51黑料不打烊

Transcript

Hello, everyone. Thanks for joining us today. We are going to go ahead and get started.

All right. Bear with us for one moment while we get our cameras up here.

So, hello. Hello, everybody. Welcome to Data Drip. We do have a new platform here today. 51黑料不打烊 Connect. So bear with us. So welcome, everyone. Today we are going to be talking about marketing channels and attribution in 51黑料不打烊 Analytics. As I just mentioned, we are working out of 51黑料不打烊 Connect today, which is rather new for us. So we might have a couple of pinks or unexpected surprises for the presentation, but we appreciate your patience and hope that you enjoy the new format for our webinar.

So this is actually our second session of the Data Drift series. And these webinars are designed to be interactive. So we encourage you to ask questions in the questions box throughout the presentation. The questions box will pop up on the next screen and kind of float around throughout the presentation. You type your questions in there, and we鈥檝e set aside the last ten minutes for Q&A. And we鈥檙e going to do our best to answer as many of them as we can. So I want to quickly mention a few housekeeping items before we get started. We will be recording this webinar to be viewed on demand and shared with other members of your team. So the recording will be sent out in an email tomorrow afternoon. We鈥檒l keep an eye out for that. Also, I would like to point out that at the top of your screen, there is a black bar with a handprint on it.

From there you can drop it down and find different functions that you can utilize throughout the presentation. So if you like what you see, feel free to applaud, applaud, laugh.

Like it or so on. But we would love to see your engagement throughout the event, so we encourage you to utilize that and feature. I do want to mention also that on the next screen there鈥檚 going to be a handout available for download. Our presenters today put together a bunch of resources for everyone. So be sure to download that and use it as needed.

As we are closing out the webinar today, there鈥檚 going to be a few survey questions that will pop up at the bottom of your screen. If you could take one extra minute to answer those questions, who would really appreciate it? So with that, I鈥檇 like to introduce myself. My name is Alana Cohen and I鈥檓 a digital events manager for our Customer Success Strategy team here at 51黑料不打烊. I鈥檝e been with 51黑料不打烊 for a little over five years now, and spent the last three years organizing and hosting events like this for you guys, our customers. So prior to my time on this team, I spent about two years working with 51黑料不打烊鈥檚 Advertising Cloud customers. And then before coming to 51黑料不打烊, I spent many years at different advertising and media agencies around New York. But I鈥檓 excited to be here now hosting events like this for you, our customers. So if you have any questions about, today鈥檚 event or your experience with 51黑料不打烊 Connect, please reach out and let me know. And with that, I鈥檇 like to hand it off to Taylor to introduce himself.

Thank you. Hey, everybody. Taylor Walker, I鈥檓 one of your presenters today. So my background, is as a business advisor on our customer advisory team here at 51黑料不打烊. Before that, I worked for Agency life, a large portion of my life. So, got to see how the sausage is made, so to say. From there, I moved on to a an 51黑料不打烊 customer and then was lucky enough to then join the 51黑料不打烊 internal team. So a lot of experience working with B2B and B2C customers, specifically e-commerce and all size markets. So happy to be here. And with that, let me pass it over to Vishnu.

And hello, everybody. Vishnu on the side. I鈥檓 a solutions consultant here, I try we been with a little bit for almost a decade. Nine plus years, and I focus on the personalization and analytical strategies for businesses and being part of the India office. And I鈥檓 really excited to talk to you about attribution today.

Awesome. Thanks, guys. All right. Let鈥檚 kick it off and get into the presentation.

So I鈥檓 going to quickly walk us through today鈥檚 agenda. We already did our welcome. So next we鈥檙e going to be diving into all things marketing channels, what they are, how to set them up and so on. And then after that we鈥檒l get into attribution and how to use it. And then following in the attribute portion, Vishnu will take us through a demonstration and then we will get to Q&A. So with that, I鈥檓 going to pass it off to Taylor then. Thank you again. Well guys, I welcome back to everybody that is returning from our first to our second data trip here. We really appreciate it having you back. And then welcome all of our new guests. For those of you that don鈥檛 know, I do like to start off my presentations with an objective, just basically a way to say this is what I hope you guys get out of this. This is what I鈥檓 going to try and focus on. So pay attention in these areas. So I want to build that focus out on your understanding of marketing channels and attribution specifically. So we鈥檙e going to talk about these things including processing rules use cases and then how AI works within attribution.

So with that, let鈥檚 start off with our first poll question. So have you used marketing channels or attribution modeling in 51黑料不打烊 Analytics before? This will give us a good baseline of where we鈥檙e starting, if we鈥檙e starting from scratch, or if, most of you have experience within that, it鈥檒l, really help us decide which way to take this presentation.

