When you’re pressed to deliver results, it’s tempting to chase the new, shiny trends. But to be an effective, no-BS marketer, you need to have the conviction to say “No.”
This week we have Edward Nevraumont, an experienced marketing consultant who has worked with companies like Expedia, A Place For Mom, and General Assembly. In this episode, you’ll learn how to find and execute ideas with efficiency, calculate the true ROI of your marketing activities, and spark real success.
Listen to the episode:
- Why most marketing is bullsh*t? (and examples of it)
- Why experienced marketers often fall into the “shiny object” trap
- Why over-obsession with personalization can backfires
- Why Edward takes a stand for effective, no-bs marketing
- 4-step processes to build a solid no-BS marketing team
- How to figure out the ROI of marketing channels
- How branding can help improve your other marketing metrics
- How to identify your most profitable customers
- What Edward think marketers should learn in 5, 10, and 50 years
- General Assembly
- A Place for Mom
- David Heinemier Hansson
- DHH and Basecamp’s Guide to No-Bullshit Digital Marketing
- Warburg Pincus
- Joan Lunden
- Byron Sharp
- Mark Ritson
- 4-Steps to Creating a Rock Solid Marketing Strategy
- How Brands Grow
- How Brands Grow: Part 2
- MarketingBS Newsletter
- Ben Thompson
- Marginal Revolution
Louis: Bonjour, bonjour and welcome to another episode of EveryoneHatesMarketers.com, the No-Fluff Actionable Marketing podcast for marketers, marketing consultants, founders and tech people who are just sick of shady, aggressive marketing. I’m your host, Louis Grenier. In this episode, you’ll learn how to stop marketing with your ego, and instead, start marketing for effectiveness.
Louis: My guest today owns the domain name, MarketingBS.com, so we have a few things in common. He also sends a newsletter every Tuesday morning around this topic on MarketingBS.com. You can find it there. My guest has a rather big CV. He’s been the Chief Marketing Officer for General Assembly. He’s been the CMO of a Place for Mom and actually steered the company from the brink of collapse to generating more than 30% year over year growth, and has also been the VP of marketing at Expedia, which you might have heard of.
Louis: I’m super happy to have you Edward Nevraumont on board. Welcome.
Edward: Great to be here.
Louis: Your last name sounds really French. For once, it wasn’t difficult for me to pronounce it.
Edward: Most people have a lot of problems with it. It’s actually Belgian, not French, but you guys get along. You’re cousins.
Louis: Yeah. The Belgians are basically French. They all know this. I really love your stance on marketing and I think it takes some guts to do it. First of all, you went on on taking the risk to registering such a domain name, MarketingBS.com, and going for it. Why do you think that most of what people do in marketing is complete bullshit?
Edward: Yeah. I think it’s worth clarifying at the beginning that a lot of people see Marketing BS and they say, “Oh, you think that marketing is BS.” I didn’t register Marketing Is BS.com, just MarketingBS.com. There is a difference. I think marketing is extremely valuable. I think being able to communicate what your product is to people and having them aware of it is one of the most important things in the business. The famous saying, “Build a better mousetrap and they’ll beat a path to your door,” I think that is BS. That does not happen. You can build a great product and no one will ever find out about you unless you put some effort into marketing it.
Edward: Marketing is really, really important. I think we’ve been distracted by bad marketing for a variety of reasons that I’m happy to talk about today. That’s what I talk about in my newsletter, that’s what I talk about in my book. It’s what I built my career on, which is stop doing the BS and start doing what actually works.
Louis: So what’s the BS for you?
Edward: The BS is usually shiny objects. It’s often that we are so focused on wanting to have some sort of answer and when shiny objects are put in front of us, we chase them. Oftentimes, those shiny objects are not effective. Those shiny objects distract us from doing a bunch of simple stuff which doesn’t sound as exciting or as sexy, but is often really, really important.
Edward: I’ll give an example. I was meeting with a Fortune 500 CMO and he was talking to me, he asked me my advice on how does integrate his technology stacks between his marketing funnel and his sales funnel? He had some Salesforce CRM system and he had all his marketing out of Google Analytics and so on. He wanted to know, “How do I find out when one of the customers that are talking to my salespeople, if they visit the website and a certain page? I want to communicate that they visited that page to my sales team so they can personalize their sales story.”
Edward: I said, “Hey, that all sounds fine and good, but take a step back a second. When your marketing team gets a lead and sends it to your salespeople, how long does it take your salesperson to call that lead back?” He looked at me and he’s like, “I don’t know.” I’m like, “Well, before we talk about integrating marketing and sales tech stacks, why don’t we start by calling your customers back when they want to talk to you?” I see this thing over and over again where companies are focused on personalization and big data and AI and they aren’t calling their customers. They’re aren’t responding to customer service requests.
Edward: My take is stop doing all that BS stuff and start doing the effective stuff. By the way, doing the effective stuff can be really hard, but it’s an execution hard, not a conceptual hard.
Louis: Interesting. So execution hard would be like it’s actually simple to understand, but it takes time and it’s tough to do day to day, and so in customer service calls, on time every time, and sending emails on time every time. But what is conceptually hard to grasp could be like AI. It’s a very complex topic, right? You throw technology at a problem, it’s a vast thing, it’s difficult to understand but then you don’t even know where to start executing it. For you, BS is really about this shiny object, the personalization, as you mentioned. You also mentioned loyalty programs, you mentioned this AI stuff. I completely agree with you.
Louis: What’s baffling to me, what’s really surprising to me, is that you talked the CMO of the a Fortune 500. I assume a very smart person. Most of the marketers out there and CMOs out there are very capable people. Why do you feel they focus on this BS and these shiny objects instead of focusing on what makes sense for the business, to actually focus on, as you say, marketing effectiveness?
Edward: That’s a great question. I’m a big believer in efficient markets. There aren’t many $20 bills sitting on the ground waiting to get picked up. If doing non-BS marketing is more effective than doing BS marketing, then why hasn’t everyone moved to non-BS marketing and BS marketing just gone the way of the dodo. I think the answer is there’s an incentive problem.
Edward: When I was a CMO, I would get calls from headhunters on a regular basis looking to hire me to be a CMO at other companies. The two big questions they always asked me were not, “How effective is your marketing? What is your marketing ROI.” They wanted to know how big my marketing budget was and how big my team was. If you’re a marketer CMO and you want to get your next big job, my recommendation for you is find a way to get a really big team and a really big budget. That’s how you’re going to get your next bigger job.
