google shopping campaigns, google shopping

Google Shopping Campaigns: What Retailers Now Know

It’s been nearly twelve months since Google retired its Product Listing Ads campaign format, requiring advertisers to shift to Google Shopping Campaigns, which manage both AdWords and Product Listing Ads from one place. The change was to benefit retailers by making managing ads in Google easier, giving them access to more product information and reporting data. They even added a bid simulator that lets retailers bid on ads for multiple products at once. But the new format is a challenge too, and how retailers set up their campaigns and inventory information will have a critical effect on their budget, campaign bids and overall goals.

In this session, shot at IRCE 2015 in Chicago, Brad Wolansky from Yankee Candle and Eric Best from CommerceHub  share insights into how retailers are navigating their way successfully through the first year of Google Shopping Campaigns and what was learned.


Eric Best: Thanks everyone. It’s 1:00 P.M. on Friday. You’ve just eaten lunch so Brad and I have our work cut out for us. You’re either here because you’re really passionate about ecommerce marketing, or you’re spending the weekend in Chicago and this was convenient, and we’re hoping it’s the former not the latter. Just quickly, as Don made a great introduction, but I’ll just summarize what CommerceHub is and does and then turn things over to Brad to tell you about Yankee Candle and set up the story that we’re going to talk through today. CommerceHub is a 17 year old company. As Don mentioned, they acquired Mercent, which is the company I founded about ten years ago. Together now, the platform that we provide allows both brands and retailers to expand their sellable product assortment through drop ship supplier relationships. We help both brands and retailers sell direct to consumer through channels like Amazon and eBay and also advertise through digital marketing channels like Google and Facebook and Pinterest, and really that’s what we’re here to talk about today, the Google portion of that business. We’re also connected with third party logistics platforms and carriers to help you compete on your shipping offer by delivering faster from more localized inventory. So it’s a pretty exciting business, pretty exciting time in the industry as you guys know, with all the disruption that’s taking place and I’m excited to be here. So with that, Brad, a little bit about Yankee Candle.

Brad Wolansky: Sure, thanks Eric. Glad to be here this afternoon. I’ve known this guy for a long time, probably back to 2005, I think, when I was at Orvis and we were just starting to figure out feeds and what to do, and how to do things better. Today I’m with Yankee Candle. Been there about three and a half years, bringing with me a lot of what I did in all my previous gigs along with Mercent and now CommerceHub. I’ll tell you just a little bit about Yankee Candle for those of you who need a little color around the brand. I was very excited to join Yankee Candle, doing what I’m doing, being president of direct, being president of a business we call fundraising, as well as being CMO for the enterprise. Yankee Candle is a 45 roughly year old business and it’s one of those American success stories, really. It started almost in a garage, but it started at a kitchen table, so it’s got that lore around it. Started by a guy by the name of Mike Kittredge and he made a candle for his mom for mother’s day. The neighbors saw it and wanted to buy it, so of course, what did he do? He sold it to them and made another one for his mom. He really did that for a while. Things progressed. For about the first 10, 15 years, he opened some stores. We opened a big store. Eventually we ended up with a lot of stores. In 1987, it started to be a consumer direct business as well as the local stores, sending out catalogs. In 1996 of course, as so many folks did, we started into ecommerce and got into the game. In 2013, so about a year and a half ago, we were required by Jarden. Jarden Corporation owns about 120 brands that you would recognize. I’m sure they’re all in your home. Brands from everything from Mr. Coffee to Marmot to playing cards and all kinds of things that you would recognize, and it’s been great being part of Jarden. It gives us tremendous resources and a tremendous way to move forward. Today we’ve got about 575 of our own company-owned stores. We are as omni-channel as they come. So we make the stuff all by ourselves in Massachusetts in our plant. We sell it in large retailers and we sell it in small gift retailers. We sell it in our own stores, we sell it in our outlet stores. We sell it through our catalog, we sell it through our contact center and not the least of which, we sell it on our website. So what I’m here today to do is to help Eric give a little color around what we do, what we see in terms of Google and what’s happened in the past year. In terms of where we are right now with Google, we launched AdWords in 2004, started doing Shopping in 2005. I’m one of those guys that remembers Froogle, and of course, I loved Froogle, because it started with the two words free, but those days are gone. We’ve really run the gamut of how we’ve handled search and how we’ve handled PLAs and Shopping and everything else. We did it on our own for a while. We’ve used agencies, we’ve used other agencies. We’ve brought it back inside to do it on our own and then we’re back using agencies again. If you wanted to wrap your arms around me for about two hours and a couple of beers, I could tell you why and all the wherefores of that, but let’s just say, I’m very pleased with where we are today with the relationship, with CommerceHub, in terms of how they’re helping us. It’s a critical program for us. We measure things really by not only our return on ad spend, but probably the most important thing to me, which is new customer acquisition. New customer acquisition is the life blood of what we do, not only for our ecommerce business, but for our stores. I’m responsible for driving traffic to our enterprise, whether that’s stores or whether that’s ecommerce. So the new to file rate, NTF, is what I’m looking at, relative to how much I spend. So Eric’s guys know better than anything, you can get lots of sales but if the new to file rate isn’t high, I’m not really interested. So those are some of the things that we’ll be talking about and I’ll let Eric take us further.

