We recently had the pleasure of interviewing Eddie Lichstein, an entrepreneur, investor, and eCommerce consultant. He’s a co-founder of Cycleplicity and a managing partner at Rejoiner. We spent the call talking about PPC for eCommerce and specifically how to avoid cannibalization between different types of marketing campaigns.
In this interview, you’ll learn:
- Why eCommerce stores need to run attribution models to know which advertising channels contribute to the final sale.
- Why using too many channels can lead to cannibalization and wasted advertising dollars.
- How you can measure your campaigns to know which channels are a good idea and which ones are superfluous.
- Eddie’s personal recommendation on how to use coupons to generate sales without handing out extra money to affiliate sites.
- How a personal touch can increase conversions for online stores.
Joe: Hi Eddie, I really appreciate your joining us today to talk about advertising for eCommerce. To get us started, can you tell us a little bit about your background, what you do, and your experience with both eCommerce and PPC?
Eddie: Absolutely. Glad to be on here. Our background, or mine specifically, comes from running an eCommerce company in the automotive space for 11 years.
I’m also heavily involved with a company called Rejoiner. It’s an email abandonment recovery platform for mid-market eCommerce stores. Working with Rejoiner has given me a really good idea about how SaaS companies work and also the ability to apply what I already know from an eCommerce perspective into the world of SaaS.
That’s my background. I’ve worked in the nitty gritty of the industry all the way through highest level you can possibly get for building these types of companies. It’s been a very unique experience to be involved in every aspect of these businesses.
Joe: One of the things we’ve talked about before is cannibalization when it comes to advertising, i.e. advertising on multiple channels and paying more than once for the same customer when you could have just paid one time to get the sale.
Could you talk about that a little bit? The concept and what it is that you’ve found to be problematic for eCommerce stores.
Eddie: Sure. When it comes to cannibalization, it’s one of my points of passion. In our modern world, when you’re looking at advertising from an eCommerce perspective, or even in any digital space, there are so many different advertising mediums you can use and each one has their own dashboard with each promising X% growth, etc.
We’ve all seen it before with claims of 200% to 300% growth for traffic, etc. What tends to happen is this: while these platforms are not lying to you because there is growth and benefit in what they’re doing, the value added is not always equal for every single company and every single medium.
If you have a marketing manager on staff, he knows very well that the more campaigns you add, the more cannibalization there is, especially if you just blindly start turning one on after the other.
I’m speaking about the point where it’s very important to run individual tasks or attribution models if you’re using a product like Google Analytics in order to know what’s contributing to the sale and what’s not.
Even deeper, there are really great ways to see where your traffic is going, where it’s escaping. At the same time, take a look from a UX perspective. Do people really need every single extra touch that you’re giving them? Do they need to find you in every type of medium? Or can you take a few of them and grow those as much as you possibly can?
Joe: Can you describe a little bit more what this looks like? Let’s say a store is doing PPC with AdWords, and they’re also doing Facebook ads. On top of that, they’re doing retargeting, and then maybe they’re doing coupons, RetailMeNot, or something like that.
What does this look like from a real world example? How do stores and companies measure the attribution and get a better idea of which channel is giving them the best bang for their buck and which one they don’t need to be doing?
Eddie: You’re going to want to look at three things. Let’s say you’re using an affiliate, which in our sense is a coupon through RetailMeNot. You also have PPC ads running, and I’m not referring to product listing ads. I’m talking about PPC ads that either push a landing page or a product.
If you’re in the fast game, you’re really pushing to a pricing page or something like that. You have those two, and you want to add on retargeting as well. Let’s say you become an AdRoll customer, and you’re doing retargeting traditionally, which is on the PPC end a lot of times, then you’re also doing Facebook Ads.
The very first point of contention that would come into play is comparing traditional PPC with retargeting. Retargeting companies are looking to obviously push a product.
You as a customer have been targeted. You’ve landed on a page. We can predict that if you’ve looked at some item, you looked at some shampoo, you want to continuously look at that shampoo if you didn’t purchase it. Retargeting’s job is to make sure that anywhere you go, that shampoo will always be in front of you.
Let’s compare that to PPC. You have a shampoo product. For that product, you’re running your own PPC ad campaign that either you put together yourself or you have an agency do for you. At the same time, you have a retargeting campaign going on AdRoll
Let’s say someone searches for your site and arrives organically, which is one of the most beautiful ways to get customers. Then they decide to leave.
Afterwards, they’re going to be followed around by an ad anytime they search anything that has to do with shampoos and almost anything to do with body care. Depending on how you have your retargeting campaign set up, your ad will pop up.