Edward: One way to get a bigger team and a bigger budget is to be more effective in marketing, but it’s not the only way. I would argue it’s not even the most effective way. Once you get to a certain scale and you’re a sizable business, it’s really hard to measure effectiveness and it’s easy to BS effectiveness.
Edward: A great example of that is retargeting. Everybody loves retargeting. Vendors love retargeting. They do last touch attribution on retargeting. If someone comes to your website and then you’d cookie them. You’d follow them around the web with your ads. Some of those people click on those ads, come back to your website and buy and you give the retargeting the credit for that purchase. Well, if somebody is already on your website, some percentage of people on your website are going to come back and buy from you in the future. When you stick retargeting ads next to them, all you’ve done it provide a navigational tool and you aren’t actually getting incremental sales.
Edward: Now, some of those sales are incremental, and I’ve run tests against this, where you do retargeting where half the people see a retargeted ads and half the people you just show Red Cross ads, and you measure what the actual lift was. The actual lift ends up being small. There is a real lift, but it’s 10% of the impact you think you’re having. If all you care about it growing your marketing budget and all you care about is making it look like you’re effective, if you’re an agency say, “Hey, look at this. Your ROI on your retargeting is 1000%,” then the marketing person says, “Oh, I can spend a lot of money on that and I can go to my CFO and say, ‘It’s 1000%'”. The CFO says, “Well, why isn’t our business growing?” And they’re like, “Oh, it must be someone else’s problem, not marketing. Look at this: my marketing attribution says it’s extremely effective, so there must be some problem in sales or product or somewhere else.”
Edward: A lot of marketing is like that. It’s hard to measure and the people doing the measuring, the coaches and the referees are the same person. It’s easy to gloss over what’s actually working and what isn’t, especially when it gets complicated.
Louis: I love the fact that you’re talking about retargeting. In fact, that’s one of the lessons I’ve learned from David Heinemeier Hansson, who is the co-founder of Basecamp. One of the first episodes of this podcast more than two years ago, he talked about this bullshit of retargeting. I like the fact that you’re mentioning this. Massive businesses are being built on this technology, right? To explain it simply to people who might not know, but it is. It’s simply a way to show ads to people who’ve shown interest to you in the past. They might have visited your page or your homepage and you basically show them an ad that basically tells them to come back.
Louis: As you said, this technology works well, but it does not necessarily create new incremental customers. That’s because if you’re interested in a product, you’re going to buy from them anyway, and if you saw an ad, you might click on it, but it’s not going to make you buy necessarily from them.You mentioned this test you’ve done, which is a 10% lift, nothing compared to the biggest ROI that this company tends to talk about in terms of entirety. That’s a great example. Do you have any other examples of this technology solution or something that marketers tend to really focus on that seems to work but actually doesn’t really bring a lot of effectiveness?
Edward: Yeah. There’s lots of things that … These complicated things that I harp on, sometimes they work but they work on a small amount on the margin some of the time. It’s not that you should never do retargeting or you should never do personalization or you should never do loyalty programs, all of which I think are generally BS. It’s that you should not start there. You need to start on a bunch of simple stuff first. By the time you get to personalization, well, you better be freaking good at all the basics because personalization is not going to help you very much. If you do it really, really well, it’ll help you a little tiny bit, but it’s not going to blow away your business.
Edward: Personalization is a great example. I look at Amazon. Amazon is one of the best personalization companies in the world because they have so much data on you, there are so many categories. I live here in Seattle and I know a lot of people working at Amazon. They hire really extremely smart people to go and work on this stuff. You’re probably not going to be as good as Amazon at personalization, and yet, if you go and buy a printer on Amazon, the next thing that they recommend is that you buy another printer. All right? They’re the best in the world at this stuff and if you buy a toilet seat cover, they think you’re collecting toilet seats. If that’s the level that Amazon is at after investing in this like crazy, how good is the personalization effort that you’re going to do be?
Edward: What do you do instead? Well, why don’t you start with, instead of doing personalized content, why don’t you try to create some good content? I joke about this all the time too, how many companies do you subscribe to their email lists where you’re excited to get their email every time they send it to you?
Edward: The only people on email lists are there because the momentum of unsubscribing is too hard and they just basically keep it on the list and they don’t open it. Open rates on company based emails dropped down to about 10%, some as low as 8%. They never seem to drop below 8%. There’s always somebody who’s going to open them. There’s no value there. People hate this stuff. So what do we do? We try to optimize by changing the subject lines and changing the sentence line. If you talk to other content marketers and they’ll say, “Oh, the number one driver of open rates is the subject line.” Well, that is true at the micro-level but on the macro-level, if you have an awesome subject line, your open rate might jump from 10% to 12%, but meanwhile if the emails that you sent out to your people listening to your podcast probably have open rates of 70%. Why? Not because your subject line is awesome, but because you’re sending out good content that they want to read.
Edward: If you’re a company, before you start worrying about personalizing your emails, why don’t you just try to create a good email? If you can create a good email every week, you should send that to everybody and not a personalized email to anybody. If you can send out two good emails a week, instead of sending out two personalized emails to half of your base, just send out two great emails to your whole list twice a week. People will be very happy to read a great email. They’re not going to be happy to read a personalized crappy email.
Louis: Before we go into the basics and what are the basics for you and how to do it, because I’m really curious to know how you would do it step by step, especially a business that tends to focus on the BS and making them go back to basics and convince them to do so, before that, I just want to come back to your experience. You’ve been a chief marketing officer in a few companies, a VP of marketing at Expedia. You basically have experience with big companies and you’re not saying that from the perspective of a freelancer who only started yesterday, right? You have a lot of experience there. Why do you take such a stand? Were you doing this type of BS when you were in these type of companies and realized late that actually you had to do things differently? Were you doing effective marketing from the start and were just baffled by all of the other companies doing it differently? What’s the story behind this fight that you have?
Edward: Yeah. I went to business school at U Penn, at Wharton’s school. When I was there, I trained under some very, very smart professors doing some very, very cool stuff in marketing. I came out of business school really excited to do the stuff that they taught me. I started my career post business school at McKinsey. I was traveling the world helping out companies and I was just pumped to be like, “Hey, I can go in and help these guys do customer based valuations so much better than they’re doing right now.” But then I’d walk into these companies, and the problem wasn’t that they weren’t doing this stuff well. The problem was they weren’t doing good stuff at all. It wasn’t a matter of improving their customer lifetime models, they just didn’t have customer lifetime value models. I spent a lot of my time at McKinsey charging ridiculously high fees to these companies to get basic stuff set up.