Eric Best: Okay. Thanks, Brad. So as you mentioned, Google, for about 12 years, Google Shopping was a free program by many different names over the course of that timeframe. As most of you probably know, we transitioned to a paid model in June of 2012. It was pretty disruptive. We saw a big drop-off, of course, in participation, as the industry was learning what the implications were of moving to direct paid advertising through Google Shopping. If you fast forward today, Google has been innovating and investing heavily in the evolution of that platform. Looking back over the course of last year, we had a pretty significant change to what is called Google Shopping campaigns and it’s key to what we’re going to talk about here in terms of how things have evolved over the course of the last 12 months. So the first thing I wanted to just touch on, CommerceHub as a digital marketing platform and service provider manages both paid search and shopping programs for our customers. Because of the conversion characteristics that we think are directly related to a richer ad unit on Google, we see higher conversion rates, therefore we see higher return on ad spend. That generally means that we’re seeing a shift in spend from search to shopping. That means traffic is going up on the shopping side of this equation and therefore revenue is growing on the shopping side of this equation. So the bottom line is, more than 50 percent of the spend that we manage on a same customer basis is on the PLA or shopping side of the equation now, and more than 50 percent of the result and the revenue comes from that side of the equation as well. All right, so there’s four questions we’re going to touch on here over the course of the next few minutes and we believe that these are sort of central to success formula on the platform. Am I achieving my revenue and return goals? Is my total product assortment visible and relevant on the channel? Am I maximizing my mobile and local opportunity? And then, if you’ve been watching the news and you’ve seen that Google has now confirmed the imminent availability of their own shopping cart, we need to talk about what that means and ensure that you guys are ready for that. So four topics I’m going to touch on to try to answer those questions today. We’re going to focus heavily on data quality and data structure, because that catalog readiness is really key to campaign performance, and that’s a recurring theme that I’ll talk about here. Again, we’ll touch on mobile and local as a discrete topic and then we’ll talk about transactional PLA’s or the Google Buy button as we wrap up. Each of these topics, by the way, is a three slide format. I’m going to talk about what’s changed on Google, the opportunity or challenge that that creates and then I’m going to talk about how to solve that. So first is data quality, how have things evolved since Google rolled out shopping campaigns roughly a year ago today? Well, Google introduced new attributes and custom labels for segmentation. Basically, the structure of the shopping campaign got more catalog oriented, it got more granular and it now gives advertisers, particularly retail advertisers of course, both brands and retailers, the ability to take more control over the campaign by segmenting it around retail relevant variables, like brand or category or custom labels might be tied to a merchandising promotion or they might be tied to a product attribute. So it’s very important today to structure that campaign around the fundamentals of your ecommerce catalog and your merchandising strategy. If data compliance, in terms of getting the right product information up to Google, is important from a listing standpoint, then product content is important from a relevancy standpoint. That means on the one hand, solving for mismatched data and ensuring that you’re normalizing attributes and variations across your catalog. On the other hand, making sure that you’re populating missing values, something as fundamental as a manufacture part number or UPC, or something more relevancy oriented, like the verbose or non-brand description and title of the product.