The issue is that a lot of times, retargeting companies will go ahead and use PPC to display what you’ve looked at beforehand. It will either be directly on Google itself, on the search, or it may be on another page.
The cannibalization part becomes difficult when you get to the top. If you’re the customer, and now you’re searching the product, all of a sudden the retargeting campaign comes up, and also the companies PPC ad pops up.
The customer doesn’t really care which ad they click on so they go ahead and click on the retargeting one. They’re back on the page, they look at it, and they’re interested, etc.
Now, you’ve already paid out some percentage for a click-based campaigns to your retargeting company.
Then it happens again. They’re sitting around, and they come back to the same page. It’s pretty standard for people to research the exact same thing that they did initially. They re-enter that keyword, and they go back again, this time around clicking on your PPC campaign ad.
You end up selling the product. The customer checks out, and everyone’s happy. The only problem is now your cost of sale has doubled. You’ve paid out two mediums rather than one. Now you have to stop and consider, “Where does this traffic come from?”
If we take instances like this, let’s say multiply them out so we have a good data set to look at, you look through an attribution model on Google and then you can actually tell which medium people have gone into with the first click and which ones purchased with the last click.
You can see what source brought the traffic in and which one is responsible for the sale. From that you can decide, “Going on in the future, do I want to run a retargeting campaign on the shampoo and a PPC ad, or do I see one converting better than the other? At what rate can I afford to pay for both or should I be focusing on just one?”
Joe: When you’re doing this, are you looking primarily at the first click, so which channel is getting the initial visit, are you looking at the one that converts, or are you looking at them both and weighing them equally?
Eddie: I would weigh them both pretty equally. It’s just important to find out why someone has left your site or from where they’ve left and where they came in.
For example, I remember quite some time back, we worked very heavily with an affiliate company, who, from our front end, looked like they were just bringing in a ton of traffic. If you looked at the straight analytics side of it, it was huge. However, when you looked at the attribution model, you saw that the first click was really strong but the last click showed a massive drop-off in revenue, which means people were using the site to just gain knowledge.
From there, with the marketing team you have to figure out, “How much from a branding perspective are we reliant on the service? Should we be worrying about it?” It almost falls into a different bucket from a marketing scheme or mix of what you’re doing.
Originally you thought this was driving traffic directly and converting, now you’re looking at it more from a branding perspective and it’s only the first part in the funnel. You’re going to be looking at it from the first-click perspective and ask, “How well is it working to get me new customers.” After that, you can see how long it takes to convert them. Each one has the first click and the last click, two different approaches that you have to look at both together.
Joe: That makes sense. It also seems like, when it comes to PPC and retargeting, those seem to play together pretty well, so you might be looking to pay to get them to your site, and then have an ad follow people around. And then hopefully those are the ads that help you to eventually convert the sale.
It seems like those work together pretty well, but I see, potentially when you have a third or fourth channel coming in, coupons or affiliates or something like that, that’s when the cannibalization can go up.
Do you have anything to talk about along those lines? What combinations tend to work well and what should people be looking for? Specifically, let’s say somebody clicks on a Google ad, and you’re also doing retargeting so visitors continue to see your product.
Let’s say they come back and they’re about to make the purchase, but before they do they’re think, “Wait a second, let me Google and see if there’s a coupon for this.” They find a coupon and all of a sudden now you’re paying for a third piece of marketing, which at that point is going to start getting expensive.
What can you say along the lines of what combinations you have found to work well and what combinations tend to be a little bit more tricky and people should consider more closely whether or not those combinations are cannibalizing their sales?
Eddie: Sure. I don’t want to suggest that what we’ve done in the past is the formula for perfect success. We’ve gone through our trials and failures. What I can say is that there’s one more piece I would add to what you’re talking about.
You’re talking about just mediums and how to mix and match them. I’m sure a lot of people who are reading are waiting for that golden goose answer of if I choose A and B, but not C, this is going to be successful, or whatever mix it is.
One of the biggest things we’ve found out is you must be selective, and the user experience is extremely important. Let’s say you have three mediums. You have your PPC ads for shampoo, for example. You have retargeting enabled, and the product also features a coupon that you work with an affiliate, such as RetailMeNot, which has the coupon featured on their site.
What I would start to look at is the font. You can drill down by one product, depending on the analytic software that you have.
Let’s say you’re looking at, hypothetically, a specific product. You notice that most of the people are coming through a PPC campaign. However, you’re bringin people back through retargeting. Start to look at where they’re coming from on the retargeting end. What do you have enabled?
If it looks like they’re all coming from a search result page that has a PPC ad on it, it’s great idea to turn the retargeting off. You can run an experiment and see that maybe your PPC campaign, right after you turn retargeting off, will start to grow.