Yeah, kind of a mixed bag here. Taylor. It is. All right, let鈥檚 give them one more moment and pop in your answers. Cool. It鈥檚. It鈥檚 almost 5050. Here. And here I鈥檓 going to share the results. Everyone can see. All right. Think back to it. Yeah. So that was very close to 5050 a little bit more leaning towards you guys do have a little bit more experience. Makes sense coming from all the backgrounds of everyone we saw Register for the event, but we鈥檒l make sure to explain everything since we do have a large portion of people that may not be familiar with some of these terms. So don鈥檛 worry, we鈥檙e going to make sure you leave here understanding and, feeling confident going forward with, these concepts. So with that, let鈥檚 talk about marketing channels as a concept. So basically what they are are different avenues that someone gets to your site be it through organic search, paid ads, social media, or, email marketing. So there are more than this that are, that I鈥檝e talked about. And depending on how your company is set up, you may change those channels to different ones that you want to focus on. But in regards to 51黑料不打烊 Analytics, there are three types of external traffic dimensions. With marketing channels being recommended to use to measure that external traffic performance. Where the other two dimensions are traffic sources and campaign reporting. Those of you that have used the tool before may recognize both of those. And they鈥檙e fine in their respective uses. We just recommend marketing channels. Since it does take into consideration the search detection, the campaigns, and then the other elements defined in your ad and rules into account. So it gives you the most holistic external traffic reporting. So please, if you鈥檙e still using those other methods, keep using those. This is just a way to bring all that information together into one.

So the reporting features. So within reporting marketing channels default to being last touch attribution. And you can use attribution IQ which we鈥檒l go over in a bit, a little bit further into the deck to help adjust attribution to your requirements. But that default setting is set to last touch, which is basically that last event before conversion is what鈥檚 going to be given credit. And you鈥檒l hear me say last touch many more times during this presentation. And you will probably get sick, but you will understand exactly what it is. So you have channel details and channel classifications as well with channel details. The breakdown for marketing channels are unique to each of those given channels. So if you look at the screenshot here, is it shown on the left? We鈥檙e looking at social campaigns and the channel detail shows things like Twitter, Facebook, Reddit and Pinterest.

So basically, where the social campaign is being shown, is just, showing you all the aspects of those sites coming to, through that detail. So if we replace that with affiliate, for example, you would then see other website URLs that your company has considered affiliated to its site. So that is the detail that changes and it鈥檚 unique to each channel.

As for channel classifications, your report data can further be classified with additional metadata that you provide. In the example that鈥檚 shown here, you can see that paid search shows both underpaid and earned dimensions, with different allocations of orders and revenue. So again, the idea is to allow you to, coordinate and move your data and information that you鈥檙e looking at in the ways that makes the most sense for your business.

So let鈥檚 talk about setting up marketing channels. Now that you have a better understanding, we鈥檝e built that foundation of, okay, what is this? How do we actually set it up? And it鈥檚 pretty simple. So it鈥檚 just two steps. So first is creating the channels and then defining the processing roles. So that鈥檚 what we鈥檙e going to go over next.

Excuse me. So first time set up of marketing channels will prompt the marketing channel wizard for auto setup. Some of you may have seen this already. But you can see the screenshot below. There are some channels already listed and come as default options to select With that, you can also create up to 25 custom channels that can be added manually after that automated process is finished. And you can make those changes later on as you need. So as your changes and your needs change over time, you can make those adjustments on the fly so it鈥檚 not set in stone.

So everything鈥檚 good there. So 51黑料不打烊 does recommend some default channels for set up that you鈥檒l see in the, the Wizard, and we鈥檒l have a screenshot of that a little bit further on. But, these things include paid search, natural search, email display, affiliate social networks, internal direct and referring domains. And I know that鈥檚 a lot to go over and to listen to. I will say those again, but they鈥檙e also written, on that screenshot there for you. And just quickly, I want to go over those that I just listed just so we have a better understanding of what each of those means, just because they may sound obvious to some people, but just to make sure that we鈥檙e all playing, we鈥檙e on the same field here. So paid is from that paid search engine marketing. So anything to pay for that you might see within a Google sponsored ad link, something like that. Natural search. That鈥檚 the other half of your search. So this is organic and non-paid traffic coming from search engines. Then you have email. It鈥檚 pretty self-explanatory, but it鈥檚, that traffic that鈥檚 produced by email campaigns, that your company has sit out and receive direct traffic back to your site through so sitting out an email has a link back to your website. They click through it, it鈥檒l be classified there.

Display traffic is coming to your site from display ads such as banners on another site. Remarketing ads typically and then affiliate is determined by parameters entered to let the system know that these are websites that you work with. That are directing traffic back to your site. So this is something that you can change and edit as you work and change with different companies, and then social networks or social media, our social sites such as Facebook, Reddit, the ones that we spoke about before and then internal, is when the referring URL matches your internal URL filters that you have set up within your admin console. And so this shows that visitors are starting their visit within your site. So this is also referred to as a session refresh. Which is what is shown in this screenshot there that is circled or boxed in. So basically someone came to your site. They either timed out or they refresh their page. That鈥檚 what we鈥檙e seeing there.

