On-Demand Webinar Overview
Are you still relying solely on last-click attribution tracking? If so, you’re answering your performance questions with incomplete data. Cross-channel marketing campaigns require cross-channel, multi-touch attribution tracking. Though insurmountable for many in the past, today’s technology has made cross-channel attribution accessible and affordable. The most successful marketers are tracking their campaigns holistically, and you can too. In this webinar, Sparkroom’s Akeel Haider & Anthony Nater will explain why last-click attribution doesn’t work in a world of fragmented media and how you can get cross-channel attribution tracking quickly in place.
Akeel Haider - VP Innovation & Strategy @ DMS
Akeel is the Vice President of Innovation and Strategy at Digital Media Solutions, responsible for the Sparkroom software and media services evolution to meet current and future market demands. Akeel has over 12 years of experience in performance-based marketing, cross-channel advertising, product development, and marketing analytics. Akeel has successfully lead the creation and execution of hundreds of digital media campaigns and services related to lead generation, mobile media distribution, and digital subscription products.
Anthony Nater - Product Manager @ DMS
As a Sparkroom product manager, Anthony blends his strong focus for pixel-perfect, user-centric product design with his online marketing sensibilities to stay on the cutting edge. While keeping a close eye on inquiry generation needs and continually assessing market trends, his most recent priorities include road-mapping and developing the integration, mapping and display of cross-channel marketing data to provide marketers with actionable attribution data.
Kathy: Good afternoon. My name is Kathy Bryan, and I’m the Vice President of Corporate Marketing and Communications here at Digital Media Solutions. I’d like to welcome everyone to today’s webinar, during which we’ll dive in to the importance of cross channel attribution and how you can get started with enhanced tracking efforts. Before we get started, I wanna briefly tell you a little bit about Sparkroom.
Sparkroom is exclusively focused on education and is a proven leader in enrollment marketing and management. Through the deployment of a world winning technology and services Sparkroom helps institutions of higher education grow and sustain enrollment. We offer our solutions in three formats as you see on the screen. Performance marketing technology is the foundation for everything we do. Enrollment marketing and enrollment management are the other two services as you see. Before I pass it off to our presenters, Anthony Nater and Akeel Haider to introduce themselves, I want to let everyone know that we’ll be assuming certain level of marketing proficiency as we present this topic. But, don’t worry if you hear any terms you are not familiar with, please feel free to submit a question for clarification. I will either respond via the gotowebinar chat tool, or if appropriate I will interrupt the presentation for clarification.
So, thank you for joining us today and now I will pass it off to Anthony and Akeel.
Akeel: Thank you Kathy, I’ll be introducing myself, my name is Akeel Haider and I’m the VP of Innovation Strategy at Sparkroom. Previously, I was in charge of audience marketing. My role is more on the product side. I have 15 years of additional marketing experience in all the channel experience and cross channel marketing and attribution. Beyond that recently I took on the role of product in terms of innovation. My job is to align our product to the market needs and future in terms of innovations to meet our client demands as well.
Anthony: And, I’m Anthony Nater, I’m a Product Manager here at Sparkroom, I’ve been with the company over five years and collectively I bring over 13 years of experience within digital marketing and web development. Here at Sparkroom, I’m primarily responsible for driving parts of our software around integrating our reporting and analytics in to third party data sources. So, things like Google, Bing, Google Analytics and AdWords, and as well as some of our marketing services APIs that we have available for client websites.
So, today, what we’re going to talk about is centered around cross channel marketing attribution. So, different than our previous webinar that Akeel and I also have posted. We’re gonna take it a bit of a different approach. So, we’re gonna start off by talking about, what is really expected of marketers today to really understand cross channel marketing attribution and how it applies to them day by day. We’re gonna touch on last click attribution and why it may not always work, and the type of analysis that you need to do. And, how cross channel and multi touch attribution reporting and analytics can really help you do the types of analysis that helps drive more optimized campaigns and marketing mix.