Brad Wolansky: This is actually how I got into it. So back in 2005 when we were first sending a feed that we had built ourselves at Orvis to Amazon, we realized that we couldn’t keep up with all of this. The feeds that we were sending just weren’t that good. I’m not that smart. I can’t keep up with what the latest changes are by all of the different marketplaces, with all of the different CSE’s and now Google. So in my mind, the big value in what these guys do initially was to clean up the data, make it good, make it what the recipient wants to hear.

Eric Best: Yeah, and we’ve got a great example here. So in the context of Yankee Candle’s website, this Juicy Watermelon makes perfect sense. It’s categorized within a candle category or subcategory. In this case, it happens to be within a semi-annual sale category within the ecommerce website. There’s a SKU number, there’s a family SKU number, so there’s information about the parent-child relationships. If you take that data out of the context of the Yankee Candle website, however, and put it on Google, you have a problem, because consumers are not going to be able to discern in a vacuum what this product is or whether or not they want to buy it without the addition of some critical information. So what CommerceHub does is, we take that initial dataset from Yankee Candle and we optimize it and we augment it, before we send it on to Google. We send it to Google, by the way, via their content APIs and data quality APIs. These are new software interfaces that sit alongside the AdWords API and are critical in terms of the speed and frequency with which you can update your information, which becomes very important as we’ll talk about in a minute. Anything else you wanted to touch on here, Brad?

Brad Wolansky: Other than the fact that about three-quarters of the time, I don’t know what he’s talking about, so if you don’t either, that’s fine. But my job isn’t really to understand the granularity of this. My job is to understand how it drives sales. So that’s why I really firmly believe in having a partner. There’s other ones– sorry, Eric– there’s other ones besides CommerceHub, but of course, I believe they do it the best. But whoever you have, you should have a partner, really from the largest of websites to the smallest. This is not something you should do on your own. It pays for itself almost immediately.

