Lessons

Lesson Info

Understanding Intent Signals and Visitor Engagement

how do your marketing fans and welcome back? Imagine what you're looking at here is a freeze frame of your local Bloomingdale's department store. Okay, the ceo of Bloomingdales hired you to be their advertising agency and you're running a re marketing campaign for Bloomingdale's but it's not a traditional remarketing campaign. This is a live event re marketing campaign. The chief marketing officer of Bloomingdale's tells you the advertiser that you have the opportunity to invite three current shoppers to a to an event later that evening. That's going to be promoting their newest line of luxury handbags. Us, the advertiser have to choose just three people from this freeze frame that are going to get invitations to go. So pretend that who you see in this image is is your entire population. Okay, let's say this is everybody that's in Bloomingdale's at that time. I want you to you can pause this video and I want you to mark down three of those people that you would choose right now, let's ...

say, you know, this is this is like our all visitors list, right? We haven't created any segments. It's are all visitors list, it's our entire population. We have to choose from Bringing it back to online re marketing. This would be all our visitors to our website that we have in our remarketing audiences. So go ahead, pause the video and choose three people and just mark them down on a piece of paper that you would bring. Okay, hopefully you did that. Now, obviously it would be really silly. Um and it would be pretty ineffective to just use an all visitors list. And I can't tell you how many times we've seen clients of ours bring bring bring campaigns to our agency and we're not talking about small campaigns. We were talking about campaigns that might be spending 50 100 plus $1000 a month and they're re marketing campaigns which are spending a lot of money is just simply re marketing ads. The same adds to all their previous website visitors in one big batch. Now granted that's better than a display network campaign. That's re marketing to people who have never been to the site. But as I'm going to illustrate to you right now it's so much less effective than it really could be. I'm going to create fake profiles for each of these visitors to Bloomingdale's at this time. So here we have a female she's browsed the handbag aisles every day for the past month but she never made a purchase. She's a Bloomingdale's employee. Next we have a female she's been to Bloomingdale's four times in the past week but she has never made a purchase. Next. We have another female. She's made multiple purchases but she never purchased a handbag. Next we have a female first time ever at Bloomingdale's Next. We have a male buys a handbag and perfume for his wife. On average once every three months Next we have another male frequently comes to Bloomingdale's but has never stayed more than five minutes in the store Next We have a female buys lots of handbags but only when they're on sale. She has never bought a handbag at full price yet. I actually have a female, She's brought three new full price handbags in the past year alone. Next we have a female was at Bloomingdale's yesterday, waited online with two very expensive handbags but once it was her turn to check out, she decided not to buy the actual handbags. Next we have a female first time in the store has been there for over an hour and she told the sales rep that she saw an ad in the local paper for new handbags and decided to come check it out Next we have a female her 20th time in the store, she's never bought anything but she spends most of her time checking out women's luxury clothing items, so she's pretty indecisive. So as you can see from here we started to create this this profile and understanding of each of these people in our store right now. Now take a look at this at this profile that we created for each of these individual shoppers and go back now and decide which one of these three. Remember you only have three invitations which one of these three you would want to actually invite to the party that night that evening. Once again remember the party is an event that's promoting the new line of full priced handbags from Bloomingdale's. Now we have a lot more relevant information and you can start seeing that with this information and let's bring it back to online advertising with this information where we could see which specific users are browsing, what areas of the site are they adding items to the cart, What types of items are they only buying handbags on sale or are they also buying full price handbags? People who are buying perfume, does that mean that they would be that they'd likely be interested in an event um for handbags? Maybe, maybe not. That gets back to understanding who your customers are. The males, the males who have been buying gifts for their wife. That might be an excellent um type of person to invite to that event. Now that we have much more information, much more detail about who these users are. We could make a much better guess as to which one of these people or which three of these shoppers would be the best candidates for this event that we want them to come to and obviously the goal of that event is for somebody to buy one of these new handbags. So that's the concept of segments, we're gonna start really, we're gonna start saying, okay, maybe you as the advertiser for bloom for Bloomingdale's, you decided with the CMO and the Ceo of Bloomingdales, that um only people and you look, you look at the data from the last year, only people who have bought a handbag in the past, whatever potentially buy a brand new full price handbag through a promotional event in the evening, if that's the case, then you would only invite people. The only your population, your sample would then be much smaller portion of the population. And the requirements to be in that sample would be people who have at least bought a handbag in the past year. And we could use we could create that segment and