We’re then gonna go in to what solutions are currently available for you today to get started with this type of analysis. Including what Google Analytics can offer for free, and we’re gonna dive in to more education focused. Why it’s important to join your pre-lead to some of the post-lead data, as well as some of the milestones that go beyond just the leads that are generated. And, we’re gonna look at different ways you can gain some insights and drive some actions from some of this type of reporting. I’m gonna pass it over to Akeel to kick this off.
Akeel: Thanks, Anthony, so, let’s talk about the expectations for marketers today. It’s really not enough for marketers today to do predictive audience targeting and one cross channel media companies to reach that target audience. What’s expected of them, what we see, the client demands of the market is that not only to do that, but to be able to measure and understand the full value of each marketing channel beyond the last click. To understand the impact on a channel on each of the marketing channels and campaigns you’re doing on one another and the overall ROI.
Also, to understand what combination and touch points yield the most with the least cost. In essence, there’s a need to credit a channel for the real value it’s providing. And, that’s what we’ll be talking about. Specifically, beyond impressions beyond the last click, mainly what are the impacts of impressions and clicks on basically your ROI and your media mix.
Right, so the best analogy for that, I think if you take a look at a sports team. Maybe, we start with saying, okay, what is really attribution? Let’s start with that. So, it’s really trying to accurately assign a value to each touch point across the user journey that they’re taking to becoming your customer. And, beyond that is understanding which combination of channels and touch points that are leading that person to becoming a student or customer. Or in the case of e-commerce maybe a sell. So, in that a good analogy would be really something like a soccer team or a basketball team.
Now, what you want to know is what is each player best at? What are the attributes of each player and then what is the combination of that team, in terms of how you put it together to provide the best value for you. So, in defense you want someone that’s going to tackle well and in the midfield, you want someone who is going to be able to pass the ball well, and definitely in the offence you want someone, I’m talking about soccer in this case. You need a good finisher somebody with skills to finish and so this is how we kind of look at marketing channels. Some are going to be openers, some are going to be influencers and others are going to be closers.
So, looking at things from just the last look at attribution and saying only the closers and the finishers and the ones that I’m going to attribute all the value of how my marketing mix is working, is really not a good way to look at that. And, our job is to understand, okay, what is the value of again each of those players and what is the best combination of that team to get me the best deal. So, let’s move to the next one.
So, in that sense, if we look at our last look at attribution, we have a problem already. So, someone who has seen an app and clicked on that ad, so you have an impression in play, then went to your landing page or clicked on the call number to generate that inbound call or filled out a form, and then became an enquiry, and then moved down the funnel. That is really missing the point of how that person got to that last click. What is that user journey that the research that happened. What influenced him along the way to basically get to that point. So, that’s what, if you’re looking at just last click, again that’s what you will see.
So, let go to the next slide. We have another challenge. The challenge is, in general CRMs are not good at tracking that user journey. That pre-lead user journey or housing it for that man. The challenge really is one of tracking for those touch point, housing the data before even being able to start analyzing and visualizing it. The users are taking various paths of data, Google’s data, if you run multi-channel campaigns, you will see very clearly that there’s a nonlinear, but it’s nonlinear and there’s multiple paths and users take multiple paths to get to you, and beyond that there is then the offline impact and online and offline there is also the cross-device problem.
Where people are crossing devices just to confuse. So, there is multiples challenges. But, since we’re talking about last click attribution, let’s move to the next point. All right. So, I’m going to show you a few campaigns that we’ve done in the last couple of years to give you a quick example of the data. Kind of what we see when it comes to the complexity of that user journey and the nonlinear path. So, here we have an example of a display company. What we’ve done is this is an example of a predictive display targeting campaign, that means we’ve created a lookalike audience and we went out to target that audience regardless of device or website.
On basically using display banners and a little bit of social media to be able to track what’s going to happen to them. Since we know this is the right audience for us, we’ve done all the modelling already. What is the impact of that display campaign? What we’ve seen is very interesting results. And, the reason we’re seeing the interesting result is we’re able to track that impression all the way to the bottom of the funnel converge in a long bleed life cycle, like education. We’re able to attribute that impression back to say a start, or whatever you would consider a conversion in your case, like an enrolment.