The next part to look at is user experience. You just brought up an example where customers are looking at a certain kind of shampoo. They’re about to check out and then say, “You know what? I think I need to go find a coupon.”
This is the point where you may lose a sale or your last click will come from RetailMeNot because that’s where your coupon is. Do yourself a favor and add that coupon to your checkout, or add a coupon page on your site so you don’t have to pay an affiliate fee to RetailMeNot.
If you can run a test and see that 80% of the people that purchase the product are using a coupon, it’s going to be in your best interest to add the coupon directly to your site.
The person can convert faster and the user experience improves. This is the approach that I take, but the hard part is finding out where the cannibalization comes from.
It’s not one data point. It’s not just a quick look at your analytics and running a model and saying, “You know what? This is working, but this is not.” You really need to dive in and make sure you understand where your customer is coming from, how their user experience is, and segment it based on different product sets. Or whatever semi vertical you have on your website from a product perspective to really understand, “Yes, we’re not cannibalizing as we were before or maybe we’re just throwing money at every single medium.” Like I said, the more channels you add, the worse it gets if you do it blindly.
Joe: It seems like the take away from this is, first of all, for people to realize that cannibalization is something that you will experience. What you’re looking for is an optimal or semi-optimal combination where maybe three particular things work well or four things for each company. You’ve got to measure it; you’ve got to look at it yourself.
Another takeaway to keep in mind is that, the more channels you do, the more likely the cannibalization is happening. It really makes sense to look at the numbers, to dig into it, and to consider the user experience and things like that.
To think through it and see, “Can we cut one of these out? Is one of these more valuable than the others or is one less valuable?” Would you say that’s generally the approach people should take for this?
Eddie: Yeah, absolutely. Look at those third-party companies that you use and where they’re marketing. If they’re heavily marketing in the same space, what’s the point?
It’s the same thing if I went to a supermarket and I’m selling a product and my product is featured on five different shelves across the supermarket. There may be people who wouldn’t walk to one side of the supermarket and pick out my product which means they’re excited to find it in a different spot, but more than likely, they would have seen it more than once if there’s five locations spread throughout the store. Eventually more spaces become unnecessary, especially if you have to pay for each additional space.
Be sure to look at the medium the third-party companies are using, how they’re marketing themselves, how they’re remarketing your product, etc.
Take for example, our product, Rejoiner. We retarget customers, in a sense, because we remarket via email. If you don’t have any kind of email recovery software that works, our tool reminds customers what they’ve left behind in their cart which means it’s a really good chance to bring back people that would be really expensive to reach in other ways.Those people would have forgotten about their cart otherwise and retargeting ads can start to get expensive for people who’ve abandoned carts.
If you’re running Rejoiner alongside retargeting on Facebook and Google, it’s going to figure out what should get credit for convincing customers to come back to your site.
Joe: It makes sense that both of those work differently, but when running together, it’s impossible to know which one brought the customer back.
Eddie: Separation. This is a very big strength of where I see the future of how we, as marketers, are going forward and where there’s going to be different strong players that really conquer specific market segments.
To give you an example, let’s look at eBay. eBay is, from the PLA perspective, Google’s biggest customer. They have an algorithm built in to decide which products to push with Google ads.
Let’s say you have product listed on eBay, you have a PLA ad running for your own website, and you’re on NewEgg which also has PLA ads. If everyone is sitting on PLA and that’s their main form of advertising, now when someone searches a product, you have four listings for different prices from different people, but they’re all coming back to you.
That actually creates confusion. If I was a first time customer and I saw that, why would I choose your site over someone else? Maybe I like a certain platform and I like the security of eBay and I don’t know your company, so I’m going to go ahead and click the eBay ad and buy through eBay.
You still get the sale, but why not focus on making your site look better so that if someone does come to your site, they’re actually going to checkout there rather than going back and saying, “I’d rather go to eBay because I trust them more.”
This is where the user experience comes in, i.e. where you don’t want all of these other channels cannibalizing the sales you can make on your own site.
Joe: That makes sense and is very interesting. Something else I’m interested in, zooming out a little bit, based on your 10+ years experience in eCommerce is this: how much has advertising and marketing changed? It seems like every year that goes by something new happens. What are you getting really good results with right now? What are your top go-to advertising and marketing channels or methods? What’s at the top of your list and what are you getting good results with?
Eddie: Absolutely. Obviously, I’m going to start with our own product. Rejoiner has been fantastic. We’ve been extremely happy with the non-cannibalization it provides for winning back customers who leave. And I’m speaking from the perspective of being a customer of Rejoiner and not just someone from the company.