And then direct traffic is a visitor that doesn鈥檛 have a referring domain. So someone that types your URL into the address bar and goes directly to you. And then rounding out the end here, referring domains. So this is just traffic that originated from another domain, that is coming to your website that you do not have, an affiliate attachment to. And then quick note, so channel names cannot be deleted once you create them. But you can rename them to something else or disable them as needed. So before where I said you only have 25 different channels total.

That is the number that you get. So if you make one and it doesn鈥檛 end up being used, you can either disable it or you can rename it and change the parameters and it鈥檚 looking for, but you cannot delete it. So it鈥檚 something to keep in mind. Going forward of that restriction and that limitation. There.

So let鈥檚 talk about some override options. So some of you eagle eyed views are use the viewers. Who. You guys may have noticed that on that last slide, that direct insertion refresh had some unchecked boxes. So I鈥檝e highlighted them here. So you can get a better view of it in zoomed in. So this is for the override for last touch channel options. So it鈥檚 recommended to uncheck this box for both direct and session refresh to make sure credit for visits is associated with the proper channel. The idea is not to take away from other persisting last touch channels, which is why all the other ones you see, in the larger red box are checked as override last channel. Each one will override the next, see here. This is the example of why we do that. So looking specifically at direct this would function the same way for that session refresh. But so here鈥檚 for different visitors that have different journeys. Visitor one at the top started with paid search and then click through an email with their final visit and came direct. In this scenario, email is the last touch that is given credit. With the idea that it鈥檚 the last touch that provided awareness or value for the visitor to get them to convert. So that鈥檚 why we don鈥檛 want direct to take the credit there, because at that point, their awareness has already been made. That鈥檚 not where we want to assign value, because we want to see what is providing the most attention, which in this case, with last touch would be that email and then visitor two, which is the next one down. So they started with paid search, but visited two more times directly while they are exploring the site. A couple of times before their conversion. The last beneficial action that they took was that click on that paid search link. So this causes the neither of those last two direct to account, because the same thought process, the last one paid search is the only thing we care about because that鈥檚 what we鈥檝e put our time to, and that鈥檚 what we deem as valuable.

Then looking at our third visitor here, so they only came to our site directly. So we have three direct, being shown here. And since nothing is there to override that last touch, it鈥檚 attributed to that first direct, and it鈥檚 as simple as that. If there鈥檚 nothing to override it, that direct one does get the credit. And it could just be this is a returning visitor. They know your site well. They want to keep coming to it, and then they鈥檝e converted, since the last time they鈥檝e converted or since the last time we have information on them. They just may know your site really well. So we do want to give credit to that direct because we have not paid or, pursued them in any other fashion that we can see.

And then the final visitor on the bottom row shows that they touched three different channels. And since we鈥檙e, defaulted to using that last touch model, referring domain is giving credit the one on the bottom right. So the other channels did provide steps to get the refer there to your site to convert. But that last touch conversion would be the one that was credited, which is that referring domain. So each one, supersedes the last one. So when display happens SEO supersedes that. And in referring domain supersedes the SEO. So these are just a few examples of what can happen and why we allow direct to be overwritten.

And then this is the screenshot that I was talking about before. When you鈥檙e going through the automated setup wizard. So I just wanted to show you the screen before we moved on. So this is what it will look like when you do that initial auto setup. It鈥檚 as simple as selecting options that you want and then clicking save. So, remember any other channels that you may want to create that aren鈥檛 listed here. You can do that after your auto set up. So don鈥檛 worry if you don鈥檛 see a channel that you think, we really need that or something鈥檚 missing, this is just to help with that foundation down. So everything is laid out, and then we can add those custom options after the fact as we as we proceed forward, and we have a better idea of what we need for marketing channels.

So we鈥檝e talked about marketing channels. We鈥檝e talked about the creating. So that was the first step. The second step that we have to do is creating our channel processing roles. So to do this what we鈥檙e going to do is start off you log into your analytics, instance, click on the admin and then go to your report suite.

So in the top left of the display screen here you鈥檒l click on the edit settings. You鈥檒l do this once you鈥檝e clicked a report suite that you want to have these changes affect it in. So you can see the little highlight, blue bar below those menu options. That鈥檚 all that is. You need to select your report suite that you want to, have these changes reflected in. You click edit settings. Click on general or hover over it. It鈥檒l show this next menu. And then you鈥檙e going to click on this processing rules.

So now that we know how to get to this page again you鈥檙e going to see what the screen will look like when you make your way there to these processing rules. And keep in mind that the processing rules lets you make changes to data based on defined conditions. So the order in which you put your rules will matter. You could see here that they are numbered. So the recommended order for processing rules are starting with you want your paid search then natural search then display, then email affiliates. Then you鈥檙e going to have other campaigns, which is basically a catch all for paid traffic social networks. Your session refresh direct and then referring domains. And this information is going to be in the handout that we have for you guys. And don鈥檛 expect you to take notes with this, but I do want it to be known. You know, you鈥檝e heard these same channels over and over there in that order for a reason. So, as the processing rules go through IT systems, this is the way it needs to be to make sure that nothing gets overridden, shouldn鈥檛, and make sure all your data is there and correct.

