or at least work with.

Chris Gadek: Sure. Yeah, agencies. We want to be the plumbing for everybody. So irrespective if you're an advertiser, an agency, we have a value proposition that transcends those two personas. But we think about agencies in the following way. There's the large holding company networks: your WPPs, your IPGs. Those are slow moving conversations. Everybody in B2B knows that the larger the company gets, the longer the sales cycle gets, the more consideration it is. A lot of the times, these are multi-year discussions. And so we haven't really worked a whole lot with those. But there is a very large mid to long tail of agencies that we do work with. And so there's two buckets that we kind of put them in. One, are they specialist agencies? These are people who are pros at buying out-of-home. Albeit the old-fashioned way, they probably have a list of 200 media salespeople in a spreadsheet somewhere. And they do a lot of the stuff kind of the old-fashioned way. We've seen a lot of those folks just put all their media planners on to AdQuick in a white-labeled instance that shortcuts their entire workflow by like 70%. And so we've seen some success there.

But the way that we operate most closely, given that we place an outsized amount of importance on the measurement piece, is we typically work with agencies that are full service but are missing that out-of-home piece. So they'll do like the digital. They'll do SEO, or they'll do TV. Then they don't really have an out-of-home partner. So we'll typically work as like a concierge or a managed service and help them deliver the best campaign to their clients. So we don't have a preference. But we do prefer that all campaigns are measured so we can kind of start to understand how the channel impacts different business models, different companies of different sizes and across different use cases for launching. So it could be like a product launch, geo expansion. You name it. We wanted to collect enough data to have benchmarks, so we can further excite the market about the media format.

The Future of Out-of-Home

Alex Kracov: Can't do a podcast these days without talking about AI. And I know you've been running some AI experiments. Can you talk a little bit about AI, what you've been playing around with? And I don't know. Do you think it'll have an impact in the out-of-home industry? How do you think about that?

Chris Gadek: Sure. So, Alex, just to confirm. Are you talking about how we're using AI for AdQuick products or how we use AI internally as part of our growth team?

Alex Kracov: Both. I'd be curious with both.

Chris Gadek: So we've been working closely with probably five or six YC companies in the AI space over the last 18 months. I think the channels that have been impacted the most by AI have been our SEO content writing, obviously. We're doing a brand-new refresh. We have 14,000 pages on the website programmatically built that feed off of our database. Then a lot of the content that gets populated basically takes the data. And our database packages it up at the city level and tells a story. So that's one way that we're using that internally. We also take firmographic job title data. So for our sales development efforts, we then take all the data that we have in our CRM. Then we take current events and financial statements and a whole bevy of other data points. We'd be like, hey, this seems like it'd be important to you. Or, this was important to you. And we use that for our outreach as well. So 1/10 of all of our scaled outbound is now AI generated, and we're refining those models.

But in the context of product for AdQuick, I guess you could think of AI in, I guess, a couple phases. I think we're still at the read and retrieve phase, for lack of better terms, which is basically hey, this information exists somewhere. Go find the information and then return it to me. We just shipped to AI Assistant. Its goal is to basically answer any questions - city-leveled market data around impressions, pricing, CPMs, number of participants in market, number of inventory in market, just rolled up metrics that a media planner might or a marketer might care about. So we took that knowledge gap or knowledge base and packaged it up into a custom GPT.

Then we also took - one of the neat things that we've been doing at AdQuick is, most of the industry is not set up like a B2B revenue organization. So what I mean by that is, you'll have full cycle sales for most of the participants. So they'll open up the account, and they'll do all the mid funnel stuff. Then they'll close the deal. We kind of separated that out to resemble the businesses that you and I have worked at historically. And what we've done is we've done taken all of our Gong calls. So we've been recording our sales calls, our strategy calls, since, I don't know, like 2019, I think. Gong, by the way, is the best piece of software to come out for us since Segment. No joke. And so it's been super helpful. So we took those conversations. We got the transcripts out via API. We package them up. We built a knowledge base, and then we stitch that together with the market level data. So now we have an AI Assistant that you can now install in Slack. We'll drop the link after.

Alex Kracov: Yeah, I've been playing with it. It's been fun.

Chris Gadek: Yeah, and it's been rough because open AI has been - they've been shipping a lot, so a lot of stuff has been breaking. But overall, it's been pretty positive. And so we're now getting to a point where all those answers that used to have to crawl or you have to call a salesperson for or look through all these SEO-optimized links on Google, now it just spits out the answer. We're working on refining the answers and still building in controls to make it a more seamless experience and have it hallucinate a little less. Because, obviously, marketers don't want to be reading paragraphs. So it's coming together nicely. In the retrieval stage, I think we're there. And I think the next iteration is going to be actual, the creation of set campaigns given a set of parameters. So like, hey, find all the billboards that are this size within two miles of Starbucks in these markets. The goal will be, hey, you could write that in common text, and then it'll spit out a campaign for you.

Alex Kracov: Super cool. So many fun AI experiments. I'd love to end today's conversation just talking about the future of paid media. AI is definitely a part of that. But a lot has been changing over the past couple years with paid media, like Apple's tracking changes, rise of privacy regulation, death of cookies this year. What does the future of paid media look like in your opinion? Are we just going to be buying more billboards and direct mail and going back to the old world? What do you think?

Chris Gadek: Well, short answer, yes. But individual user tracking capabilities have been limited. We've seen that across the board. And so a lot of folks are transitioning to cohort and interest-based marketing and utilizing universal identifiers such as an email address as opposed to an IP address, because that stuff's getting kind of phased out. And so, basically, the new world that we live in is going to require a lot more data rooms, a lot more data partners. And so, basically, you will have to utilize a lot of data partners. But you'll be able to get the same amount of targeting that historically you've had pre-14 and a half or 14.5 iOS change. So it's still pretty promising. I think we're reaching parity between direct response marketing online and with what we're building, which is pretty cool.

Andrew Chen, if you read his book The Cold Start, The Law of Shitty Clickthroughs, marketing channels get less competitive and less useful over time. The funny thing is, out-of-home has never been connected enough to really use it as a proper national channel. And so for the first time, it's going to be a race to see who can start using this as a national format across the board and I think they'll be able to - basically, we're starting to see growth teams use out-of-home as one of their performance channels. And a lot of the iOS privacy changes don't impact that at all.

Alex Kracov: Super interesting. Thank you so much for the time today, Chris. If people want to find you, if people want to go buy billboards, where's the best place to go?

Chris Gadek: Adquick.com. Then you could find me on Twitter. I'm dappermarketer.

Alex Kracov: Awesome. Thank you so much.

Chris Gadek: Thank you, Alex.