The immediate impact would seem that only a small portion – well first of all, the campaign yielded in 11 percent lift in starts, but here’s the interesting part. The interesting part is that only two percent of those came from a last click attribution that we were able to track through a last click. Everyone else came through either somebody who clicked, but then decided to do some more research and come back through another channel. Or in some cases multiple channel enquiries. Or someone who was influenced just by looking at the ad again, and having clicked on it but the impression alone was valuable,
And, then because again we’re targeting like audience. They went and done their own research and came back, days of sometimes weeks after to enquire and then eventually that enquiry became more than that. The immediate thing you can see here is that, I think one way to measure those top of the funnel campaigns, even if you don’t have complicated tracking in place, like we do. Is, to look at the brand impressions. The Google brand term search volume, and from that you will see that for this campaign specifically about two weeks in, we’ve seen that immediately there was a lift of 26 percent that were clearly attributed to this campaign.
From just people who had started searching on Google for that brand. That’s beyond the click or what not, that’s just an impression that happened from someone searching on the brand and that you can get. This type of data you can get from Google AdWords or from the Google Consul or it used to be called Webmaster Tools. Yep, so anyway, so if you were just looking at the last click attribution, clearly this campaign would not be successful, at least from an ROI perspective based on the cost that happened to it. And, yep that’s whys it’s important.
Anthony: Akeel, so even though this display campaign doesn’t really directly drive those leads or conversions, but it’s fair to say we can conclude this is a good brand play. It’s a good tactic to start a branding campaign. Because, it actually it started those users along that conversion path?
Akeel: Yeah absolutely, in this case we’ve seen that the value of the impression alone justified the cost of the campaign, because we were able to calculate our line beyond just the last click. I mean, each campaign you’re going to produce, especially top of the funnel [inaudible] [00:16:05] will have some impact. It’s all a question of how you measure it and there’s different techniques for that.
All right. The second example that I want to talk about very quickly here, is that an automated lead life cycle email campaign. So, this is an email nurturing campaign, that’s fully automated that will help the prospect move down the funnel, between the different milestones. So, what we’ve noticed, again without talking a lot about what the campaign was. What we’ve noticed is again, every campaign we look at has those kinds of interesting trends. So, what we’ve seen what people who have opened the email but never clicked on it. Have a 2.5 higher enrolment rate than those who have actually clicked on the email.
Those who have clicked the email have a four times higher enrolment rate than those who have not clicked. Clearly higher engagement. Us, being able to measure that it shows clearly that the path was not linear and then if you can have a more sophisticated attribution you can tell how they came to become – where it happened, was this the last click or somewhere else in the middle or was it the first click, or the first touch point. Beyond, that let’s take a look at a Facebook campaign that we’ve done recently for a large ground school campaign. What we’ve noticed here is again, same type of trends, just in a different way.
Social and this being specifically Facebook, is influencing all parts of the funnel. Since almost working as an awareness campaign as a last click campaign and an influencer as well in the middle. In this case for every 100 leads we were able to generate on a landing page from a direct click on the Facebook paid ads, and this was a lookalike targeted campaign based on the client’s data, and sort of our enhancement of that data. What we’re seeing here is that for every 100 leads that were generated on the landing page of that campaign, another 70 to 80 came from as a view through conversion through the client website. So, it provided an incredible life to the website.
And, again if you were just looking at last click attribution you wouldn’t be able to see some of those.
So, moving on for us to be able to understand attribution, how should we look at it, attribution itself is not a technology it’s more of a process and a framework. I think it starts from understanding that online user behavior, which we talked about moving between the different touch points and channels. It’s a long journey. And, there’s multiple touch points along that journey. Beyond that, tracking that media correctly is very important. How you are housing the data including the data is also important, Then, joining what we call that user journey from the upper part of the funnel to post enquiry milestones and down the funnel, that’s really the key here.
The key is to be able to join that data to see the full journey. And, then can we visualize digest that data in a way that makes it actionable. And, then the final part is about the really, what point we’re talking about, attribution we’re talking about attribution modelling and the idea here is, can you then create data driven attribution models that will help you do forecasting, testing and eventually optimizations of your campaign. To either lower the effective cost, ECPA, effective cost per acquisition and also increase your volume across the board.