It really does work, and it’s non-cannibalized traffic that would cost a lot to bring back with other methods. Aside from that, we have been doing pretty well with PLA. PLA is the product listing ads that Google’s created for Google Shopping.
They made the product more robust. It’s popping up on more search results. I would say that’s a really strong point for us. Aside from that, user experience has been our number one for not only growing revenues from the organic side but also improving the conversion side. We’ve really focused on that.
Unfortunately, I can’t give you that many more that have been overly successful. I know for some, a company like AdRoll has been completely amazing, and they’ve done such a good job with growing their business. But it’s not always as applicable to everyone, and it hasn’t worked that well for us.
Our biggest focus right now is on better categorization while also focusing on the site itself, user experience at checkout, etc. Those have been our number one areas to improve.
Joe: Interesting. I was tweeting with someone last night about shopping cart management, and I think a stat that gets thrown around a lot is somewhere around 55% of carts get abandoned. I was comparing that to an in-store experience. I don’t know what the number is, but I’m pretty sure 55% of people aren’t leaving their shopping carts in the middle of the aisle in Target or Walmart or something like that. Anyway, the wheels were turning, and I was curious, I was trying to think what the difference is between the two.
Some things you can think of is, in a store you get to pick up the product, look at it, feel it, if you’re in Best Buy or something like that, a salesperson can help you so you can have a personal touch, personal assistance.
Based on what you just mentioned about your user experience and conversion rates being important, is there anything specifically that you guys have focused on and have brought good results and you’ve realized, “Wow, we need to pay more attention to our checkout experience because that’s the last final leg and if we miss over there, we lose a sale.”
I’m curious to know what it is you guys found from a user experience to be the most helpful.
Eddie: Sure. You actually just touched on two very important points. The first point is somebody can pick up the product, look at it, toss it around, play kickball with it, whatever you want to do, granted it’s not too fragile like a laptop.
The first part that you mentioned is you need more robust descriptions. This is what you can provide. You want better pictures. You maybe need a video to explain how the product works. That leads me to the second part that you mentioned.
This makes your site like a boutique store where you’re going to have someone guide you along and help you with the process itself. That can be done in one of two ways.
There are a lot of stores that have very high end products or have a really high AOV where they’ll suggest for someone to contact you or chat will be a really big part where you can have a specialist help you.
One of my favorite examples is Crutchfield. Crutchfield.com is impeccable when it comes to their specialists.
I feel every single time I visit there for any kind of audio that I need that they can build me a house, boat, car, or whatever that will produce the symphony I want to experience, and they’ll know down to the smallest piece of how to make something work. They bank very heavily on that, on having a specialty sales team.
The other side, if you can’t build a really heavy sales team or have too many products to focus on, as you mentioned originally, someone can show you an initial set of products and you’re probably going to buy one of those.
Let’s say Best Buy has 100 different headphones available, but if you go to the aisle, your mostly going to look at what’s at your eye level. You’re going to see 5 of the 100 products with the best packaging at eye level standing out the most.
Essentially, do the exact same thing on your website. Give people predictive choice. Give people suggestions about what they need.
If they’re drilling down, let’s go back to the shampoo, if someone is searching for shampoo, and they click through a PPC ad, make sure you show them your top three shampoos. That way the conversion will be better because you know they sell, the branding is there, etc.
Don’t bring them to a page that has 30 different shampoos and each one is equally described and has a tiny picture next to it. Try to really bring a tailored experience to your site. I always suggest to my clients from a consultant point, think back to the traditional store, think back to what you got in a 1950s or a 1960s shop.
I don’t want to overly push Gary Vaynerchuk’s butcher shop mentality, but in a sense that applies. Use a little bit of that with service, but then also make it a really pretty butcher shop or whatever you’re running so the customer is comfortable and will actually trust you and check out.
Joe: That’s really interesting. I have one more questions, and then I’ll leave it open ended for you if there’s anything else.
Along the lines of the last thing you mentioned, you talked about making your shop pretty, making it look good, making people feel comfortable checking out. What have you found to be helpful or to really be a deal breaker from a trust perspective when it comes to design?
Let’s say your site doesn’t look very good, so people don’t trust you, or you don’t have trust symbols so people aren’t sure they can trust you with their credit card. Have you found anything specific along those lines that helps to boost your credibility and can improve conversion rates?
Eddie: We’re an automotive eCommerce store, so I can only speak from that perspective since our demographic archetype of customers can be very very different from an online shoe store or something of that nature.