Edward: One of my favorite stories, I was doing a churn reduction program on prepaid wireless plans. This is when you don’t even know when the person cancels, they just stop using the phone. Then if they haven’t used the phone for one day, have they churned? What if they haven’t used it for a week? At what point do you think that they’ve actually stopped using it? You’re trying to build these models to predict when the person is going to churn, and if you wait too long, they’re totally gone and you can’t get them back. If you go too early, then you’re subsidizing them for the people that wouldn’t have churned anyway.
Edward: We’d build these models out. We were working across six different countries. We’d build these models out and then we’d create programs for churn reduction and then we’d roll those programs out and then we’d test them to see what would work and what didn’t. We finally found a model that worked. It took us a long time, almost a year project. We finally found a program that consistently worked in a few different countries. We started rolling out across all their country network. It was working everywhere except for one country. We couldn’t get it to work there. We kept trying to change things and modify things. We’re like, “Is the culture of that country different? What’s going on?”
Edward: Finally, I sent my analyst to the country and said, “Hey, I want you to work your way through the whole thing, from the beginning to the end to figure out why this thing isn’t working in this country.” He spent a ton of time looking at the model, figuring out if there were any problems with the model, understanding the customer base and seeing if they were significantly different than other customer base, segmenting the customer base to see if it was working with some segments and not others. It wasn’t working with anybody. Eventually, after a couple of weeks of working through the system, he figured out that their machine that sent out the SMSs wasn’t plugged in. It wasn’t working. They weren’t sending the messages out. Their system was telling them the messages were going out, but they weren’t actually going out to any people. We were charging half a million dollars a month to this company and they hadn’t plugged their machine in. Those types of things happen all the time. The challenge isn’t, “How do we build a more sophisticated model?” The challenge is, “Let’s make sure our computers are plugged in.”
Edward: I left McKinsey to go to Expedia, to do this in-house. Again, it was the same challenges. It was so easy to get distracted by advanced segmentation stuff. Meanwhile we were sending out emails to drive attribution rather than to actually drive performance. A lot of it was how do you deal with the incentives within this big company to make sure we’re doing the right thing, rather than trying to figure out what that right thing is?
Edward: When I got to A Place for Mom, I was lucky enough to be at a company where everything was broken. The basic foundation of the company, it’s a company that helps you find senior housing, a senior housing referral service. It’s called Expedia for senior housing. If you understand marketplaces at all, they had the marketplace setups. They had more inventory, more properties, senior housing communities than anybody else, and once that happens in the marketplace, it’s really hard to beat you. But they were doing everything else wrong. They were less the Expedia for senior housing and more the Craigslist of senior housing.
Edward: There was just so much to do that we had to focus on the basics because the basics weren’t there. It made me learn how many basic things there were in order to get something to work and how hard execution was, and how when you fix one thing, something else breaks. By the time I left A Place for Mom, I started helping at other private equity backed companies in that $20M to $200 million revenue range and began to see the same parallels everywhere I went. Whether the company was a marketplace for lawyers or a software tool for urgent care clinics or an enterprise solution for security, they all had the same issues: they weren’t answering the phone. We had to go and find a way to help these companies answer the phone.
Louis: Right. It’s important to know those basics. As you said, from your experience working for different industries, different maturity levels and all of that, it seems like all of those companies struggled with the same stuff and they don’t have the same basics in place. Let’s take the scenario of when you joined the Expedia for senior homes, as you mentioned, where everything was broken, right? Let’s take this scenario and let’s talk about the steps you took to actually fix things, but you actually turned things around quite drastically in this company. What is the first step that you took? What did you put in place that wasn’t there before?
Edward: Oh man, there’s just so many things. Actually the number one problem when I was there was … There was one guy doing marketing and he was doing it terribly, so there was lots of opportunity around marketing. It’s interesting when you think about that. There’s nobody thinking about marketing, but that wasn’t our biggest problem. Our biggest problem was our sales organization, where at the time, we had 450 advisors who were turning over an 80% per year. Our reviews on Glassdoor were one star reviews consistently across the board. It was just a mess of a sales team. The compensation structure was 100% commission. Could you imagine going up to a good salesperson and saying, “Hey, I want you to leave your company where you’re making good money, come and work for our company where it’s an new industry that you’ve never been in before. Our current salespeople hate the job. They’re giving one star reviews on Glassdoor and leaving every year. Trust us, come and work for us.” Well, no good salesperson was coming to us at all.
Edward: We were left with the people … We had some good salespeople, but we tended to find … The good salespeople that we had, we got almost by pure luck. Oftentimes it was stay-at-home moms who were trying to get back into the workforce and were having a hard time getting jobs. We would hire them because we hired everybody and some of those people were very, very good, so we had some very great salespeople. We also had people who would sign up to work for us, we’d send them their computer and then they would disappear. They were basically using us to steal computers. It was just a disaster.
Edward: We had to go and totally change the compensation system. I had to change even the workflow. Those salespeople were doing everything from call screening, they were working with families, they were doing property sign-up, they were doing property management, they were doing hospital relationship management and they were doing collections. If you ask one person to do all those jobs and you’re not going to find a good person. We had to split that job into six different jobs, create a collections department, create a centralized call center, create a property management team, create a property sign-up team, create a hospital relationships team, so we can focus our salespeople on actually doing the sales job of working with families.
Edward: You’re trying to do all that while the ship is flying. If you built this business from scratch, that’s one thing, but we were an operating business and we were spitting out $40 million in revenue a year. We couldn’t afford to stop that. You try to change things as you go along and all the changes you were making were for the better.
Edward: Now we made some mistakes, but in general, we were on the right track. Even when you’re doing the right thing, change hurts you in the short term. We were living in this world where we couldn’t change too much because if we did, our company would go bankrupt and there would be no future, but if you didn’t change fast enough, well, there would also be no future. Balancing all that together, it felt a little bit like we were that kid in the Dutch dike where you’re putting your finger in to stop the water from coming out and it would spit out someplace else. It was a solid year and a half of struggling to make that work and making sure we hit payroll every two weeks.