Anthony: Great, thanks Akeel. So, we talked about the value that looking at your data from a more multi-channel cross attribution stand point really has. And Akeel has demonstrated through some of the examples. But where do we go from here? So, what I usually like to say, a good place to start is understanding your goals and answering these three basic questions, before you even start to really get your hands dirty with some of this type of analysis. So, really understanding what is your current digital marketing mix. What are the types of channels that you’re currently running? What channels are you considering adding in to that mix, or maybe even removing from that mix?
And you really need to prove some of those decisions you need to make. And do you have any idea what your typical conversion path currently is? Does it take a user or prospective lead several weeks to convert because their hopping around these different marketing channels? Is it more direct to lead from one marketing channel and then they’re from a last click stand point, they’re directly converting? What are the goals you really need to track? So, do you wanna track the lead or are you tracking maybe some micro events that you want to attribute to some of these marketing efforts? Or are you trying to track something deeper down that conversion funnel as we are in education?
What are the solutions that are available? There’s a host of software providers out there that allow this type of analysis that you can plug your data in, they get pretty robust, like you can imagine with some probabilistic analysis of cross device attribution and online to offline. But, like I mentioned in a previous webinar, a good place to start is actually with Google Analytics, I don’t wanna go in to the details of how to configure that again, because that was discussed last time, but I do wanna stress that it’s a free tool that’s available and they do offer a host of different types of reporting that can really get you started.
And, at least understanding if this is something you need to really dig in to deeper, and apply to your day to day marketing efforts and optimization. So, one thing Google provides is different types of reports around your conversion paths, understanding your assisted conversions, different types of attribution modelling you can apply to that, and how the channels that you’re running may overlap. Now, this is important to consider when you’re looking at other software providers as well, because these are kind of the basic types of attribution reports you really need out of the bat. So, if you’re considering one solution versus another, and you feel that they might be lacking in some of these areas.
This is something you might wanna way against what options you have available for using something like Google Analytics, which is more cost effective, because obviously, it’s free.
Akeel: Anthony. Does Google Analytics take its consideration in things like offline cross devices or is it only digital?
Anthony: So, primarily it’s only digital, so whatever you have Google Analytics configured for, so like your website and/or landing pages. So, it’s strictly digital. However, they do allow you to import offline data, offline cost data as well. It’s a little bit more manual, but they allow you to do that. But, right out of the box it tracks all of your Google sources seamlessly, anything that’s not a Google source, obviously, you have to get in to specific tagging methods and things like that, so it’s tracked appropriately. So, digging in to one of the reports here. A Standard multi-channel attribution path analysis.
It’s important to look at this type of report to really understand which channels are working together and influencing one another. How they can really help you optimize your messaging and the creatives. If, some of these channels do work well together, you might wanna plan and coordinate some of these campaigns between these different channels simultaneously. Looking in to how we can analyze some of this data. So, taking for example tying back to what Akeel mentioned earlier around social network marketing.
If we are looking at this report and we’re interested to see how social network was performing. Just by looking at this table we can see that, all right, our social media campaigns have overall contributed and can be attributed to 28 total conversion. So, what does that mean. That essentially means that it took place somewhere within that conversion path, before the user finally converted on the last click stand point. We also can determine that it assisted 11 conversions and also started 11 conversions. So, that again, is where it fell along that conversion path. It necessarily wasn’t the closer, but it assisted that overall marketing mix in securing that lead.
But, then traditionally, we can also see that it closed or generated 17 good leads for this example, and this is traditionally how you would be looking at this type of reporting. Well, you’d be looking at your standard reporting and say, all right, our social media campaigns, generated 17 leads and it would kind of end there. But, if you look at this type of analysis, you can really understand that, it generated 17 leads, but it actually can be attributed to 28 conversions or leads from different various combination of these marketing channels, and assisted 11 conversions. So, it’s important to look at your data in this way.
