How the Abacus Agency Leverages Facebook for Content Distribution

Co-Founders of Abacus Agency Peter Reitano, CEO, and Jeff Goldenberg, CSO, join the Content Pros Podcast to share how they squeeze every dollar out of Facebook using dynamo content.

In This Episode:



Please Support Our Sponsors:

Huge thanks to our amazing sponsors for helping us make this happen. Please support them; we couldn't do it without their help! This week:

Full Episode Details

Finding Facebook Dollars

Facebook and its ever-changing algorithm have kept content marketers on their toes for years. As soon as you find a way to snag a spot on their feed, everything changes and your posts go missing again.

Jeff and Peter have found a way to keep their content boosted to the right people through clever use of Facebook’s own tools, reframing how they use data, and shaping their content around the algorithm’s interests.

Peter and Jeff think of content funnels as conversion funnels. It helps adjust their approach to content, making it more focused on the end result of conversions than just engagement. This puts Facebook metrics and data in a new light where something like click-through rates become a north star for the content that works.

Better content means more clicks which means more pick-up by Facebook which leads to even more clicks and ultimately more conversions. It’s a happy cycle of clicks and sales that every content marketer dreams of starting.

In This Episode

  • Why creating a successful Facebook advertising campaign means spending less money on boosting and more time on content
  • How smart use of Facebook tools leads to conversion rates and data that can magnify the impact of your content marketing strategy
  • Why being a successful content marketer today means marrying new data to new creativity
  • How succeeding at one point of the Facebook Ads Triangle leads to amazing click-through rates

Quotes From This Episode

“We saw the world going specialist rather than generalist.” —@digidharma

“I’ve always been forced to think about marketing in a different way than just spending money and hoping for a result, because you can’t outspend big competitors.” —@jeff_goldenberg

“On Facebook, using click-through-rate as a metric for resonance or context, you can effectively tell which messages are resonating the best amongst what audiences.” —@jeff_goldenberg

Content funnels can be conversion funnels. Click To Tweet

Good content has leverage that can’t be competed on a dollar-to-dollar basis with a bigger spender.” —@jeff_goldenberg

“It only works if it’s authentic, it only works if you’re solving problems for customers, and it only works if you really know the customer you’re speaking to.” —@jeff_goldenberg

Creative thrills but data pays bills. Click To Tweet

“Data helps us segment and map the buyer journey, but creativity is what carries your prospects and customers through it.” —@digidharma

“It’s only if you’re able to do something really well that you’re going to get a good result on Facebook.” —@jeff_goldenberg


Content Pros Lightning Round

For Jeff:

What is a really interesting fact about Peter? Peter listens to really loud aggressive music.

What is Peter’s go-to lunch when you guys go for lunch all the time? He’s always on some extreme diet, and he doesn’t need it. He’s a very, very handsome man. Always on some extreme version of a diet. There’s always something that you’re allowed zero of. So it really depends, but he doesn’t eat a lot of bread. If I could follow his nutrition I’d probably have less bread, but I love bread.

Who would you put more as the art and copy person versus code and data? It’s really hard for us to separate the two.

For Peter:

What is a really interesting fact about Jeff? Jeff is obsessed with a band called Phish. I think he’s been to see them maybe 70 times.

What is Jeff’s go-to lunch when you guys go for lunch all the time? Jeff’s not on a diet. That’s all I can say. He likes good, nice fried food and stuff.

Who would you put more as the art and copy person versus code and data? If I had to choose one, I’m the artist… Jeff’s a nerd.

What Great Brands Do That Good Brands Don't in Content Marketing

Okay content is easy. Killer content is hard. This nifty eBook shows you the difference, based on our real-world work with dozens of brands. A must-read!