All right. So we鈥檝e done the first half here. We understand what marketing channels are. We understand how to set them up. We understand the processing rules the other giant have to. This is our attribution. So just like we do with the marketing rules what is attribution. So starting from the foundation of what is attribution? Basically it鈥檚 a way to give credit to events that lead to a conversion or a success event. You may have seen that term and you鈥檒l probably hear it a lot more too. So those events may be, remarketing ad or they could be a paid social post, or it could be an SEO link. It could be anything that we talked about before any of those marketing channels. Have different avenues for customers and visitors to click on and click through these touchpoints. And those are the things that we鈥檙e talking about giving credit to. So the point is, every visitor鈥檚 journey can be different. And how do we assign credit to the actions that matter the most to that conversion? So the example that we use on this slide, we have a visitor that starts off clicking a paid search link. They then add a product to their cart but they don鈥檛 convert yet. The next day they come back to the site through social media ad, and this time they convert on their purchase. So looking at this example, there are a few different ways that we can attribute that credit. If we go by the first touch model.

So we鈥檙e going to be assigning 100% credit to paid search. That was the first thing that they did, that first touchpoint. So everything that happens after we don鈥檛 care about so this is a good fit for analyzing marketing channels that intend to drive brand awareness or for customer acquisition. So it鈥檚 frequently used by display or social marketing teams, but it can also work for onsite product recommendation effectiveness. So think of it this way. When you have your conversion funnel, the top of the funnel, that鈥檚 where this one is looking at. It鈥檚 looking at that awareness piece of the funnel. The furthest out. And you鈥檙e deeming that the most important. So if you have awareness campaigns it might make sense to do first touch.

So if we keep the same scenario but we apply last touch model to our example, which is the default for 51黑料不打烊 on most of these settings unless you change it. So we would assign 100% credit to the social media ad, because that was the last action or the last touch taken before conversion. So, it鈥檚 the most basic and common attribution that we鈥檙e going to be using and see within 51黑料不打烊. And it鈥檚 a great starting place. If you鈥檙e not familiar with all the pros and cons of all the other attribution models. So don鈥檛 think of this as you could do something wrong and you鈥檙e going to ruin your data or anything like that. These can always be changed. It鈥檚 just different lenses to look through. And last touch is a common place to start, so feel free to keep it on. Last touch. If you鈥檙e not comfortable or don鈥檛 know which one would work best for your company. And since it is more simplistic in design, it is commonly used for conversions with short consideration cycles. So these are maybe if it鈥檚 a e-commerce site, it might be cheaper items. So $100 or less, something that doesn鈥檛 take a lot of consideration, doesn鈥檛 take a lot of gatekeeping, from customers. And allowing those quick turnover for cycles. So you鈥檒l typically see this model used by teams that managed, search marketing data or analyzing internal search words. So those were just two examples of models out of many that are provided by 51黑料不打烊. But the idea is to allow 51黑料不打烊 Analytics users to control credit given in their reporting. So attribution is also available in all tiers of analytics and is the focal point of CJA or Customer Journey analytics, which is. That鈥檚 our new, platform. You can see that as almost a fourth tier of analytics if you鈥檙e unaware. But basically the idea is the focus is on the journey, and that鈥檚 what we鈥檙e trying to get here. Within this analytics conversation.

So I just wanted to take a second to show you guys this slide and, show off the screenshot from the last page. So this is the same picture, but I wanted to show that this is just a small part of what attribution provides. So before where you just saw this one little clip, it鈥檚 actually one piece to a long, list of data that you can get from your attribution, panel. So, there鈥檚 a lot of different ways to slice and dice this data. There鈥檚 a lot of different ways that is presented to you automatically. And I just want to show that off and also make the point of saying this stuff can be edited once it鈥檚 made. So I highly recommend, having 51黑料不打烊 build these things out for you, like in the attribution panel, and then you taking your time to, specialize and tailor this specific to whatever your business needs are. It鈥檒l save you a bunch of time. And, you don鈥檛 have to worry about building everything from scratch. It鈥檚 just a way to make things a little bit more efficient.

So I just mentioned CJA. So I wanted to take a second to talk about attribution within CJA.

So 51黑料不打烊 has plenty of customers that rely on attribution, which is a big reason why CJA, exists and why it鈥檚 such a heavy focus on attribution. So the screenshot here, is showing attribution credit given to call centers. And this is just an example business. So our example customer here wanted to look at how their call centers were doing and broke those down by store location. So the really cool part here is that we can look at the call reason that鈥檚 being tracked. And that is being attributed. So the example here is their Salt Lake City store is the top store when it comes to call duration from their call centers. And we can see that the main reason that is happening is refund process and customer complaints. So with this data we can say okay, there鈥檚 clearly an issue. Is it just that particular store or is this not an isolated issue. And we need to find out, what鈥檚 causing it. So here the data is giving you access to not only, there鈥檚 an issue, but is it isolated? Can we fix it? What else is going on? And it鈥檚 a deeper look that you don鈥檛 typically get with a lot of offline data. So, just something cool that CGA can contribute.