But, going a bit beyond just the enquiry. So, this applies to some verticals and specifically education where we’re looking beyond just the lead that’s generated is very important. So, in education a lot of times we required to do a type of analysis beyond the leads that are generated to really ensure the quality of those leads tie back to the marketing efforts that we’ve been running. So, in education it’s enrolments or starts, and ensuring how some of that post-lead data actually ties back to your pre-lead marketing campaigns. And, the common conversion paths, is really important to understand what’s the most optimal mix of your media.
To drive the most quality leads that then translate in to these deeper conversion milestones. Briefly, what are we doing about some of this in Sparkroom? So, we’re actually trying to address a lot of this with Sparkroom by integrating directly with Google Analytics, and taking the data that’s available within Google Analytics, all of your pre-lead marketing data, joining it to that post-lead data that we have within Sparkroom and expanding on that pre-lead data that Google provides to really show you how these deeper milestones can actually tie and map to some of your pre-lead conversion paths.
So, by looking at something like this, you can see how things like enrolments or starts for example for education actually map to these conversion paths, so you can better understand that. All right, even though this channel might be generating a ton of leads, are they really translating in to what really matters? The goals that we really want to count as enrolments and starts.
Akeel: Anthony, so I have a couple of questions for you here very quickly. One is, if you can clarify what the X2 and X3 are on those and I should just also briefly add a comment that this is our kind of attempt at solving that and joining the data. Understanding it and the solution for it might take different shapes. This is just one way of –
Akeel: We’re not saying that this is the ultimate play, we’re saying that this is one way you can have insight in to this –
Anthony: Yeah, so the times two, times three this we kind of just lifted from, we followed how Google presents its data, and essentially just means this specific channel grouping occurred two times or three times. As in the example here, with page search times two or times three. It just condensing the data a little bit more so it’s easier to read. Because, as you start to go a little bit deeper and granular in some of these different –
Akeel: It might be a long path.
Anthony: It could be a long path in to see like page search, page search, page search, and then, display, display, display. It can get a little confusing.
Akeel: So, what you’re saying is that there’s a group of users who whether they clicked on it or not and then became a good lead.
Anthony: Well, they interacted with that page search campaign.
Akeel: Yeah, so whether that’s by impression or a click.
Anthony: Exactly. And, we’re doing this again by joining the data that we’re getting from Google in joining it to the post-lead data that we have within Sparkroom. So, this is really expanding and adding more value to the free data that Google provides you right out of the box. And, just another couple of screenshots of what we’re actually doing to provide this. So, we’re taking it down a little bit deeper in to how these channels are actually interacting with one another. So, again showing something like total contributed conversions. Which conversions can be interpreted as any milestone, so you can filter by good leads, enrolments or starts.
Last interaction or first interaction conversions. Also, really understanding what channels are more assisters or closers. So, like that last kind of donut widget on the bottom, if you’re looking at a specific channel, it can really help you determine that. This channel that I’m looking at, example would be page search. It’s closing more leads than it’s assisting. So, that might help you derive some insight in around how you wanna allocate your budget for that specific channel verse another that you might be looking at. But, yeah that’s just something that we’re doing, that I kinda want to throw in there, in to the presentation. But, yeah, so back to you Akeel.
So, just wrapping how this all ties back to your first point around, what is really expected of marketers today?
Akeel: Yeah, so if we again, bring it back up to the kind of the first level we talked about. In terms of what we’re looking for, if you are not seeing beyond the last click, the conclusion here is that you have to see beyond the last click to really measure and do attribution more correctly. Now, sophistication of how you would do that, there is multiple levels to that, but just the mind shift of understanding that and trying to resolve it, might get you closer to accuracy than not. So, I’ll leave that at there.
Anthony: Okay, so I’ll go back to Kathy, who will end up here.
Kathy: So, I’ll explain that, that quick picture is, so instead of the last click where you’re just inside the box and only seeing what’s inside that box at the very end. This is being freed from that box. So, with that thank you Anthony and Akeel. We’re gonna take some questions from the attendees right now. If you have a question, please type it in to the question box on your webinar control panel. If you can’t see the control panel, look for the orange box with the arrow, click on the arrow and the control panel will extend allowing you to type in your questions.