Episode Transcript

Randy: Welcome to another episode of Content Pros. I'm Randy Frisch from Uberflip. As always, I've got Tyler Lessard joining me from Vidyard. Today on Content Pros we're going to dig into an area that we don't really talk enough about I think. It's Facebook. It's this big place that we all know is so important, but it's figuring out how do we leverage Facebook today. How do we leverage it for distributing content, for distributing our advertising? The guys who are going to bring on I think are going to bring more of data-minded approach than perhaps the creative approach that we always harp on here being content-minded people. But Tyler, I think you'll agree that data side is just as important, if not more important these days.
Tyler: Right. I couldn't agree more. As a B2B marketer, a lot of what we do centers around the use of data to understand how our audiences are engaging, what's working and how we can use those insights to really fuel the optimization of our programs. But to your point, as a B2B marketer I don't think we've thought enough about the role of social and the role of places like Facebook to not only act as an advertising engine for us, but to really fuel the conversations we have with audiences outside of the normal stream, such as email and banner ads and things like that. Much like you said, super excited to have Peter and Jeff from Abacus Agency with us today to share their insights on this world. Peter, why don't you pick things off with a quick introduction about yourself and what you guys have been up to at Abacus?
Peter: Yeah. Thanks for having us guys, it's great to be on the show. My name's Peter Reitano. I'm the co-founder and CEO at Abacus. A little bit of background, I've been in marketing and digital marketing for say 10 years. I ran another agency called Spark before, which was a full service agency. We sold that agency two years ago and then set up Abacus. During the outro of that last agency I did a lot of brand research, market research on where I saw the agency marketing landscape was going, and started to do some consulting work with my now partner Jeff. Jeff's been in the startup marketing acquisition space for 10, 15 years. We both shared a vision for where we saw the marketing and agency world was going. We bottled it up and created Abacus from that. The main hypothesis behind Abacus, the main value prop was we saw the world going specialist rather than generalist. My last agency was full service, we wanted to be the opposite of that. We didn't think you could be really good at lots of different things, so we really wanted to focus in on one thing, which was performance Facebook advertising. We saw the world going more data driven, so a move from art and copy to code and data, and a need to tie everything back to revenue as far as possible. Those three things really drove the creation of Abacus.
Randy: Amazing. I think the things you talked about are obviously just as relevant in core marketing teams today and I see it happening more and more, of specialization with depth in certain areas, the focus on data and science. I think you're aligned exactly as you should be with how the modern marketers are really evolving, which is great. Jeff, did you want to quickly introduce yourself? A bit about your background.
Jeff: For sure. Thanks so much for having me. I've been in the startup world since 2001, which was the tail end of the dotcom bubble. I've always been leading customer acquisitions with startups, often competing against companies that were much bigger and had much bigger brands and much bigger budgets. So I've always been forced to think about marketing in a different way than just spending money and hoping for a result, because you can't outspend big competitors. In 2015, I joined a startup here in Toronto, Canada, called Borrowell. Borrowell is an online lender that lends money through Canadian banks. As the show goes on I'm sure we'll touch on Borrowell as a case study because Borrowell used content to compete against the big five banks in Canada in a way that they weren't really agile enough to do so and had really amazing results. I've lived the marriage of content and acquisition for a couple years and came to some really interesting conclusions, so really excited to be chatting with you guys today.
Randy: That's great. I know we're going to talk a lot about the data behind these programs and how to really take advantage of Facebook, but before we do that of course it all comes down to the messages, the content you're actually putting out there and what it is you're going to measure and distribute. What are you guys seeing in terms of the evolution in content, both for advertising and for broader communications with customers and clients? Where are things starting to shift in terms of the kinds of media, the ways people are using these different channels? What are the big trends there that get you guys excited?