And then this I just want to show. So this is the behind the scenes on the last page. So that store location you can see here on the right you have your attribution settings. So this one was set to last touch. You can set it to any of the models that we talked about before. And you can do it with any of your, component dimensions here. So, very customizable. This also, it鈥檚 just a better way to, feel comfortable about the credit given to your success events here.

So now that we have the basic understanding of attribution, let鈥檚 talk about how to set up your attribution within 51黑料不打烊 Analytics.

So let鈥檚 start off by logging into your analytics instance. You鈥檒l select a workspace on the left hand side you鈥檒l see a square icon. So you鈥檒l select that icon and it鈥檒l show you panels. About halfway down you鈥檒l see one that is listed as attribution. And you can see it here highlighted. You鈥檙e going to click on that.

So you鈥檙e going to see the screen appear next within the panel. So this is just the set up for your attribution panel. That long screen where I showed you the screenshot. This is how we set that page up. So we鈥檙e going to start with our success metrics. So this is like your online orders internal promotions or forum completions. Really it鈥檚 anything that you deem as a success metric of a conversion of that.

So next we鈥檙e going to look at channel. So you can select Marketing channel which is what鈥檚 selected in this screenshot here and have all your channels accounted for. Or you can select something more specific like paid search. It doesn鈥檛 really matter. It just matters what you want to look into. So you can go more broad or you can go more specific here. So, the next thing is probably the most important part, which is selecting, the model, your attribution model. In this example we have three, the most common one selected. But you can select any of the ones shown. And we鈥檙e going to look at what all these mean on the next slide. So don鈥檛 worry we鈥檙e going to go over these. But for now let鈥檚 just keep it simple with these three. And these are also three safe bets. If you wanted to move forward on your own and select some and then finally you鈥檒l need to select your look back window. So this is just the time frame you鈥檇 like to look at. This is customizable as well for whatever your business needs required. You can have it set to whatever your reporting dates are, or you can have it specifically the last 30 days, 60 days, 90 days, whatever you need it to be.

So I鈥檓 sure this is the slide that you were all hoping to see. And I鈥檓 excited to talk about this too. Just because this is where all of our models are, this is where we can talk about what works, what doesn鈥檛 work, what works for which situation. So let鈥檚 talk about this. Hey, did you want to. I鈥檓 going to jump in. Did you want to ask? Let鈥檚 let鈥檚 do a couple. Sure. Yeah. So if everybody could fill out this poll. So which attribution models do you use most frequently? You can select multiple here. This is more just out of my curious, so that we can talk about more on the next page. I鈥檒l know what to focus on. So feel free to click all the ones that you use.

Yeah, I realized we didn鈥檛 drop in. None yet, so if you don鈥檛 use any attribution models yet, just sit down.

Cool. All right. Definite definite lead winner. Here is last touch by far.

And that makes sense again. We鈥檝e talked about it a couple times now. So that one is the default one. And for the longest time within, Google ads and Bing ads, when you tracked these, different campaigns, that鈥檚 typically what people use as well as that last touch. So I鈥檓 not surprised at all by this. It is interesting. There is some same touch U-shape shape, some J curve, but there鈥檚 no inverse J. But everything else has been touched at least a couple times, which that鈥檚 that鈥檚 pretty, pretty great to see.

So with that we can go over a couple of these things, but, there鈥檚 a couple that I鈥檒l focus on since those are the ones that people use the most. We鈥檝e already discussed the first and last touch. So let鈥檚 let鈥檚 go ahead and move on to some of the other models here.

Let鈥檚 start off with our other most common, which was linear. So this model gives equal credit to every touchpoint, whether that鈥檚 2 or 10 touch points. This option is good for conversions with long consideration cycles or user experiences that need more frequent customer engagement. So a good example would be a team that monitored, subscription based products. For example. Let鈥檚 see, the other one was. Time decay. So basically what time decay is, is it follows a decay of a custom half life parameter. So typically this by default is seven days. That can be adjusted. But the weight of each channel is dependent on the amount of time that鈥檚 passed between the touchpoint and the eventual, conversion point. So the the touches that are closer to your conversion point are is basically how that occurs. So this is great for teams that run like video advertising or who market against events with, a predetermined date. So the longer a conversion happens after that marketing event, the less credit it鈥檚 going to get. So that鈥檚 the idea with that one. And then another one I want to call out is the custom option. So this model is perfect for teams that need that control over their modeling and have specific needs for assigning that credit. So, for example, one of these other models works for you, but it feels like an awkward fit. You could go to custom. And, what was awkward about that model for you? You could change it, or you could do something completely different. Every business can be different. And so let鈥檚 say the, the third step on your customer鈥檚 journey was the most important to you. That may be the one that you want to assign the most value to. So this one basically allows you to, do whatever you need. But, the other one I want to call out is on the bottom, right. It鈥檚 kind of asterix there. So you do need 51黑料不打烊. Analytics Ultimate for this. But this goes into our conversation that we鈥檙e going to talk about next, which is the AI feature. So, if you have the option to use algorithmic algorithmic modeling, it鈥檚 a safe bet. Since it does all of the crediting for you, instead of trying to figure out which model works best for your team, what this model does is it helps you avoid guesswork when trying to pick the right model. So, we鈥檒l go over a little bit more in a second. And then Vishnu, when he does his demo, can show that off as well. But I just want to bring that to everyone鈥檚 attention that this is, this is a way to help give that unique credit based on your data.