So, I am going to start with a question of mine. So, as you’ve been going through this, you were talking about kind of the lack of value of last click. I’m really summarizing or really over summarizing what you were saying. Are you suggesting that last click attribution has no value or are there times when you still would use last click attribution?
Akeel: So, it’s not like last click attribution is out of the picture. Last click attribution will always have a value and probably the value of that is more than others. Because, again if you’re in to basketball or soccer in this case, you understand that players like Messi, or Christian Ronaldo, I mean, again, so I’m a soccer fan. You understand that the emphasis is those players are the highest paid in the world. It’s because they’re the last click attribution. They can close seven out of ten. They can put a goal in. And, so there’s definitely a lot of value and that it doesn’t go away.
It’s just that the last click is always gonna be there, the question is, well what about all those players that have really played very well to get the ball to that person to score.
Kathy: As a defender, I feel ya.
Anthony: It’s not always the case, so we’re just expanding on that point quickly. It’s last click attribution in some cases yeah, like if we’re looking at the conversion path, it might only be one channel that’s in that path. But, you have to also pay attention to the other channels that influence each other.
Akeel: Yeah, and so that’s when you start doing things, like what is the percentage of my group paths that are bringing the most direct the most. And, that’s when you do multiple analysis on that and you will see that some of the shorter path might indeed be yielding maybe 70 percent or 60 percent of your actual prospects. But, maybe this other 30 percent what do you wanna do with that? Try to help you out and sometimes it’s vice versa.
Kathy, sometimes there are teams that win again being a soccer fan, I’m asking about something only Europeans would know, Leicester City last year they won the Premier League and they were nowhere near the top, never been to the top. The reason for that is they played as a team, they didn’t have necessarily a lot of finishers, but because the team overall played incredibly well, they were able to still achieve the goal.
Kathy: Okay. So, a question from the attendees. I see direct coming up a lot in attribution, but direct is a bit vague. It used to be representative of how one user typed in the url directly, but it has lost specificity. How do you think of direct traffic or is it simply unknown now?
Anthony: Yeah, I could answer that one, I guess. So, specifically the way that we are requiring some of our clients to configure their Google Analytics right, so where direct actually gets assigned to the traffic that they’re analyzing, there’s different ways to ensure that the incoming traffic is configured properly within Google Analytics, so ensuring that your channel groupings and settings and everything is as clean as possible is definitely a first step. Because, a lot of times and I’ve found this time and time again, once you start digging in to say a direct channel in something like Google Analytics. You will notice some emails getting mixed in there, you notice some page search campaigns getting mixed in there.
So, it’s really coming down to making sure your data is as clean as possible. But, in the end, direct is still a direct visit. Someone came to your website direct.
Akeel: To add to that one important point as well, is that if you leave Google Analytics without configuration it’s an excellent point, because GA will group a lot of this information that’s not configured in to direct. And, it’s not necessarily correct.
Kathy: So, Google Analytics isn’t just a turn it on and it works for you kind of technology?
Anthony: Right. No, it definitely needs to be configured properly, so you’re looking at the data and interpreting it correctly.
Kathy: Another question that came in. What CRM systems does Sparkroom integrate with?
Akeel: I’ll take on that one. We integrate with any CRM system that is able to provide sort of output data either in real time or automated. It doesn’t matter to us we integrate with almost, I don’t know, ten twelve different CRMs out there today. We can integrate with anyone that’s gonna provide us with APIs, to post in and sort of retrieve data from in real time. Or, if they have the ability to quantitate what we call a nightly file via FDP or as such.
Kathy: And there was a question specifically do we integrate with Campus Management?
Akeel: Absolutely, that’s one of the top integrations we have today.
Kathy If my campaign has only lead generation tactics, so I don’t have any branding as part of my current campaign, can I put more emphasis on last click attribution than on looking at the cross channel?
Akeel: Yeah, again it’s a very good question. So, to begin with I think if you’re looking at your lead generation, I guess what you’re saying is you’re probably paying on a per lead model. And, in that case those leads that are coming to you are still going through a path. They’re still going through a journey. Yes, you can put the last emphasis on there because your ROI is really very clear and dependent on that. But, at the same time with those PPL channels, and I can assure you, and we’ve kind of measured and analyzed that as well.