Eric Best: So moving from data quality to data structure, other changes that Google has introduced in the last 12 months is, again we talked a little bit about how product groups, the structure, the campaign, are increasingly tied to the structure and data and data quality of your catalog. So you can build now product groups around SKUs, around categories, brands, custom labels. Google is clearly rewarding granularity, not necessarily in terms of what they’re expecting from you via those APIs or through a product feed, but in terms of your ability to more granularly control budget, bidding, promotions around specific brands or categories on a seasonal basis and so on. As I mentioned, the API has taken your ability to update price and inventory at Google from a function of days, to parse now down to hours. Google, as you probably know, confirms the data that you send via API or feed by crawling your website. If those two sources of data are out of sync, then Google will penalize your quality score. So at CommerceHub, we built a platform tool set called Campaign Structure Assist, that takes all of that information out of your catalog– we’re applying this for Yankee Candle today– to automatically create the structure of your Google Shopping campaign. It’s hard to see here. It’s a little bit of an eye chart, but you can see brand, first level category, product condition, a set of multiple custom labels, item ID, product type and other inputs that we use to create again, a more granular campaign structure. What that looks like in practice is when we started working with Yankee Candle a year ago, we were managing three discrete Google Shopping campaigns, 48 ad groups and 48 product groups. Today, based on the application of this algorithmic technology, we have eight campaigns, we’re managing 137 discrete ad groups, typically built around a promotion or around a particular brand or product category and 411 discrete product groups. Again, I think to just reiterate the point you made, this becomes very difficult to manage on a manual basis, so you do want to look for a technology partner that can help you scale. Coming back to something that you opened with, this focus on new customer acquisition, versus remarketing or existing customer marketing. One of the things that we do with this newfound granularity or specificity in terms of the Google Shopping campaign, is we’re very explicit in separating brand terms from non brand terms. If you’re familiar with paid search marketing, this is not a new concept to you, but it also applies in the PLA domain as well. So here we have two examples where we’re using the same product categories, large jar candle and medium jar candle, but in the case of the brand focus campaign, we’re more concerned with existing customers– or I should say, we’re converting more existing customers. In the case of the non brand terms, we found that on PLA, like with paid search, we’re able to acquire many new customers that may not be familiar with the Yankee Candle brand, if there are any out there in the world at this point. So you can see that in practice, what happens? We take that shopping campaign that’s focused on brand terms and we are using negative keywords to eliminate 12 ounce jar candle, Apple Pumpkin jar candle, generic terms that describe the Yankee Candle catalog. On the other hand, on the non brand side of the equation, we are eliminating those brand terms to ensure that we’re focused on search sessions on Google that might be more generic search terms. And in practice, what happens is, as you’d expect, the ROAS, the return on ad spend for those brand terms, for people that know Yankee Candle and love them, is much higher while the ROAS is lower on the non brand terms. However, the inverse is true for new customer acquisition. I didn’t put the data up here but I should have. While the ROAS is lower on the non brand side, 58 percent of conversions are new to file customers on the non brand of this slide, whereas we’re seeing about 30 percent on the branded side.

Brad Wolansky: And that’s really the trick. So that is what we’ve focused on, what people miss quite often. If you’re like me, the last few days you’ve walked through the vendor hall, you’ve heard people pitch you, you’ve heard lots of shiny objects for how to increase conversion on your website, or perhaps even drive traffic to your website or your stores. When I think about all the things that I can be doing, and there’s lots of things that we all could be doing and we only have limited resources in order to do some of them, when I run things through my coffee grinder or my filter is, what kind of new to file is it going to drive? So yes, can I do it with a good ROAS? Can I do it with an affordable budget? But if I’m driving 10, 20 percent new to file, that’s like 80 percent of the people who I would’ve gotten anyhow. Why am I spending anything on them? The idea of parsing what you do through this new to file filter is extremely important, and something that some people don’t quite get, some people don’t look at, and all they look at is the return on ad spend. And so while there is a sizable difference in the return on ad spend, the more important piece is new to file.

Eric Best: The commerce and platform includes a tracking pixel or tracking tag and we just simply match the conversion to the customer file, in this case in arrears. This is not a targeting exercise but that’s how we know whether this is a new or existing customer.

Brad Wolansky: Be glad to answer some more questions on this at the end when we come back.

Eric Best: So I’m going to hit one more topic and then we’re going to talk about the results that Yankee Candle has seen and then we’re going to talk about the Google Buy button. So just on the topic of mobile and local, two big topics that are top of mind for us. One is of course, mobile bid modifiers. It’s no surprise, I think, to anyone in the room that while average conversion rate on the desktop is 3 percent or above, on table today, it’s 3 percent or below and on mobile, depending on who you cite, whose data, you’re following at somewhere between a half a percent to 1 percent. So mobile conversion is a big problem. It’s obviously becoming a larger problem. On May 5th, Google came out on their AdWords blog and confirmed that in ten countries, major countries including the United States, now more than 50 percent of impression share is coming in on a smartphone device, which means now that more than half of your revenue opportunity is coming through that device and that conversion effect is something you need to address in your bidding strategy. It’s also, of course, the motivation for Google to announce and launch this shopping cart. So we look for– oh, and there’s one other thing I’ll mention here, which is, we’ve already seen with Mobilegeddon over the course of the last month or so, that Google has adjusted their organic search algorithm to favor mobile friendly sites. If they follow suit in terms of what they’ve done in the past, those signals will become increasingly important to your paid advertising quality score. So if you haven’t built your responsive site, or you don’t have a mobile ready, mobile optimized version or your ecommerce storefront, that is the number one priority in terms of being mobile ready. Luckily, you guys came out early with a mobile friendly site. You want to just talk on the history of that?