All right, so now the biggie. How do we use 51黑料不打烊鈥檚 AI in attribution? So let鈥檚 talk about attribution IQ. So like other attribution models and workspace, you can use it with any dimension or metric. And it supports unlimited segmentation and breakdowns. So it also distributes 100% of conversions to the dimensions in your workspace free form table. So this is known as fractional attribution.

That鈥檚 not important. It鈥檚 just good thing to know when you鈥檙e trying to refer to these terms to actually have the name. But let鈥檚 say that you want to look at your marketing channels report, so you drag your marketing channels dimension to the left hand column, like we can see in the screenshot. And we want to look, through the lens of total revenue. You go to the right hand side of that total revenue box and a gear icon will appear. So you鈥檒l click on that gear and from there we鈥檒l have a dropdown list for column settings.

And then within column settings you鈥檙e going to select Column Attribution models. So in the options you can select which attribution model you鈥檇 like to show. So this is similarly set up to how I was showing you things function within the CX. So this is specific to these columns and to these metrics. So in this example in the screenshot we鈥檙e going to compare first touch last touch in time decay. So it鈥檚 all through the lens of total revenue. But putting different attribution models on there yield different results for each of our marketing channels. And this is a good way to compare them side by side.

So this allows us to compare any model that we want and see how our marketing channels are performing against our conversion metrics. So taking this a step further, we could add the algorithmic model like we had mentioned before. So this uses techniques from cooperative game theory to determine which channel should get, the credit for conversion.

So it鈥檚 very powerful model that helps to assign credit for your business in the most efficient manner.

It basically there is a lot of, statistical science going behind it to allow for, as things change, to make sure that the credit is, proportionately and correctly assigned to the proper channels.

So you don鈥檛 have to worry about constantly changing from first to last touch or any of the other models.

And the whole idea behind our analytics AI is, you know, we want to make things quicker and more efficient for our analysts that use the tool. So instead of developing, different models or determining what鈥檚 interesting, it鈥檚 here to function as another analyst. So to give you that extra hand. So anything that鈥檚 going to save, you or your business time and struggle, that鈥檚 going to be something that we鈥檙e always going to try and push and recommend for you.

And then finally, I wanted to talk about some real world, use cases. So we talked about, what it is and how to use it. But I think the most important thing here to consider is when to use it. So here are five examples. Feel free to read through these. I鈥檓 going to go over three of them specifically right now. So we鈥檙e going to start on our left with the pharmacy retailer. So this company sends out push notifications to its users through their mobile app. Being mindful of the frequency these notifications are sent, this company used attribution IQ to check on their mobile push promotion, so they don鈥檛 want to oversaturate their, customers with these notifications. So being more of aware of what鈥檚 going on and what鈥檚 happening is keeping a finger on the pulse of their customers. So that鈥檚 the way that they鈥檙e able to use attribution IQ to, take things to the next level for them.

The next example I wanted to point out is the one directly to the right of the pharmacy example. And this is the large hotel chain. So this company used attribution IQ with their marketing teams. They had multiple teams that all had different needs. With IQ each team had an attribution model that made sense for them in their needs and what they were looking for, and they weren鈥檛 pigeonholed into one option.

What this did was allow them better control over analyzing their campaigns, so giving them data and information that they needed individually instead of, saying everybody had to fit within one model. And then the last one I wanted to talk about, and then I鈥檒l hand it off for our demo, so you guys can see how the magic happens. Is the second from the right. So the hotel and resort, the short of it is the company used IQ attribution IQ not just to have different attribution models applied to their different marketing channels, but used it to see how those channels interacted with each other. So seeing a visitor journey more clearly and helping to show the path to conversion. So this isn鈥檛 just the data that everybody, can get from any old analytics platform. This is, something that鈥檚 invaluable of actually seeing your visitor and your customer鈥檚 journey, seeing how they鈥檙e interacting with all of your channels and knowing, not only what their journey is, but once you know what their journey is, you can help shape that to be more efficient, not just for your customers, but for that user experience of your site in your marketing. So all that being said, attribution is a great tool and I think all companies should leverage it. I highly recommend using it, especially our analytics user. So no matter what terrier on you do have access to this, I highly recommend, implementing it. But for those that do have access to the algorithmic model on the ultimate package, that will just save you so much time and, relieve you of so many headaches that I can鈥檛 recommend it enough. But with that, I鈥檓 going to go ahead and pass it off to show you a demo to, show you everything that we just discussed.