Louis: If you had to advise people listening right now who might be struggling in their marketing, that might be in a company that’s struggling as well, not generating enough revenue, from your experience then, apart from, as you said, fixing the sales team like you’d done for this company, what are the basics, marketing-wise, that they need to set up, to have in place, that you, from your experience, don’t really see? What are those kind of things?
Edward: Yes. That’s actually what I do a lot right now. I help with a company called Warburg-Pincus, which is a private equity firm. They have hundreds of companies in their portfolio and I help out a lot of those companies get their marketing stuff set up. I like to tell them that it’s a three or four step process where the first step is stop thinking about marketing as a cost center and start thinking of it as profit center, which means you need to tie your marketing costs into revenue. You need to know that every dollar I’m spending, I’m making back more than a dollar with some sort of way to tie those two together. It’s never easy, but sometimes it’s relatively easier than others.
Edward: Your money is spent on paid search. You should get really, really good at knowing that this click I get on this ad turns into … This is my website, this many leads, that many leads turns into this many sales qualified leads which turns into this many opportunities, which turns into this many sales and the average sale value of this is X. Measuring that full funnel with some sort of attribution methodology is extremely important. You want to do the equivalent of that across all of your marketing spend.
Edward: The higher up in the funnel you go, the harder that is, but it’s never impossible. You should have what you believe. If you’re going to spend $100,000 on rebranding, you better believe in your heart of hearts, that’s going to generate more than $100,000 in value, discounted cashflow basis. If you don’t believe that, then you should not be spending that $100,000 in branding. Anything lower in the funnel than branding, you should have a much better idea of what those numbers are so you get confidence around all of your spend.
Edward: When I got to General Assembly, they were spending $100,000 a month on Snapchat filters. I said, “Does that work? Do we have any idea that this is adding any value?” They said, “Well, we don’t know. That’s why we’re only spending $100,000 a month on it.” I was like, “No, no, no, no, no, no, no. You can’t do that. If it doesn’t work, we should spend zero and if it’s does work, we should be spending $1 million or $10 million a month. We should be capping it out as much as possible,” but this, “I don’t know if it works, therefore I’m just going to throw a little bit of money at it,” is never, never, never a good idea. You need to have some way to have an estimate or a belief anyway, of what value something is providing on per dollar spent and then scale the ones that are and reduce the ones that aren’t.
Louis: Okay. Excellent, thanks so much for going through this first step. You mentioned paid search. Paid search typically is the easiest to tie back to revenue and ROI because like Google AdWords and Analytics and all that gives you a lot of good attribution naturally, so you can see how many people search for this keyword, how many people click on your landing page, how many people converted, etcetera, etcetera. It sounds like even with just an Excel spreadsheet and that kind of calculation, you can get an overall view of you put one dollar into AdWords, you get two to five dollars back or whatever, $10 back.
Louis: It sounds easy enough, right? Then when you start switching to channels that are a bit more complex, maybe content is another one, maybe event is another one, maybe brand, as you said, this type of higher purpose campaigns, how do you advise these companies to tie everything together? What’s your process for that?
Edward: Yeah. You made a good point. Start with the easiest stuff. Make your paid search is doing it. Most companies I talk to are not doing that well, or if they’re doing it at all, they’re doing it at the gross level where they have a line on their spreadsheet that’s called, “Paid search,” and they do it for the overall channel of paid search with no level of granularity. You really want to get to the point where you’re not doing it at the keyword level because you’re not going to be able to learn anything at that level of granularity but there’s somewhere between the whole channel of paid search and the level of an individual keyword and you need to be able to tie those together and figure out, “Hey, I’m spending a dollar to make $10 here, and I’m spending a dollar to make 50 cents here. I can scale up the $10 dollars and scale down the 50 cents.” You’ve got to get that right. Yes, it’s easier than the other stuff, but man, companies struggle with it. Get that going first.
Edward: Once you have that, you can work your way up the funnel. The second type of leads that you get are leads or customers where you get their email address, but they don’t buy. They don’t turn into a lead. Maybe you’re a B2B company and you have a white paper and they download the white paper. You’re creating awareness in the category but you’re not driving an immediate purchase. At General Assembly, we had a guide for how to get a job at a startup. We made no money on that. We got an email address. Now you have to figure out your email attribution programs where I collect an email address. What’s it costing me for that email address by marketing channel? What activities am I doing to that email over what period of time? What percentage of those emails over what periods of times turn into leads, which then turn into revenue down the line? It’s further back in the funnel and it’s takes you longer to measure that. Your feedback cycle is a lot slower, but you have stuff early on in the funnel that you can at least start to figure stuff out.
Edward: You want to create ways to make that attribution happen. If you have that white paper, you could gate it or not gate it. Well, the nice thing about gating it … The thing about not gating it is that you expose it to more people and you create more brand impact, but by gating it, you collect the email address which just helps you a ton on your attribution down the line. Even if you give up gating it later on, you could start measuring an impact per view of the white paper so that you can quantify that value down the line.
Edward: I’m a big believer of doing things … The example I gave of the Red Cross. You’re not going to do all your retargeting where half the ads are going to be Red Cross. You’ll throw away half your budget, but doing it once or twice to get the measure of the incremental lift of that retargeting, is really, really important so that going forward I can take away the Red Cross ads, but still assume that 90% of those clicks are invalid.
Louis: Okay. So paid search, we talked about emails. So let me just rephrase it in my own words, making sure I understand, all right? Correct me if I’m wrong, please. We collect, let’s say, 1000 email addresses from a white paper, using a CRM or some way to try those emails, let’s say we send them a campaign of 10 emails over the course of six months. After six months, we realize that 20 of those 1000 actually turned into customers, right? What we do here is we look back at the cost it took us to create this white paper potentially, and then we also look at the value that those customers brought, and then we basically look at, “Okay. From 1000 we got 20. This is the return on investment.” Is that roughly what you’re talking about?
Edward: That’s exactly right.
Edward: That takes time because you collect the email address now. They’re not ready to buy now, so it takes you some time before they turn from an email into a lead. You don’t have the fast feedback cycle that you might have on paid search. Even if you use paid search to generate leads to paid search to generate revenue, it takes longer, but that doesn’t mean that you shouldn’t do it. You need to take the early funnel metrics, which is just collecting the email address. Do they open the second email you send them? Anything earlier on in the funnel, it creates an estimate of what that’s going to be down the line.