Brad Wolansky: Yeah, mobile reminds me of the original discussion about websites. So back when I was a catalog guy and there was no ecommerce, people were walking around the office in the late ’90s saying, “What’s with this web thing? Shall we spend any money there?” We kind of cogitated about that for a few years and then everyone went there of course. Later on, it was the same thing with mobile. You’ve all been through it. Should we spend anything on mobile? Anyone going to get there? Anyone going to shop on such a small screen? I think most of us are past that. So we did do a generation 1.0 version of mobile. We’re now on our third version as of last year and clearly, more focus is being spent on it, particularly after last Christmas. So I want to say that last fall, we as well as many people, really saw an acceleration of the use of mobile, not only for browse but for buying. It really happened between September and December of last year, much, much faster than any of us had anticipated. So to some extent, we’re still catching up to that, but luckily we did have the fundamentals in place.

Eric Best: Thanks. Particularly for omni-channel retailers, there’s another use case here that I’ll talk about, and that is Google local inventory ads. You may or may not be familiar with this program. It is a derivative of the product listing ad or Google Shopping, only in this case, the ad is surfacing against physical brick and mortar store SKU level inventory and pricing information. So if you operate stores and you have the ability to extract near real time high confidence data about your inventory and pricing at the store level, you can syndicate that or we can syndicate that to Google in order to surface that ad when a shopper is walking down the street near your store or driving, hopefully not driving and searching at the same time. But they’re going to be looking for that product and as they’re searching on those branded terms or non-branded terms, Google is going to say, “This item is in stock 0.3 miles away. The price is this and by the way, the promotional offer is this.” It’s a fairly new program. I would say that the primary area right now to adoption is the level of readiness in terms of point of sale and inventory management systems. So at CommerceHub, we’re working hard to bring our customers up to speed in terms of the ability to present that near real time store specific SKU level inventory and pricing information to Google. So in summary, when we’re talking about the pure online model here, for the time being, until the cart is ready for primetime and officially out the door by Google, look at your negative mobile bid modifiers to control mobile traffic. Typically the first thing that we see, or the first thing that we saw when Google rolled out device specificity with Google Shopping is that too much budget was going toward mobile to justify, given the low conversion rate that I’ve already talked about. So ongoing attention to that as you dial in your mobile shopper experience is important. It’s one of the biggest variables to performance, now that it’s more than 50 percent of the campaign. And then if you have physical stores, think about that real time store specific, SKU specific inventory and price availability in order to capitalize on shoppers nearby through Google. So let’s pause for a moment and talk about the results then of this work that we’ve done across data quality, data structure, mobile and local over the course of the last year. Do you want to hit the punch line or shall I just summarize.

Brad Wolansky: Go ahead, keep going

Eric Best: Okay. So with the application of everything that we’ve talked about so far, Yankee Candle is up year over year through their Google Shopping campaign 76 percent. Campaign efficiency has increased by about 36 percent. So we’re generating not double, but significantly more revenue through the program over the course of the last 12 months, and at a measurably better, double digit better return on ad spend. I wish I could point at one individual tactic that we’ve talked about today that we the primary driver, but I think as you probably suspect, it’s a combination of all of these tactics applied in concert.

Brad Wolansky: Yeah, there’s no hockey stick to this. I’d love to point to one thing and say, do this and you’ll get the hockey stick of results. But it’s really paying attention to the details, it’s paying attention every single day. So there’s a guy on my team who works very closely with Eric’s team. He knows this stuff inside and out. He breathes it, along with his counterpart at CommerceHub. Every day they really look at this stuff and CommerceHub says to us, “Can we get this? We can do that.” Jesse says, “Can I do that?” Between the two, they work together and all of these details are combined to give you the result. One thing is that over time, and particularly over the past year, more knobs and levers have become available to both us and CommerceHub, working with Google and working with the other marketplaces as well, to better fine tune what we send them and therefore the results we get from it.