Jeff: Peter always talks about context over content. It's a very interesting example, to use Facebook as an example of context, because traditionally marketers think about their segmentation, so the targeting, the who, and they think about their messaging, the what, the value proposition. But on Facebook, using click-through-rate as a metric for resonance or context, you can effectively tell which messages are resonating the best amongst what audiences. That's a really interesting confluence of data that a lot of people aren't taking advantage of, just using Facebook as a mass message testing platform to truly confirm product/market fit. It's really important on Facebook because the audience ... I'm sorry, the algorithm looks at that relevance and that click-through-rate and that proof of context as important to the kind of advertising or posts that they want to promote. So it just greases the algorithm and everything starts moving better when you're measuring it from a performance standpoint. That's another trend, is that a lot of people are now realizing that content funnels can be conversion funnels, and you can measure the results from content the same way you can measure the results from advertising. Often with content you have a lot more leverage than you do with advertising, meaning good content has leverage that can't be competed on a dollar to dollar basis with a bigger spender. For small companies and startups, there's so much leveraging content and if you can truly prove that context matching the what and the when, you can do really special things on the Facebook platform. That's what we did at Borrowell when we realized that a lot of the people who were taking our online loans were taking it to pay off credit card debt, it opened our eyes to a whole new world of content and really proving to the customer that we understand the pain that they're going through. Things dramatically changed in terms of the funnel when we were able to drill in on that very specific segment and make the messaging just be a home run with them.
Randy: Jeff, maybe you can dig in on one of the points you said within there. A lot of people who tune into Content Pros, as Peter said earlier, are still generalists versus specialists. We're going to a specialist route but to really dumb down this idea, you can measure, as you said, on Facebook what content is working and what content is not working. Maybe you can take us behind the scenes through a couple of simple steps, how that's done. How does someone actually reflect on the spend that they're putting towards that channel?
Jeff: Sure, I'd be happy to. It's probably one of my favorite topics in fact. I lived that content marketing moment at Borrowell when we produced a case study on someone who refinanced their credit card debt with a low interest Borrowell loan and effectively ... one of the personal financial bloggers here said that the number one thing you can do to improve your wealth overnight is to pay off high interest debt. We wrote a case study about how great this person felt about paying off the debt and having more control over their financial freedom and stuff like that. I posted that blog post from my personal Twitter account and a friend of mine read it and asked me if I knew a company that did that kind of thing because that was exactly the situation and the emotion and the weight that he was feeling right now. I said, "In fact, I do know a company. I'm working for them." He went through the process, even though he was skeptical, even though he knew me. He went all the way through, refinanced the debt the exact same way the person in our case study did and wrote us this letter saying that it felt like there was a huge weight taken off his shoulder. I really lived through that moment and said, "Ah ah, this content marketing really works." But it only works if it's authentic, it only works if you're solving problems for customers and it only works if you really know the customer you're speaking to, which we were able to figure out once we knew we were talking to credit card debt people.
Randy: That makes a lot of sense on a one off scenario. I'm not suggesting this doesn't work at scale, but back to the idea of measurement, how do measure the success at scale? It's one thing to know that we broke through with one prospect, and depending on the ECV of our deal or the average deal size, maybe one deal is what we need but for companies that are trying to figure out investing in Facebook at scale, what are some of the best indicators to know if a content's working or not? How do you actually go through those functions to test?
Jeff: Sure. One of the really nice things about Facebook is they have this thing called the Facebook Tracking Pixel. It's a little tiny piece of JavaScript code that you put on your site and it basically tracks conversions that you can setup. If your blog post at the end says, "Find out if this is right for you. Apply in two minutes, it won't affect your credit score," we can track the number of people through the Pixel that come and complete the application from the blog post, and then we can also track how many of those people converted into a loan offer. By being able to track both of those conversions and driving it back not just to the platform but to the specific ad campaign that drove it, we can then divide the number of conversions by the amount we spent to have a CPA, Cost Per Acquisition. The beautiful thing about CPA, the thing that makes really, really difficult marketing really, really easy, is if you use your Cost Per Acquisition as a common denominator across all your marketing activities, both offline and online, and you try to figure out what the CPA is by channel, then you can simply take the top 20% of the lowest CPA and double down on it as you scale strategy. Because ultimately, it's your Cost Per Acquisition that's going to determine whether you have the unit economics to support scale. We can now measure content to the same standard that we measured ads, which is super exciting because now we can put content to specific personas at specific stages of the funnel and track the waterfall down to conversions.