Louis: Then we’re getting to more and more blurry territory. Let’s say that we get a lot of referrals, let’s say we get a lot of word of mouth. Typically what would happen here is you might have a lot of direct traffic coming to your site. You don’t know where the fuck they’re coming from. They land on your homepage and they convert. How do you then tie that back?
Edward: Every channel has a multiplier. It’s either synergy or cannibalizing. For retargeting, that cannibalization rate is again, it could be as high as 90%. Branded paid search is another example of a really high cannibalization rate. If you advertise on Google on a branded search, like EverybodyHatesMarketers.com and you advertise on Google on the keyword EverybodyHatesMarketers, your paid search is going to come at the top and the next thing should probably be your organic SEO. People will click on your ad, but a lot of those people would have clicked on your SEO ad anyway, so a high cannibalization rate.
Edward: Other channels have a really high synergy rate. The best example of that is television. You run a TV ad and no matter how good you are at attributing the impact of that television ad, it’s going to have impact beyond what you’re able to measure. At A Place for Mom, we did a lot of television. We had Joan Lunden as our celebrity spokesperson and we did everything we could to attribute that television spot. We did pretty direct response marketing TV spots like, “Hi, I’m Joan Lunden. You should get senior housing. Call now to get some senior housing help.” We put a phone number on the TV spots. Anytime anyone would call that phone number, we could attribute that to the TV spot. We had a dedicated phone number for every TV spot we ran. We could tie those together. It was really obvious.
Edward: That was only a small percentage of the impact that came from TV. The bigger impact came from a branded lift we saw in our brand traffic the same day the phone calls came in. We saw a super high correlation between lift and branded traffic and number of phone calls on any given day. It doesn’t stop there.
Edward: We also longterm brand lift from our television spots. This was the only brand advertising we were doing, the only top of the funnel stuff we were doing, so any brand traffic lift, we knew that’s where it was coming from. Over the course of the next 12 months, we got as much lift in brand traffic as we did on the day the TV spot ran.
Edward: Now we’re looking at three effects. We have the effect from the phone calls, the effect from brand traffic lift on the day you ran the ad and the same day you got the phone calls and you got the brand traffic lift over the next 12 months. But we’re not done there. Now we have the fact that Google has made it very, very clear that they love branded companies because there’s a brand trust when you’re a brand versus some fly by night organization. Our unbranded SEO started lifting during this same time period. Now, was that because we had an SEO team that was doing all sorts of great work? Absolutely yes. But the other impact of it was that our television was lifting our unbranded SEO.
Edward: Our click-through rate on our paid search was going up. Again, if you’re looking at paid search and you search for hotels in Seattle and one ad is from Expedia and the other ad is from Hotel-4-You.com, which one are you going to click on? Well, you’re going to click on the brand. As our brand grew, our click-through rate on paid search went up, which our quality score went up and our cost per click went down. But it doesn’t even stop there.
Edward: Recruiting employees got easier. Signing up supply got easier. Partnerships got easier. The impact of television was one, the directly measured phone calls, two, the indirectly measured brand traffic on the same day, three, the indirectly measured brand traffic over the next year, four, the lift in our other marketing channels, and five, non-marketing impact like HR impact and regulatory impact. The synergistic impact of our television was at least 3X or 4X what we could actually measure at the time.
Edward: Once you get this synergistic or cannibalistic rate, you can apply that to all of your marketing. When I do remarketing, I better be getting at least a 10 to one return that I’m measuring. If I’m not, I’m underwater. If 90% of that traffic is cannibalized, then I need to get at least a 10 to one return. On television though, if I’m getting a 5X synergy, if I’m getting a negative 80% return on television, I’m actually breaking even. A negative 80% return on measured television impact was the same as a 10X return on retargeting.
Louis: Okay. Thanks for this thorough answer. Let me deconstruct it. By the way, I think you’re the first person I’m talking to who makes the word synergy sexy again so thank you. Let’s say we spent $1 million euros or dollars … Let’s say dollars, $1 million on a TV ad and the direct return I got on the day the ad runs is only $200,000, which is minus 80% ROI because you’ve done your research over the course of the year, you know that the lift is basically five times. Whatever the direct response you get from the ad, the actual maximum impact is five times higher than that from all of the snowball effect that you had. Basically you have $200,000 multiplied by five equals $1 million, so you go back to your original investment. That’s what you’re describing?
Edward: That’s exactly right. You may or may not want to do that. If you’re spending $1 million now to make $1 million over the next 12 months, that’s probably not good enough.
Louis: But if you are making $500,000 instead of $200,000 and you know for sure that the lift is 5X, then you’re like, “Fuck, that’s a good return.” How did you actually … Then we can move onto step two of your process, but for those very high level ads stuff, how did you compare? Because you need to have a before and after, right? So you just basically say, “Before we run the ad, this is status quo. After we run the ad, this is what happens,”? And nothing else changes?
Edward: Yeah. It’s hard, right. It’s not obvious and it’s not easy. The bigger your brand, the harder it gets. We were small enough that when we started television that every time we ran our TV spots, we saw a short term brand bump that happened on the same day. I guarantee that when Coke runs a TV spot, they don’t see any lift in sales the day they run the TV spot. It’s not because the TV spots aren’t working for Coke and they’re working for us. The reason is that Coke’s base level of brand, base level of purchase levels was just so much higher that the noise overcame any signal from their TV spot.
Edward: The solution is to, at least initially, is to spend high enough in a small enough area … Sorry. Have a high enough spend and a low enough base that you can actually measure the impact. TripAdvisor did this and so did, I want to say Zillow, where to test their television spots, instead of running national TV spots, which would be far more cost efficient, they did intense TV buys in one specific city. They saturated the market to see whether or not they could see a lift in traffic in a specific city by spending intensely on TV and then not spending anywhere else. From that they can measure all those indirect effects and see what’s happening.
Edward: We did something similar at General Assembly where we did intense subway buys in different cities so we could see what lift we were getting in the cities where we were buying subway ads versus cities that were not.
Louis: Yeah. I like that. You really targeted the local markets, meaning that you’re unlikely to influence any other market around and you can really do it in some sort of A-B test in a sense. Local market A gets the ad and local market B doesn’t, the same population, roughly the same type of people. We see 50% more branding surge compared to this one after the ad is running. You basically do a real life A-B test as much as you can measure.