Eric Best: Yeah, I will say, unlike some of the newer programs, like Facebook dynamic product ads, which is now SKU level ads, surfaced in the Facebook feed, which tends to be a nascent program and therefore present a lot less competitive pressure, Google Shopping at this point, now three years in to the change to a paid model and one year post the transition to campaigns, is a very competitive space. So every one of these variables that we’re talking about, particularly on the product catalog, upstream of the campaign variables like bidding, they’re really critical to competitive success. All right, so last topic and then we’ll just kind of summarize what we’ve talked about here today. May 15th, Wall Street Journal confirmed– or Google rather confirmed and it was covered by the Wall Street Journal, that the Buy button is imminent, so we know that this is coming. What does this mean for you? Well, it raises a number of, I guess, strategic or even philosophical questions about ownership of the shop or ownership of data, branding and co-branding. But of course, what this effectively means is Google looks more like a marketplace, a la eBay or Amazon. Unlike Amazon, they’re not operating as a retailer, so this should be less contentious in that regard. Anything else you want to talk about as you consider the opportunity here?

Brad Wolansky: I’ve always operated pretty simply and that is, go where the shopper is. For Yankee Candle, that manifests itself in the fact that we have so many channels of distribution. The reason we do that is because more rather than not, the purchase of our product is really at the last minute. It’s an impulse purchase. You don’t wake up in the morning, put on your pants and say, “I’m going to go buy a Yankee Candle.” It’s more like you see it while you’re out shopping. That’s why we’re in Bed Bath and Beyond, we’re at Kohl’s, we’re at some of the largest retailers, because while people are buying other things, they also buy a Yankee Candle. So we’ve believed in being in wide distribution for that reason, to be where she is shopping. So when it comes to this topic, and I’ve been through this race before with other marketplaces and with the philosophy on whether you should be here, whether you should be there. Should you be on any of the marketplaces that you have your choice to be? My philosophy really is to be where she is and be where she wants to be, she being my consumer. So we’re looking at this very carefully, if it’s something that we believe will be brand accretive and sales accretive and of course, don’t forget, new to file accretive, which is going to be my number one measurement of this thing, then we’d want to be in.

Eric Best: Why is Google doing this? Well whether you call it one click, which is a patented Amazon term, or anything else, your objective is to reduce friction in that mobile conversion. This is stored shopper credentials and stored shopper payment, so the objective of course is to improve that mobile conversion through Google Shopping in particular. We’re seeing other platforms follow suit. CommerceHub also supports the Pinterest buyable pin, which is effectively a buy button by any other name, that they announced this week as well. So of course, all of the things that go along with marketplace selling are in effect here. Most importantly, if you’re off base in terms of your product catalog information, you’re going to end up selling an out of stock product, or even worse, you’re going to fulfill the wrong variation, the wrong size or scent of the product that you’re selling. So that focus on catalog data quality and accuracy becomes even more important because the stakes are much higher. You certainly don’t want to be accepting orders for items you can’t fulfill, or even worse, accepting returns from items that you inadvertently fulfilled incorrectly. We think that the focus on bidding really will reflect what we see, what we observe in terms of better conversion rates on mobile devices. Clearly that will have implications on those mobile bid modifiers within the shopping campaign. And we think that this is going to open more strategic options for Google in terms of post sale value add. We don’t think they’re getting into the retail business, but as we all know, they’ve been experimenting with local delivery and ship from store and other models. So to summarize, and then we’d love to open it up for questions, data quality, focus on your catalog data, quality and structure first and the campaign structure and campaign performance optimization follows. From a data structure standpoint, take the time, partner to get granular and descriptive with your product catalog. The same way that you apply search engine optimization tactics to pages on your site, every SKU in your catalog is an opportunity for differentiated content on Google Shopping and you should view it that way. Mobile and local, as things evolve, continue to bid those negative mobile bid modifiers to your conversion rate, your actual conversion rate. If you’re operating physical stores, get those point of sale and inventory management systems in line to be able to provide that near real time data to facilitate a nearby shopper engagement. The last piece is, if we do have a Google Shopping marketplace, think about the implications there in terms of significantly better mobile conversion, but also those questions that we raised together, namely data ownership, shopper ownership and your ability to brand the experience through the Google checkout. I hope this was great. I haven’t seen anyone fall asleep since the lunch ended, so hopefully this was relevant content and Brad and I are available to take follow up question.