Randy: I love that. That is the perfect explanation of evaluating channels. I think it's amazing. This is something I realized at Uberflip, where I have both the marketing team reporting in to me, as well as our finance team reporting in to me, in terms of how much overlap there is in terms of how marketing is actually determining the investment strategy of companies. I think it's really important that marketers understands, as you put it, putting the CPA as the denominator and figuring out which channels are most effective. I want to ask you guys, and maybe more for Peter perhaps, but on your guys' website, which I was checking out at, you have this concept of the idea that marketers are moving from art and copy to code and data. The argument we just walked through, that makes a lot of sense, but what's happening to art and copy in your mind? How is that going to be managed in the future? Is that going to be in-house? Is that just going to be such table stakes? Where do you see that going?
Peter: I love this topic. The truth is we need both. Abacus itself, we're heavily focused on data-driven marketing, but we're also super focused on the creative side of making it work for Facebook itself. We're not saying it's going anywhere. I think it was David Welch, the former VP of marketing at Adobe, he was asked last year to sum up the future of marketing in six words. He said, "Creative thrills but data pays bills." All the creatives in the audience are cringing, but the truth really is we need both. Data helps us segment and map the buyer journey, but creativity is what carries your prospects and customers through it. So when you can connect to the right audience, to the right place but in a way that surprises and delights them, you can drive a much deeper impact. But it's not as simple as say taking tried and tested creative that we always made and then layering in some data. It's not old creative plus new data equals success. One of my favorite authors is a guy called Neil Postman, he's kind of a ... He's written a few books. One of them was called Technopoly, kind of questions technological progress and what we get out of it, in contrast have a look at the negative as well as the positive. Super cool author. He said in Technopoly, "Technological change isn't additive or subtractive, it's ecological." One significant change generates a total change. If you remove the caterpillars from their habitat you're not left with the same environment minus the caterpillars, you have a new environment. After TV in the United States was introduced you didn't have an old America plus TV, it changes the ecosystem and that's the way we need to think about creative going forward in a data-driven environment. The ecosystem has changed so we need to generate and build creative around that change that's holistic and takes it into account. There was a really good ad that I use in example in my class at a college I teach. A couple years ago Coca Cola ran a Superbowl ad that was made for traditional but had digital in mind as well and data capabilities in mind. The ad was this four minute long stirring rendition of America the Beautiful sung in multiple different languages, celebrating America's diversity and multiculturalism and all of that kind of stuff. Then they worked with Facebook afterwards to identify the affinity audiences for 10 second segments of the ad and spliced the ad up in random to those audiences across Facebook straight after the Superbowl ad ran on traditional. That's the kind of thing we need to be thinking about with new data plus new creative.
Randy: The one thing that gets me super excited about what you're talking about is there's the data side of understanding what's worked generally in the marketplace, as you put some content or an ad out there at broad scale. See how it's impacting the results on the backend and then say, "Okay, I'm going to double on that," or, "not on this," based on my CPAs. But the other piece to it is starting to use the data to then target people with very specific messages. I'm going to use your example there, where let's say I'm a big brand, like a Coca Cola, and I might not want to do one general video or ad that's going to hit everybody. I want to start to get to the point where I can create 10 different variants of that and depending on the demographics or the interests of different subgroups, I want to hit them with more targeted versions of it. That's what gets me really excited as a modern marketer, where I think that ability with tools like Facebook, to better understand who these audiences are and segment my message accordingly, that's when I can get hyper targeted and really optimize my performance. My question to you guys, as people who live in this space, how much of a reality is that today? Do you see platforms like Facebook continuing to open up the data sets or the ability to really customize what messages get to what audiences based on a growing set of parameters?
Peter: 100%. I'll give you two examples, personal examples. I'm from a small town in England called Derby. We've got a really bad football team called Derby County, who I support unfortunately. A few weeks ago I was hit with an ad on Facebook for a t-shirt for English expats who live in Toronto and support Derby County. There must only be one or two of us in the city that fit that.
Randy: Nice.
Peter: But also, another example on a broader scale would be ... it may not be a super popular example but Trump, during the election his team wrote masses of that on Facebook. They ran very specific messaging to very specific audiences with content they knew would appeal to them that was kind of updating in real time. It was incredibly effective as we saw in the election.
Jeff: I just want to make a quick moment about that updating in real time because I think that's something that people really don't understand about Facebook. If you handle Facebook properly and from an advanced point of view and you get into conversion campaigns, you allow the Facebook algorithm to figure out on the fly who's converting and try to deliver even more of them. It's almost like taking complete micro-segmenting, micro-targeting but putting it on autopilot, because Facebook's algorithm has access to so much data behind the actual accounts that are converting, that it can do a better job of figuring out what the audience is that you should deliver at scale than you can by working it back from personas. It's really, really interesting technology. You don't have to touch it, it automatically regenerates based on the most current cohort of conversions. It's really, really powerful.