Edward: Yeah. Again, it’s not going to be perfect at all. No scientist is going to love what you’re putting together, but it’s enough that you can get a rough idea of what that synergistic effect is. You still want some sort of attribution stuff. At General Assembly, we still ask people when they come in the door, “How did you hear about us?” We still put a special URL on the subway ads. We have some sort of measured attribution from the subway ads, but then we see what the overall lift in the market is. We’re like, “The subway ads said that they made us $20,000 in sales, but the overall market where we run the subway ads lifted by $100,000. Our estimate of the multiplier effect of what we can measure versus what’s actually happening is a 5X on the subway ads.”
Edward: Once you have what that multiplier is in the cities that you run those tests in, you’re not going to run those tests everywhere. You’re not going to continue to run regional television and pay a premium for that. You just have to close your eyes and cross your fingers and join hands and say, “Hey, we believe the synergistic effect on how we’re measuring television is 5X. Let’s go out and spend television all over the place, measure what we can measure and believe that the impact is going to be 5X over what we’re measuring.”
Louis: Yeah. That’s really, really interesting. It makes me think a lot about a lot of stuff. I hope that’s helpful for you, if you’re listening to this episode right now. What you said, when you came back to talking to those startups that you advise, this is step one. We’re not going to have time to talk about so many steps, but the depths that you’re going into is super interesting. I want to keep going for a few minutes anyway.
Louis: Once they know roughly that marketing is now a profit center, they understand whatever money they spend they get that out, once they have that basis sorted, what do you like to do after that?
Edward: Step two is now that you have all of these marketing channels and you know what your return is on these marketing channels, you’re going to find some marketing channels where you’re spending a dollar to make 50 cents. The first step is to stop doing that. Just turn off all these channels where you’re spending a dollar to make 10 cents. Turn off your SnapChat filters, turn off, like you said at the beginning, your egotistical channels. A lot of times in marketing, dollars are being spent with your ego. You really like the local sports team so you buy the jersey sponsorship on the sports team. Turn all that shit off.
Edward: Then step two … Step B I guess of step two, is find the marketing channels where you’re spending a dollar to make $10. On those channels, just start spending more. Accelerate. Take all that money from the ones that aren’t working and throw them into the ones that are. See how much you can accelerate those channels before you start hitting diminishing returns. It’s not uncommon when I help with these Warburg companies where I find this marketing channel where they’re spending … Oftentimes it’s paid search or part of paid search, where they’re spending a dollar to make $10 and the answer becomes, “Instead of spending a dollar, let’s spend $100 and make $10,000.” We just push, push, push, push and we find ways to push more and tricks to push those channels. Finding a marketing channel that works really effectively is really, really hard. You probably have some right now that you’re not measuring. As soon as you find them, don’t let up. Just push them as hard as you can. It’s the easiest thing to do because you’re not trying to figure anything out at that point. You’re just doing what already works.
Louis: Chances are then once you squeeze as much as possible from those channels, you might reach our local maximum, you might reach a plateau, so it’s not these magical beans where you find out that one channel generates 10X the initial investment, and then you can just multiply that by one million and you’ll be a billionaire. There is always a limit. How do you handle that? I don’t want to derail you too much from the step that you’re taking, but do you see that that happens a lot or do you see that usually the plateau is really high and you have a lot of ways to improve?
Edward: There’s always diminishing returns. There has to be. As you said, no company can grow faster than the growth rate of the world for very long before eventually they become … If they couldn’t, they’d eventually take over the entire world and the growth rate of the company would become the growth rate of the world. Diminishing returns is a fact of the universe. The question is when is that diminishing returns and when is it going to happen?
Edward: Often, for whatever reason and I’m not sure exactly why this universal rule is true, but I see the square root rule all the time. Once you get to a certain size in a given marketing channel, you can always double your spend and if you do, your revenue will go up by the square root of two. Every doubling of your spend, if it’s 1.6 or whatever that number is, whatever the square root of two is, 1.4, 1.5 … I should know my math better, but that square root of two rule happens again and again and again. You increase your spend by 10%, your sales will go up by the square root of 1.1. That’s your rough diminishing returns curve.
Edward: Sometimes you’re lucky and you’re early enough in the funnel that you don’t hit those diminishing returns for a while. The diminishing returns are probably there but you can’t see them, but once you get to any reasonable scale, that’s often what happens. Yeah, that’s the answer. Diminishing returns are always going to be there, but you want to push them as far as you can. If you’re making 100X return on a channel, man, you can have lots of diminishing returns before you run into trouble. If you have a 10% return on a channel, you’re probably pretty close to the boundary of diminishing returns given what you’re doing right now.
Edward: Which gets me to step three which is do what you’re doing better. Once you find the places where there’s gold in them there hills, get better pans and better pickaxes and focus your attention on your big channels that have high ROI. I know I keep going back to paid search, but it’s probably the easiest one to talk about. Paid search in general, people do it terribly. Again, going back to the thing that we talked about in the beginning, they chase shiny objects and they don’t focus on the basics. Agencies are focused on spending your money, not on being more effective. The way agencies get more clients is that they have really good sales teams and really good customer service teams, not really good marketing teams. It’s actually a struggle for agencies that are really good at marketing. I’m sure they get frustrated because if they get better at marketing, they don’t get more revenue. They need to get better at sales and better customer service if they want to grow their business.
Edward: Paid search has just done terribly across the board. I find we go in and we take these channels and do them better. It’s improve your site speed and all of a sudden your SEO will get better because nobody’s been looking at site speed. It’s on paid search, it’s making sure that your ad copy matches the keywords people are typing in, which is again, very easy if you’re selling red umbrellas in Austin. Just make sure your ad says, “Red umbrellas in Austin.” Now again, most people use creative ads and they don’t do that. It’s fixing that first and making sure all your ads repeat the same words the person types in. Then doing that at scale is really hard. If you have 1000 products in 1000 cities with 1000 colors, all of a sudden you’re into millions of ad copy. That’s difficult to do and most companies aren’t doing it. We do that next. When you start fixing these different marketing channels, as long as they’re scalable marketing channels, you’re not wasting your time.