Brad Wolansky: Be glad to answer questions. We’ve got a few minutes.

Man 1: We do have time for questions. There is a mic in the aisle here and we’ve got some people waiting to ask questions. Go ahead.

Question # 1

Brad Wolansky: Go for it.

How do you determine if it’s worth the investment to place PLA’s in a marketplace like Google?

Eric Best: Okay, well the answer is simple. If we can hit an effective return on ad spend target, this is typically informed by the profit margin of the order, based on some average order value, based on the mix of products, the line item order detail and the shopping cart, it’s a profitable transaction. It’s really as simple as that. I think strategically, you’re also elevating your brand by being present in more places on the search engine results page. Seventy-five percent or more of that search engine results page is dedicated to paid ad units today, versus organic. It’s the inverse of where we were five years ago or so, maybe ten. I’m aging myself. If the profitability and return economics makes sense, all you’re doing is generating more profitable revenue and you’re elevating your brand on that page.

Brad Wolansky: But it’s a good question, and for many years, I resisted it. In fact, I’m not buying my own brand. It’s ridiculous. I was pretty vocal about it. But as things shifted and as things moved along, I started looking and folks like Eric would just show us the dollars and cents of it. So this is a huge dollars and cents play.

Eric Best: And I would say it’s not an all or nothing discussion in any way. We’re being very intentional about those products that we surface in those PLA’s or Google Shopping ad units, based on their sales velocity, based on their gross margin and based on that new to file versus existing customer dynamic.

Brad Wolansky: Yeah, thank you.

Man 1: Next question, there.

Question # 2

Eric Best: It’s not an area of expertise for me, unfortunately. Just given our enterprise focus.

Brad Wolansky: Let me say it another way, or let me try to give you an answer without giving a specific. Any POS system that you’re looking at today, I would expect to be able to feed back to CommerceHub real time inventory information. So it is table stakes today. We at Yankee Candle have actually spent quite a bit of time looking at POS systems lately, although we’re much larger than the scenario you’re painting. But if a POS system you’re looking at does not have the ability to output real time inventory, move on.

Question # 3

Eric Best: I would say one, but I’m thinking about the direct marketing scenario. It really depends. I know companies like P90X and going back a ways, product that fit the format of a George Foreman Grill, obviously have opportunity to be wildly successful, but they’re brand driven by other channels, right? So you’re really capturing lower funnel demand that’s being generated elsewhere. In practice, in terms of some of the category and profitability and sales velocity filtering and sorting that we’re doing in terms of campaign optimization, most of our customers in practice have at least a few thousand products. And again, the answer depends on whether you’re talking parent or size, color variation information. Some of our customers have millions of product SKUS, so it’s kind of all over the board. I think a lot of it is really going to depend on, honestly, consumer demand. Maybe not a great answer to the question. Okay.

Man 1: Next question.