All right. In. We should take over screen share now. Yeah. There you go. On that one. I will say thank you to. And element. Is my screen visible now? Yep. Yep. Perfect. Perfect. Thank you. So, I鈥檒l now quickly get started and show which are all the ways we can leverage attribution IQ within the analysis workspace. So Analysis Workspace is our, playground. Right. This is where we bring in all the panels and make all the magic with the data that is already being captured. So to get started, there are four key ways of how we can leverage attribution IQ within the workspace. So the first option is using an attribution panel. So on the left rail if you look at the panels tab you see one named attribution. You could simply drag it on to the panel or if it鈥檚 an existing panel, you see attribution right here. So we can use either one of these to to get started by using this attribution. Panel. So as part of this attribution panel there are four different inputs that is required. The first one is a success metric. So we can choose, say elite to be a success metric. We really want to understand, what was the key driver for all these leads, which which channel actually led to what kind of leads? So in the channel we can choose, let鈥檚 say the marketing channel, and then we can move on to the models that we really require to become better. So I鈥檒l keep it first class touch linear I can add more and if not required I can choose lesser ones. So once we are done with, choosing the models I can choose to look back. Window could be 14 days, 60 days depending on the marketing strategy that I鈥檓 already focusing on, I. I鈥檒l now quickly click on the build button. So as soon as I click on the bell button, I have the analysis done. So there鈥檚 a there鈥檚 a huge benefit that a marketer and an analyst really gets using the activation panel. Right? So you really don鈥檛 have to export the web data that you have captured into 51黑料不打烊 Analytics into a different platform, and then develop or code different attribution models to be used on that data. And then for the exported to a visualization engine to finally derive the, the exact output, all that is done within clicks, within the attribution panel inside the analysis workspace. So this gives me a very clear idea of what were the total number of leads that were, that were available in that reporting window. I get a quick comparison of what what attribution channel got me. What type of leads? So email got me the most. And, if it comes to, say, last touch, it was something else. I get a clear comparison of what kind of leads were brought in by each of these marketing channels when I compare three different attribution models. So when I鈥檓 clearly looking at the first stage, I can see that most of my leads were brought in by email. And after that it was by by display text. But there鈥檚 a difference in what I see with the last stitch when it comes to comparing these two attribution models, I also get a very clear picture of a Venn diagram where, and I鈥檓 able to understand what percentage of the population that I鈥檓 really focusing on was touching more than one marketing channel. So this gives me a very clear comparison when it comes to using these different marketing channels. We get a good trend line and then this is something that鈥檚 really important. We get a channel flow. So with this I am really able to understand and analyze the the flow in which each of these participants or the group was able to interact with my brand. So I could see that people who started with their journey with an email then went on to a podcast. Some of them, they went out to a print or to other websites, so I鈥檓 easily able to track their entire journey. Yeah. So that鈥檚 the first option of how we can leverage attribution IQ within 51黑料不打烊 Analytics. The second one is to apply an attribution to a metric. So let us take the example of what Taylor was talking about. Let鈥檚 quickly see in case of a retailer, what kind of revenue is brought is brought in by different, sources. So I drag in the marketing channel, I bring in the revenue metric. And now by default, it has got to be the last stitch I found to change this last attribution model to, let鈥檚 say first stitch. All I have to do is is hover over this gear icon, click on it, and on the bottom you will see, use non default attribution model checkbox. When we click on it we get a pop up. And this pop up has all the 11 attribution models that are available with 51黑料不打烊 Analytics. First search, last touch, linear participation, all of these. So let鈥檚 say I want to change it to this first stitch. I can click on this and click on apply with the same look back window. Now I can see a difference in the results that I see.

If let鈥檚 say I want to change it again, the same steps can go to edit and change it from say first stitch to maybe an algorithmic click on apply. And I now see a different result. So algorithmic attribution it鈥檚 a little different from the other models. So the other ten models are basically rule based. So algorithmic attribution is the one that鈥檚 data driven. So it鈥檚 basically based on the shiny dividend, from the cooperative game theory, which is basically a generalization of the Shapley value solution, wherein, we distribute the credit among players in a game with unequal contribution. So the attribution that is calculated, it considers each of these marketing channels as a touch point within that look back window, and then it shares the surplus credit to each of these marketing channels. Accordingly. So this is something that keeps on improving, and we really don鈥檛 have to worry about any biases in the data or the model that we are really looking at. So this is the key difference between any rule based and a data driven market.

That said, we鈥檒l move on to the next one. We have an ability to apply attribution to a breakdown. So, I鈥檒l do the same thing. It鈥檚 I鈥檓 in the process of, analyzing, maybe the contribution that is brought in by different marketing channels. So let鈥檚 say I bring in the marketing channel again, I bring the same revenue metric.