From our experience, what we found is this: First, don’t over-clutter. You don’t need every single trust symbol to pop up. You don’t need to have buySAFE in conjunction with the resell ratings and eCom. Start to choose which ones are the best for you, really look at where they’re placed on site rather than how many that you have.
The second thing I’ve seen is to harness the people that have bought from the site. Customer reviews are excellent. Think about Amazon. It’s almost no touch on their end in terms of having a customer service rep. It’s an entire company built on a ton of reviews.
Let’s say you’re looking at a cigar cover, and there’s five of them to choose from. More than likely the ones at the very top are going to be the ones that are rated really high. Everyone knows that if a lot of people have rated them, more than likely this is a strong candidate for being a good product.
Something else we’ve seen has been extremely helpful is an About Us page. Many different companies pop up in search results and you don’t know who they are. When I walk into a store, I see the people who are standing and are selling to me or who can help me with a product or who are assisting me at checkout. There’s a human touch when you’re shopping in a store.
Why should you delete that altogether online? Many times trust starts to fall off when it looks mechanical. That behind here it’s just A, B, and C and a bunch of different systems that run, and you’ll never get to speak with someone or you’ll never know who this company is.
This depends on what you’re selling. If you’re in the high volume business, maybe it’s not the greatest of approaches, maybe it will work or not work for you, but I know that the smaller you get, the more boutique it is, the more people want to know who’s standing behind the products.
That has to do with being active on the social side. Make sure you have a Twitter account. Make sure you have a Facebook page. Get Instagram going. Share things about your customers. I’d like to mention one of the most beautiful designs period, a company called PetFlow out in New York. I think they’ve won awards as of last year. They have an excellent model and are in the business of selling high end dog food that gets delivered repetitively to your house or to your home.
However, to spread the word, they hit at the heart. Every customer has some kind of pet. We know that in America, and around the world, we absolutely love our pets.
People want to share their pets, so they enabled everyone to do that. Based upon that, more people started coming to the site, and more people started trusting the site. There was a human touch.
That’s what I feel like in this day and age we’re missing out on a lot when companies go to market. Try to focus on that. Doing an About Us page, have some pictures of who is on your staff. At the very least, let people know who’s working at the company.
Joe: It’s interesting you mention that. I think in my head, I was thinking for eCommerce stores, especially one like you have, a customer comes there and they’re just interested in buying a car part. Getting in and getting out.
It makes sense, if you go into a Best Buy, you get to meet the people and have an interaction. You go into a locally owned store, you might purchase from them over Best Buy just because you meet the owner and you like him a little bit more.
It’s really interesting that you mention an About Us page, pictures, maybe a short bio, things like that helps. In my head, I was overlooking that and thinking that would not be necessary for an eCommerce store.
Eddie: Yeah, very much so. You’d be surprised how many people start to value your brand based on the person rather than based on the good looks of your website.
Joe: Right. Here’s another application of that in a different space. The first two companies that come to mind are Crazy Egg and KISSmetrics. Neil Patel and Hiten Shah have both done a lot of personal branding in addition to building their company.
For myself, it makes me trust the companies more because I know they know what they’re talking about when it comes to internet marketing, so I trust the products they’ve built more.
That’s just one example where personal branding can give people more trust, and it also goes along with what you mentioned earlier, more of a butcher shop mentality where you meet the owner, you get to know the people behind the scenes, and you start to like the company and the people behind it. Not just because it’s some kind of robotic company that ships out the best products, but because there are real people behind it and you start to get to know them a little bit and like the team. That makes sense.
Eddie: Yeah, very much agreed. You got to think back to, why do you go to a different restaurant? What type of restaurant do you like? In America, we have a ton of options. Obviously in the cities, we have more boutique ones. In rural areas you’re going toward a more corporate approach, although there are a lot of great restaurants there as well.
When you choose where you’re going to go, granted you live in a metropolitan area, if you could meet the owner of a restaurant, or there was excellent service and it’s more boutique, you’re more likely to come back.
For instance, I’ve met Fabio Viviani and he has, aside from all his Food Network stuff, he has a restaurant here in Chicago called Siena Tavern. After meeting the guy and talking to him, he’s absolutely fantastic and has a great personality. Now, I want to support his restaurant. I’m going to go as many times as I can.
If I have an event in town or clients or something, I know I’ll be treated well and at the same time I have a personal relationship. Plus, there’s a story. That always helps.
Joe: That makes sense. I’m the same way. That also covers all the questions that I have or had thought of. I really appreciate your taking the time today to talk about eCommerce advertising, cannibalization, and how to improve conversions
Eddie: My absolute pleasure.
Joe: Thanks Eddie. I’ll talk to you next time.
Eddie: All right, sounds good.