Edward: One of my favorite stories when I was a consultant, we were helping at a grocery store and helping them improve their operations. They were stacking bananas. We were following these stackers around through the store and watching what they were doing and when they were stacking bananas, their system to stacking bananas was to go and create the bottom layer of the pyramid and then do the second layer of the pyramid, the third layer of the pyramid and they worked their way up to the top of the pyramid and they’d stack the pile of bananas.
Edward: We timed them doing it and then we said, “Hey, let’s try this. Instead of working your way around the layer by layer of the pyramid, what if we just did one row of the pyramid and did it all the way to the top of the pyramid and then go to the next one and go from the bottom to the top, the bottom to the top, and just work your way around. Instead of doing multiple circles to stack the bananas, you do one circle around one time by stacking all the way up to the top?” We found that by doing that, it saved two minutes on the stacking of bananas. Which again, if you’re one grocery store in one place, you probably don’t care. You’re paying your staff member minimum wage. Whether he takes 10 minutes to stack the bananas or eight minutes to stack the bananas, you don’t really care. But if you’re a national grocery chain with thousands and thousands of stores, you take two minutes off the stacking of bananas across all the stores, all of a sudden you save $2 million a year.
Edward: Once you get things up to scale, small improvements can make a big, big difference. You find your biggest channels that are already effective and you focus on making your improvements there on the margin and those marginal improvements can all have significant impact on your business.
Louis: How do you prioritize such improvements? Let’s say we have five marketing channels that are ROI positive and are working very well. Underneath, as you said, are multiple things you can optimize. How do you like to prioritize, “Okay, this is the first thing we need to do. This is the second thing that you likely need to do,” etcetera, etcetera? What do you look into?
Edward: One is the scale, so you focus on your bigger channels first. Second is your ROI, so you focus on the valuable channels first. Third is the insights you already have from other places. Figuring out something new is really, really, really hard. Copying something that already works someplace else is comparably much, much easier. Where you have a toolkit of things that you’re not doing right and someone else is, focus on that first because it’s a lot easier to have impact.
Edward: Once you get to top levels of performance, what gets you there is very different and esoteric by business. Every business is different in its own way, every industry is different in its own way and it’s hard to take findings from one place to another, but the basics are the same everywhere. The idea of, “Hey, call your customers back when they want to talk to you,” is the same whether you’re in travel or you’re in car sales or you’re an education provider. Everyone wants good customer service. If there are things that you’re doing badly that other people are doing well and have had significant impact, fix that stuff first before you start thinking about, “Okay, what’s the over the top thing we can do to really, really go to the next mile?” It’s less sexy and it’s less interesting but man, it has a big impact.
Louis: Yeah, your CV speaks for itself. You’ve done an amazing job for a lot of companies, so clearly what you’re saying has a lot of value for people listening right now. You mentioned customer service as a basic. What other basics do you feel that the foundations are not there apart from that?
Edward: Figure out who your customers are that make you money and treat those customers really well. What does treating them really well mean? It means meeting their needs and being convenient for them. Coke had a thing back, I don’t know when it was, it was the ’80s or so. They said their mission was, “Every human should be within arm’s reach of a Coke at all times.” That was their mission. My wife was actually climbing a mountain in Ethiopia. Hardly anybody was there. She hadn’t seen anybody on the mountain all day long. It was her and her guide and her friend that she was hiking with. They got to the top of this mountain in Ethiopia and there was guy at the top of the mountain selling Coke. He had climbed the mountain earlier that morning and he was sitting at the top of the mountain with Cokes for anyone who climbed to the top of the mountain that day. My wife doesn’t like Coke, but you know what? She drank Coke that day because Coke had distribution.
Edward: If you can find a way to meet your customers where they are, make it easy for them, reduce transaction costs, make it more seamless. Should you have your product be self-service on the web or should you have call center people they can call, or should they be able to do it via text? The answer is yes, yes, yes. Some customers want it one way, some want it the other way, some want it the other way. You want to make it easy for anybody to buy your product anyway they want to buy it. That’s the big thing. How do you reduce friction and make things easier for people to give you their money?
Louis: What’s your number one method to identify your most profitable customers or the 20% that make up the 80%? What’s the one method you like to use?
Edward: I don’t want to go far along that route because … Byron Sharp has done a great job writing it. There’s a book called How Brands Grow, that if you haven’t read it and you’re a marketer, you need to read that book. So much of having really valuable customers is having lots of customers. You go out and acquire a bunch of customers, some of them will be really good and some of them will be bad and the bad ones will go away and the good ones will stay with you if you do a good job. The way to get more good customers is just to get more customers. The idea of trying to target and find the needle in the haystack, don’t do that. Just get a lot of haystacks and there will be a bunch of needles in there.
Edward: Now once you have those high valuable customers, you need to do stuff that makes it easy for your entire customer base to work with you and buy from you. When I say your good customers, I don’t mean your top 20%, I mean your top 80%. Your bottom 20% are probably distracting you and getting in your way, but you need to make your service awesome for the 80%, not the 20%. If you focus on that 20%, you’re often wasting your time in that your top 20% of customers love you already and they aren’t going to churn. They, almost by definition, are not going to churn. If they’re high churning customers, they’re not your top customers.
Edward: The top customers are going to love you no matter what you do and they’re going to stick around. Investing in them is almost like retargeting a little bit because if you do a bunch of stuff for your good customers and your model says, “Oh, the good customers are still good,” well, that’s true but they probably would have stayed good even if you hadn’t done anything for them. Instead, think about what your mass customers are doing, make their lives easier. It turns out that if you make the lives easier for your mass customers, lives also get easier for your best customers. They’ll increase the chances of them sticking around just as much as the mass, only you’re affecting a much larger base.
Louis: But if you only target … If you have a mass marketing strategy where you target a bunch of people who don’t necessarily share the same demographics or psychographic, how do you make sure that the message sticks for them? How do you make sure that you have marketing that is sharp for every other segment if you don’t have a lot of cash? Wouldn’t it be better to focus on those 20%?
Edward: You’ve got to be really careful, I think. Demographic and psychographic targeting is one of the chapters in my book on BS Marketing. You’ve got to be really careful. If you want to sell tennis rackets, you could try to target by demographic, but it’s far better to sell to the people who play tennis. Behavioral targeting is much more effective than demographic targeting.