Randy: With all this talk about advertisements and sponsorships we've going to take a quick pause here from some of our sponsors on Content Pros. Hopefully it's really tailored to everyone listening in. Within a minute or so we'll be back with more questions for Jeff and Peter here on Content Pros.
Tyler: Welcome back to Content Pros with Peter and Jeff from Abacus. Now you guys, I'm like buzzing over here with what we're talking about, of this notion of automated algorithms to granularily target my audience on autopilot. It sounds too good to be true. I want you guys to put this into real life and talk about some campaigns that you've seen out there to be super successful. I expect that those are going to be ones that had a great creative and messaging edge, but then really leveraged the data and the Facebook platform to explode, how it got targeted and who it hit. Do you guys have one or two examples you want to talk to for the audience here to put it into real terms?
Jeff: Sure, I'm going to make it super real for you guys, so you'd better be prepared. We're currently working on a campaign that's in market right now for The Body Shop. The Body Shop is trying to get a humongous number of petition signatures across all the countries they operate in to end animal cruelty and that, the combined petition, to the UN to get it stopped for good. Because it's very hard for countries on their own to stop animal cruelty when they need to be responsible for the ingredients that they import from other companies, long story. Using Facebook we are trying to get people to sign a petition, a fairly simple call to action that put their voice toward stopping animal cruelty. When we came up with the original media plan, we came up with a number of different buyer personas or segments that we thought could relate to the topic and would be interested in acting. We reverse engineered basically targeting groups to form each of those segments, and then using the Facebook Pixel to track conversion to each of the campaigns. We let the campaigns run and we were able to measure a cost per petition signature, to figure out which campaigns were doing the best. One of the ad sets that we tested these dreamt up personas against was a look-alike of the people that are converting. So because we're driving hundreds and thousands of petition signatures a day, the algorithm's getting a ton of really, really good data, and it's putting its pen to paper on what a target audience would look like that correlates high with the people that are already converting. The hypothesis being, it's almost impossible to outperform the algorithm based on the amount of behind under the hood data they have access to. To make a long story short, the look-alikes are kicking butt and we are acquiring signatures at a really efficient clip. That's a live, as we speak, look into a test that's saying, "Will cats convert better than dogs? Will PETA convert better than people who say they're dog lovers? Do cute puppies or cute cats drive cheaper signature conversion?" I think that's a really cool example of putting the creative and the data and the tracking and a good cause and really good ... We've had 7.5% click-through-rates in certain ad sets, and that's like insane, an average might be .6 or .7. You can imagine how Facebook would assign that campaign a really high relevancy score and would also be really interested in showing it to a lot of people because it's doing so well. It's only if you're able to so something really well that you're going to get a good result on Facebook. We have something at Abacus called the Facebook Ads Triangle. It basically says that you need to hack one of three points of the triangle. Either your click-through rate, your conversion rate or your relevancy score. If you come to the table with average, average, average across those three metrics, you're just going to get an average or below result. You need to come to the table with something above average in one of those points to have a winning campaign. Clearly, with this Body Shop campaign we're 10 to 20 X a typically CTR and the auction is rocking.
Randy: That's super interesting. I love that. I love that notion of the triangle. I couldn't agree more because I think we all start to appreciate that these algorithms are self-fulfilling prophecies in some ways and that they reward success and relevancy, no different from how Google Search runs on the backend. Are there any tips, or any other examples that you can share, that would help your average marketing leader out there think about, "What should I be doing to try to nail one of those?" We don't have to talk about all three but how can I make sure I am creating a program that's going to optimize for either click-through rate or for relevance? What do you guys recommend? How do you approach that challenge?
Jeff: I think it starts with tracking the right metrics. I think people, they're intimidated by data and they don't do what they need to do, which is just track a few metric obsessively. I love tracking funnel conversion rates week by week, because you can see them trend and you can figure out what requires your most attention. If conversion rate is dropping week after week and you can see that by looking at it every week, then you know that you should be focusing on UI/UX and conversion rate optimization and all the on-page stuff that might be a bigger problem than your ads at the moment. Tracking cost per something that leads to revenue. It could be a click, it could be an install, it could be a purchase, it could be a lead, but something which is a direct lever towards revenue is really important, and then tracking the funnel conversion rate so you know where to focus your attention in terms of what needs the most triage the quickest.