then use that segment in our re marketing campaigns in our case over here, we could then hand out those invites just the people who meet that qualification and that's the beauty, that's how we start getting um much more efficient in spending our re marketing dollars. But understanding our customers, understanding intense signals and understanding the conversion goal. We want that re marketing audience to take, we want that re marketing campaign to have. Now our our examples was talking about a specific event even in that very contained structure. Things got very complicated. There's all these different information, we know there's all these different engagement metrics, there's all these different data and we need to figure out how best to create our segments. But let's say the conversion goal or the, the goal of the re marketing campaign that or the remarketing list that we're building has nothing to do with an event promoting specifically high end handbags, let's say we're having an event that's giving away sales coupons and the CMO tells you to pick three shoppers who you should give away a sales coupon to all of a sudden. Not only do the invitees change the type of information that we want to look up and the data we want to investigate could drastically change. We may no longer be interested in handbags, we may no longer be interested in how many times that person's visited the store, how long they've spent in the store or where they came from? Remember we had that one woman who came from a local newspaper ad. So with online advertising we could segment out our sources, somebody who came from facebook, somebody who came from Lincoln, somebody who came from an email campaign, Somebody who came from the blog on our website, we might not be interested in that, We might be interested in people's buying patterns, we might be interested in our people, mobile mobile shoppers who are shopping on the go and who might be more interested in getting a email or text message coupons are people are some of these people in store shoppers versus online shoppers. If we're going to give away sales coupons, the whole picture changes the type of information we wanted to collect. Changes. The type of data analysis we do is going to change and the type of ads, re marketing list and conversion actions that we track are going to change. Let's say we're doing a new perfume product launch once again, the whole picture could change, let's say we're doing a free newsletter subscription drive. Once again, if we're, we're trying to get people, we're hosting an event Bloomingdale's is hosting an event. You have those invites, you need to get people to sign up for the free newsletter subscription. All of a sudden the whole game changes. What type of information do we want to look up? What type of re marketing segments, What characteristics of these users will give us those intent signals that will give us the best ability to guess whether or not this user should be put on this re marketing list to get the message to join our party for the free news, the free newsletter subscription drive, let's say you're trying to run an event for mobile app downloads, right? You're having a big cocktail party and you want the people you invite to ultimately leave the party after they downloaded the Bloomingdale's mobile shopping app. Once again, all that data is going to shift and change and how that re marketing list is put together is going to change and you might be promoting products similar to your last purchase. Now, if that's the case, once again, the whole dynamics change. You want to know and you know, this is definitely something that, that you want to know is is what that person's previous buying behavior was. If somebody hasn't bought anything recently, you might not invite them to the event. If somebody has a pattern of buying higher revenue, higher margin items, that might be someone who you really do want to invite to the event. So the dynamics change now to make things even more complicated, let's say the CMO and ceo of Bloomingdales tell you we're having nine parties tonight, we're having one of each of those parties, it's your job to say, okay, what is the goal of this party? What do we want the Bloomingdale's customers who we invite to do? If that's what we want them to do, How should we segment out our entire Bloomingdale's previous website, visitor or previous shopping visitor list? Which actions for each of these parties? Which characteristics of the buyer, which characteristics, what has this previous visitor done to signal to me that they're likely going to be somebody who will take the conversion action for that specific event. If there's somebody who is likely who we feel is likely to download that free newsletter subscription, they should be on that list. If we feel somebody who's likely to buy a high end handbags, they should be on that list. If we feel that this person will likely take both conversion actions. Should they be on both lists? Should there be rotating ads? Should it be switching off by day by week? How should we structure these remarketing campaigns? Because segmenting out your remarketing campaigns can get so complicated it can get so convoluted it can get so nuanced, we need to have specific go to tools to help us do this in a manageable way. That's what we're gonna cover in the next couple of sections. Talking about the different ways that you as a marketer could start developing these segments to start analyzing your data to start figuring out who your customers are, how they interact with your brand and what intense signals they're giving off by their browsing behavior. Once you have a toolbox with these kind of best practice, go to methodologies. You can then build upon them, tweak them and hand them to your specific goals to your specific sales cycles to your specific customers and start creating and developing very powerful, very detailed, very sophisticated and ultimately very profitable re marketing campaigns. Thanks again guys, and I really look forward to seeing you in the next section.