Question # 4

Eric Best: I don’t have expertise on the particular parameters that Google is looking at. I think if you were privy to earlier sessions today, this is kind of par for the course in terms of the level of transparency that’s available from Google in terms of how their algorithm works. This concept of responsive design, where when your site manifests on a smart device, it’s easily scrollable, it’s swipeable, it’s navigable, based on that tactile interface, as opposed to a mouse click. I think that’s kind of the basics. Structuring the site so that it’s easily navigable, ensuring things like your onsite faceted search is easily accessible right at the entry point of that shopper engagement. Those are all things that, even just a readable product detail page, I think is going to directly inform your quality score as this applies to paid ads in the future.

Brad Wolansky: It is a good question and the cynic in me says, I’ve seen new things from Google before. I’ve been through the ringer of Google shopping carts and Google wallets and Google this, Google that. The fact is, when you look at what Google is successful with, basically it’s search and now PLA’s and shopping, and everything else they’ve tried has been not so successful. Having said that, I do think that you can’t ignore what they’re doing. I never will write off what they’re doing. I think that just like where our customer is, we need to be there as retailers. We need to be where Google is, because that’s where people go. So every time they bring out something, we look at it very closely. If we can get involved in a way that limits our financial investment in terms of getting involved, we will and try it, and if it doesn’t work, we move on. A lot of things will fail, fail fast, move on. Google seems like a good bet.

Eric Best: Just last comment before the next question is, Google has been very successful in search and search based ads. Because of mobile and because of Amazon, the market is changing and that frictionless purchasing that Amazon facilitates on a smartphone is driving companies, it’s forcing them to innovate. I’ve seen that.

Brad Wolansky: Yeah. I’ve heard friction quite a bit and the idea of, how do I decrease the friction in order to hit the buy button as fast as we can is something we’ll be seeing more of over the next 12 months. Just think about how many times when you get an email from someone and you click through email while it’s going to their mobile site and you sort of play the waiting game, waiting for it to resolve. I think I’d prefer to make that buy as fast as I can in the initial search results.

Man 1: We’ll take the last question here.

Question # 5

Brad Wolansky: Good, it’s a good focus, yeah. Well, someone had to bring it up, Eric and bring up the whole attribution question. We’ve gotten through the entire session and it was the last question.

Eric Best: We’re out of time, sorry.

Brad Wolansky: Brought up the famous attribution question. I will say that this is another one that I’d welcome two to three hours and a case of beer to talk about in terms of attribution. The answer is, there’s no specific best answer. Attribution is something that moves along the spectrum of what’s good, what’s bad in a mixture of attribution sources, without taking the time here to be more specific, is my best answer.

Eric Best: At CommerceHub, we punted on the question, because our customers know better than we do how they want to measure attribution. So we just allow our clients to import their order file and tell us how they’re attributing, whether it’s weighted, first click, last click or otherwise. Their business intelligence system is probably going to have a broader purview than we do, so we let you tell us how you want to do it. Hopefully that’s helpful. We can chat further after this session.

Man 1: We are out of time. I’m going to ask you two quick questions about the new Google Buy button. What do we know about the ownership of the customer? Will the retailer be getting customer information and be able to remarket to them? Secondly, what has Google said about how they’re going to charge for this?

Eric Best: So I think I can answer the latter, which is that it will be a cost per click model and by the way, I’m just speaking based on publicly available information at this point. I haven’t heard or read the answer to the prior question. I have to believe that Google understands the dynamic with retail and brand advertisers in terms of customer ownership.   They’ve looked at Amazon’s approach and they’re going to try to be competitive in terms of what they cede to the advertiser in that regard.

Brad Wolansky: It’s an important question, Don and something that folks like myself, who have a huge investment in our brand, are very much eager to see the answer to. I, like Eric, want to believe that the answer will be a balanced approach. I’m good with a balanced approach. I don’t need to be here, I don’t need to be there. I’m good somewhere in the middle for at least our brand.

Man 1: So you’d like to be able to remarket to somebody who buys you, right?

Brad Wolansky: My ideal is to know who my customer is, and so I can talk to her again, yes.

Man 1: Great. Thanks very much, Brad Wolansky and Eric Best.

Eric Best: Thanks Everyone.