Randy: Really interesting stuff guys. We're starting to run down on time here and usually, this is actually unique today, Tyler and I don't always get two guests but it's been really a nice dynamic to have this full room conversation. When we wrap up we usually like to get to know our guests. Instead of asking each of you I'm going to actually play you guys off of each other. We'll start with Jeff, the challenge for Jeff is going to be to help us learn more about Peter. Jeff, I'm going to ask you this question, Peter you can get ready, you're getting the same one. What is a really interesting fact about Peter? What is Peter's go to lunch when you guys go for lunch all the time?  
Jeff: That's a really interesting question.
Randy: This is like the dating game but on Content Pros.
Jeff: Cool. Totally wasn't prepared, that's fun. Peter listens to really loud aggressive music. Kind of picture like ... What was the movie Pete with Christian Bale who's on the drums?
Peter: The Big Short.
Randy: The Big Short
Jeff: One of my favorite movies. That kind of intensity of music. That's something a lot of people don't know about him because he looks like this respectable British guy. Then his go to lunch, he's not going to like me saying this but he's always on some extreme diet, and he doesn't need it. He's a very, very handsome man. Always on some extreme version of a diet. There's always something that you're allowed zero of. So it really depends, but he doesn't eat a lot of bread. If I could follow his nutrition I'd probably have less bread, but I love bread.
Randy: Alright, Peter your turn. Revenge. Isn't the best thing going second?
Peter: Sticking on the music trend, Jeff is obsessed with a band called Phish, which in England I've barely of them. They were always kind of on the periphery of there. They were there but no one really liked them, and then I came to Canada and there were people that were religiously obsessed with this band and Jeff is one of them. I think he's been to see them maybe 70 times.
Jeff: Even more.
Peter: He collects the artwork. He's going to go in New York, apparently they've sold out 13 shows in Madison Square Garden. There's this whole underground worship of this band that Jeff's part of. In terms of food, Jeff's actually ... I don't know what you eat downtown. I know he's a very good cook. He cooked me some momofuku chicken the other day, which was excellent. Jeff's not on a diet. That's all I can say. He likes good, nice fried food and stuff.
Randy: Awesome. I've got one more question for each of you on each other, alright? We talked today about the balance, the balance of art and copy versus code and data. For each of you, who would you put more as the art and copy person versus code and data? If one you represents one or the other, who's going to own the code and data versus who's the art and copy guy between the two of you, to build not just the company you are but you guys have a strong brand as well?
Jeff: It's really hard for us to separate the two because like Pete said earlier, code and data and art and copy is something new. We have a client called Questrade. They put together these amazing videos with people asking these pertinent questions about your investments. We got them to build a vertical version for mobile because that makes it look native to mobile, and a horizontal version for desktop. They both performed really well and I think that it's really hard at Abacus to separate the two because we need really good platform specific creative to get the data result that we need. In terms of a yin and yang of Pete and I's relationship and how we keep each other in check, I think that my head is stuck in 2020 and I'm constantly thinking about how to slightly reposition us to be the most relevant to this future that I'm anticipated. Pete, with his agency background and understanding how to run a really tight ship, is constantly looking at the quarter and the next quarter. I think that's a really good yin and yang about us, is that we both have our eyes on different horizons at all time.
Peter: But if I had to choose one, I'm the artist.
Jeff: Jeff's a nerd.
Randy: Listen guys, before you turn this on Tyler and I and making sure that we know each other, we're going to wrap up this podcast. It's been fantastic to have you guys on here, sharing an area, as I said at the beginning, I don't think we talk enough about. How to distribute content, how to leverage these channels using the data and using the insights that our content gets out there, gets read and does what we want it to, which is convert people. That's a great place to probably stop. This is Content Pros, which is part of the Convince and Convert family of podcasts and other content. I'm Randy Frisch from Uberflip. I've had Tyler Lessard as always with me from Vidyard. Special thanks goes out to Jeff and Peter for joining us from Abacus. If people want to check out more about them, they can go to Dot agency is actually the end of that URL. If you're enjoying this podcast, please follow us and please give us feedback. We're at, we're at iTunes, we're at Stitcher, we're at Google Play. Wherever you're looking for podcasts, we want to make sure that we're delivering you fresh insights to keep you thinking. We thank you for joining in to Content Pros.
Show Full Transcript