Edward: And far more effective than psychographic targeting. By all means, if you’re selling tennis rackets, advertise in a tennis magazine but really wary of, “Hey, our audience skews male and aggressive, therefore we’re going to advertise in a guns and ammo because lots of men read that magazine,” well maybe, but you’re going to pay a huge premium CPM to go in and target those people and you’d be far better advertising at a broader rate.
Louis: I personally would define psychographic … To me, it encapsulates all the behavioral stuff because their beliefs make you … Okay. You’re making a distinction between psychographic and behavioral stuff, so you would make a distinction between psychographic and the job they’re hiring the product to do or the type of stuff that they would do that makes them … The one thing that makes them together as a group, as you said, a tennis racket, they play tennis. But then things get a bit trickier about, let’s say you want to buy a car. It’s not as easy to understand the behavior theme behind it. Maybe people tend to buy a car when they move to a house, for example. We still consider that to be psychographic. For you, they are two different things, the behavior and the psychographic, yeah?
Edward: Yeah, absolutely. Behavior is something you can see, that they actually did that thing, and psychographic to me is something that you’re trying to infer based on something. Man, we’re really bad at inferring stuff. Our false positive rates are … Any small group that we’re trying to target has really, really high false positives. You’ve just got to be really careful. Now you can use that to rule out people. It’s far more easy to rule out people than rule people in. It’s much easier to figure out who isn’t pregnant than it is to figure out who is. If you can get 10% better at figuring out who’s pregnant and you say, “Okay. These are the people who are most likely to be pregnant. Let’s exclude them from our non-pregnancy ads,” that’s far easier to do than, “Okay. These are people we think are pregnant. Let’s go target them with pregnancy ads.” That second one doesn’t work very well.
Louis: It’s getting better and better. I want to keep pushing just a bit on this. I know that the book you mentioned, How Brands Grow. I know that Mark Ritson and Byron Sharp is the author of How Brands Grow, right? Byron Sharp, he is?
Edward: Byron Sharp is. Mark and Byron get in fights all the time.
Louis: Yes, exactly.
Edward: I think they agree on 80% of stuff, but they fight about the 20% of stuff they disagree with.
Louis: If you had to summarize this, the two points of view, Mark seems to be more about selective right target market, where Byron is have a mass marketing strategy and you will have the right people in the mix. Is that it? Is that the difference between the two?
Edward: Yeah. I think so. I might get killed for saying this. I think Mark is a little more nuanced than Byron. I think Byron basically says … In some ways he has a hammer and it’s a really great hammer that no one else had before and now he’s hammering things with nails, and 90% of the time he’s right. His hammer is doing great stuff. He may have over-corrected a little bit. I think Mark is a little bit of the … He tries to nuance to Byron to be like … Maybe it’s a little bit, I think he goes too far in some cases, which is probably true. I think the risk becomes …
Edward: I used to teach swimming lessons to kids and when you teach a kid how to swim, you get them to over-correct because the natural tendency is to drag your arm through the water. You get them to throw their arm way, way, way out of the water. If you’re one of the best swimmers in the world, you arms barely come out of the water. But that’s not how you teach kids. You teach them to throw their arms way, way out of the water. You over-correct their natural tendencies.
Edward: I think marketers definitely need an over-correction a lot of times.
Louis: What do you think marketers should learn today that will help them in the next 10 years, 20 years, 50 years?
Edward: To do the stuff that worked 20 or 30 or 40 or 50 years ago. I think again, we’re so focused on what the next big thing is and we still haven’t learned the lessons from what worked 50 years ago.
Louis: So basically everything you mentioned before in this episode. What are the top three resources you’d recommend to listeners? It could be anything; books, podcasts, conferences, anything at all.
Edward: It’s tough. For a long time people would ask me, “What book do I need to read to become a CMO?” The answer I always gave was, “I don’t know. It doesn’t exist.” I think How Brands Grow is a great, great resource. It’s a correction to a lot of the BS that’s out there. When I was a VP at Expedia I bought a copy of the book for my entire team. I think his second book, How Brands Grow Part 2, is actually a better book. In theory, it’s more about luxury goods and international markets, but in practice, rather than it being about, “Here’s all the things that are bad,” he talks about, “Here’s what you should do instead.” I think it’s a more readable, effective book for most marketers.
Edward: I think you should be subscribing to my newsletter. I talk about this stuff every week. It’s a deep dive coverage if something that’s happening in the news that ties back into what it really means for marketers.
Edward: I read non-marketing stuff. I think a lot of the insights I get are from things that are not marketing specific that you can pull stuff from that are helpful to you. Instead of reading another newsletter about content marketing and how you need to create a top 10 list and here’s how you trick people into filling out your email and so on, read something like Stratechery where Ben Thompson does a deep dive into tech strategy. You’ll be able to pull marketing insights out of that. Read Marginal Revolution by Tyler Cohen, where he talks … I don’t even know exactly what he talks about, lots of different things every day that I pull insights from all the time. I think reading non-marketing stuff is probably going to be more helpful than reading a lot of the marketing things.
Louis: I concur. Yeah. I would say again, if you’re listening to this episode, do subscribe to Ed’s newsletter. There’s a lot of wisdom there. I’m definitely going to read How Brands Grow. I haven’t read it yet. I’m sorry, I will, I promise.
Louis: You’ve been super helpful. You shared a lot of stuff no one has ever talked about in this podcast, and I say that with entire transparency and honesty, so thank you so much. I know that people listening got a lot of value out of what you just said. Where can people connect with you and learn more from you?
Edward: Yeah. MarketingBS.com is my website. Go ahead and you can send me emails on there. Subscribe to my newsletter and just reply to a newsletter. I read everything that comes my way. I actually just purchased a new company a couple of weeks ago called IPassExam.com which is resources for passing … Practice tests for Google Analytics and Google AdWords certification, Facebook display ads certification. If you’re trying to take any of those tests, feel free to use my new website. I’m still trying to figure it out myself.
Louis: Nice. Once again, Ed, thanks so much for your time. We learned a lot from you. Thank you very much.
Edward: Oh, thanks for being here.
I’m a no-fluff marketer living in Dublin, Ireland (but yeah, I’m French).
I believe you can treat people the way you’d like to be treated and still generate results without using sleazy, aggressive, hack-y marketing. This is why I’ve started Everyone Hates Marketers – a no-fluff, actionable marketing podcast – as a side project in April 2017.
I’m also the Content Lead at Hotjar – a powerful way to analyse people’s behaviour on your website or app and understand how you can improve their experience.