Is that they are providing a lift to your brand, your website, whether it’s a direct lead or somebody putting an ad out there, and ad out there that has your name and brand. There’s definitely an impact. How to measure it, now it’s a different story. If the providers are allowing you to tag or basically pixel some of that content, then that’s great. If they’re not then that’s when offline attribution solutions can come in to place and maybe without getting in to it there are various solutions that allow you then to do some data reconciliation. If you have the information available of the PII, basically data. But, yeah, that’s much more complex.
I’ll leave it at that.
Kathy: So, on a couple of slides you had start, assist and close as different kinds of media or different things that you could be looking at. Which should I value the most between those?
Akeel: Go ahead.
Anthony: I can take that. So, it really depends on the type of campaigns you might be running. So, for example towards Akeel’s example that he gave, he was looking at a display campaign. Generally, display has its most power around some kind of branded type campaign or initiative. So, if you are trying to focus on expanding your brand and you want to run more display campaigns, you should expect to see a lot of these different channels to become part of the mix before you start generating leads or sales, or something like that. But, that’s just one example. Did you want me to go on?
Akeel: No, I think that’s perfect.
Kathy: Okay, and then a final question, unless we get some more. Explain more about how cross channel attribution helps me forecast?
Akeel: Yeah, so I think that’s really the key there, it’s important. So, to get to the point of forecasting, you need to be able to have that framework that we discussed, all implemented very well. And, now you can see, okay, there’s multiple ways of forecasting. Clearly, there are some enterprise level software out there that you can have all your data flowing in to, and sing various algorithms and models. Everything from Xworks and game theory and sort of, I mean I can keep going in terms of the technicality of that, but what happens is how can you forecast the data, it’s all about understanding your media mix. What’s working?
Then taking a step back and inputting those specific algorithms and models and saying okay, what is the combination of that? That that’s gonna yield the most. There is no, if today you go and you say, okay, I think based on what I know so far, I’m going to use a time decay or a linear attribution model, you know that’s one way of looking at it. Because maybe just toggling between last click, first click interaction and a few of those, what we call, standard models, might definitely give better results. At the end of the day you need what I call a data driven attribution model. Based on your own media mix and what’s happening today.
There needs to be a specific data driven model that will be built to yield for you the most, and that’s really all a combination of testing an algorithmic solution out there. Without getting in to too much detail.
Anthony: Yeah, and just to quickly expand again, you know looking at your data, the volume of the data also isn’t working here, right. So, if you just set all this up and start looking at your data in a week. The probability of those common conversion paths, may shift over time. So, if you’re looking at maybe three months’ worth of data, and you’re really starting to identify common conversion paths, you’ve have higher probability that this is the media mix that I need. This I where my budget needs to be allocated. So, the overall volume of the data actually supports the probability of presenting those types.
Akeel: That’s an excellent point. So, let me take a step back and say that if we were to do it, we would take a look at your historical data. Probably going back, a year or two at least. Yeah, and then start to understand that. And, we would apply what we call and ensemble of various methods to analyze the data to see, okay, in almost a simulation type to see if this combination based on this algorithm yields the best. Or we can apply real time [inaudible] [00:41:50] network algorithm and sort of adjust like that. But, your right I guess the final point here is that, the data has to be refreshed and for you to even have a start baseline, you need I would say at least three months’ worth of data.
And, then after analyzing that data, then you need to refresh also every two or three month.
Anthony: That’s going to be essential.
Kathy: All right. Excellent. Well, thank you Anthony and Akeel. And, thank you to everyone who joined us today for the webinar. If you have any questions about cross channel attribution or if you would like some help configuring your Google Analytics, or lastly if you’d like to see a demo of how Sparkroom performance marketing technology solves the question of attribution challenge, please reach out to us at firstname.lastname@example.org. We wish you all great success in your tracking endeavors and have a great day.
Anthony: Thank you.
Akeel: Thank you very much.