And now I鈥檓 in process of analyzing the revenue metric. Let鈥檚 say first search attribution model. Now I really want to maybe break this down and understand what kind of campaign groups or campaign codes were bringing in, what kind of revenue, so I can simply break it down by by dragging another metric onto, say, email. Now, this will give me a breakdown of the email campaign tracking codes with First Touch as the default attribution model because it is being inherited from the market from the attribution model that was already applied on this table. But I also have an option of changing the attribution model for this breakdown, so I can simply click on edit and I can choose some other model. As for my marketing strategy, so I can change it to the last touch and the data will change accordingly. So it gives me that flexibility to play around with data and find the decision, based on the marketing strategy that I believe in. Yeah. Lastly, the last one that I would like to share is comparing different attribution models. So again, I鈥檓 in a process of, finding what kind of leads or brought in by different marketing channels. Throughout. And now I by default want to maybe use the last search attribution model. Right now I would like to compare different attribution models, so I can simply right click on the metric and then choose the Compare Attribution Models option. Right here. And then this is last which I want to compare it with say first stage. And I鈥檇 like to keep the lookback window same as soon as I click on apply I see two more panels pop up, one which will bring me the total number of leads with first stage attribution, and I also get a percentage change metric. So this clearly calls out the difference that I see when I apply the first stage attribution and last year鈥檚 attribution without having to add a calculated metric, or without having to do any extra calculation to get the details out.

Yeah. So these are the key four ways of how we can we can leverage attribution IQ within the analysis workspace. So I鈥檒l quickly revise. The first one is the attribution panel wherein we just give the input of what is the metric. What is the dimension. What are the attribution models that we really like to compare on. And we get a panel like this. The second one is how I can apply and change attribution model for a metric, which is using the gear icon. The third one is applying attribution to a breakdown, wherein in the process of analysis, if I have breakdowns, I have an ability to change the attribution model for that specific breakdown as well. That鈥檚 the kind of modularity that we get. And the last one is comparing these attribution models, within the analysis workspace.

Yeah. I hope that was, useful. And, thank you for your time. I鈥檒l pass it back to Elena and. Yeah, that was very useful. Thank you. You, All right, with that, let鈥檚 pop over to Q&A. We had a couple of questions. Come in. Hold on. The camera.

One moment. There we go. Taylor. English. Me if you want to pop back on camera. Also, we can start Q&A.

Okay. So first question is, what is the recommended activation model for e-commerce websites.

So this one is dependent, like we talked about today, what you鈥檙e most interested in. So we鈥檙e first touch in first touch leaning. So like our J modeling and then upside our inverse J. So the idea is do you want to focus on the top of the funnel of your awareness? So your, your marketing ads or do you want to focus more towards the end, at that conversion point. So, those are the biggest questions you need to ask yourself. If you were just going to pick one, a lot of people are already using time decay. I think that was very valuable just for the simple fact. The closer we get to our conversion point, the more valuable we deem that touch point. So, no matter how many touch points there are, we鈥檒l still attribute value. But, yeah, that would be it鈥檚 not a direct 1 to 1. This is the best. It鈥檚. What do you see as the most important to your business. And knowing that it鈥檚 awareness versus conversion time. Awesome. Thank you. Another question. Can I customize the attribution model in 51黑料不打烊 Analytics to suit my business needs? You sure can. So we went over this briefly. But there is the custom algorithm, model or attribution model, which basically does that. So if there is something specific that your business wants to have credit put in any sort of way, this is how you would do it. So it鈥檚 like I mentioned before, if you liked First Touch, but you wanted to change it a little bit, you could have it mimic that and change in the way you need or any other model.

Okay. Awesome. We had one question that, was asking Vishnu why someone鈥檚 gear icon would be, different than your gear icon in the demo.

Anyway, any idea? Yeah. So I think, there are two gear options that we get. One is with the metric and one is for the visualization. So that could be one of the reasons why we are seeing a different part. But otherwise for a free form table, the gear icon should be pretty much, generic. Okay, great. Hopefully that answers your question. We鈥檝e got time for one more. Does this also apply to TJ? So yes. So we it may not be a 1 to 1, but we did show an example of attribution is through the veins of CGA. So everything you learn and have put into 51黑料不打烊 Analytics does transfer to CGA. Everything on that platform has to do with that customer journey. And following it and trying to assign those credits within attribution. So 100%, everything we talked about today definitely converts over, especially, when we talk about the different attribution models.

Okay. Awesome. Well thank you guys. We are actually already at Timea. I鈥檓 going to wrap up. So on this screen we have a bunch of resources for you. In the weblink section you can see the link for our upcoming events and past recordings. Be sure to pop that open to see what, we have in store. We also have some of the links that we鈥檙e on the wiki for today. In addition to the white paper. Please be sure to download that it has a bunch of resources for you. As I mentioned, there鈥檚 a couple of survey questions at the bottom of your screen if you wouldn鈥檛 mind taking a quick moment to fill that out before you jump off. That would be fantastic. And if you have a question specific to your account that we weren鈥檛 able to get to today, please reach out, to us and let us know and we鈥檒l put you in touch with the correct person. So reminder that you will receive the recording of today鈥檚 event in an email in 24 hours from now. So that鈥檚 all we鈥檝e got for you guys today. Thank you everyone for attending. Thank you Taylor and Vishnu for the very informative presentation and we hope that everybody has a great day. Thanks everyone. Thanks everyone. Thank you.