{PODCAST} In-Ear Insights: Viewthrough Conversions and Attribution Analysis

{PODCAST} In-Ear Insights: Viewthrough Conversions and Attribution Analysis

In this episode of In-Ear Insights, Katie and Chris talk about the devil that is viewthrough conversions. What are they? Why do advertising networks rely on them so heavily? Do they have a place in your marketing analytics? Learn why viewthrough conversions can be greatly misleading and what alternatives exist for robust, real measurement of marketing campaigns.

Subscribe To This Show!

If you're not already subscribed to In-Ear Insights, get set up now!

Advertisement: Data Science 101 for Marketers

Do you want to understand data science better as a marketer? Would you like to learn whether it’s the right choice for your career? Do you need to know how to manage data science employees and vendors? Take the Data Science 101 workshop from Trust Insights.

In this 90-minute on-demand workshop, learn what data science is, why it matters to marketers, and how to embark on your marketing data science journey. You’ll learn:

  • How to build a KPI map
  • How to analyze and explore Google Analytics data
  • How to construct a valid hypothesis
  • Basics of centrality, distribution, regression, and clustering
  • Essential soft skills
  • How to hire data science professionals or agencies

The course comes with the video, audio recording, PDF of the slides, automated transcript, example KPI map, and sample workbook with data.

Get your copy by clicking here or visiting TrustInsights.ai/datascience101

Sponsor This Show!

Are you struggling to reach the right audiences? Trust Insights offers sponsorships in our newsletters, podcasts, and media properties to help your brand be seen and heard by the right people. Our media properties reach almost 100,000 people every week, from the In Ear Insights podcast to the Almost Timely and In the Headlights newsletters. Reach out to us today to learn more.

Watch the video here:

Can’t see anything? Watch it on YouTube here.

Listen to the audio here:

Download the MP3 audio here.

Machine-Generated Transcript

What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode.

Christopher Penn 0:02
This is In-Ear Insights, the Trust Insights podcast.

Do you want to use AI in your marketing but you’re not sure where to start? Take a class with Trust Insights and the marketing AI Institute.

The AI Academy offers more than 25 classes and certification courses to start you on your AI journey, including our intelligent attribution modeling for marketers certification.

One membership gets you access to all 25 classes, visit Trust insights.ai slash AI Academy to learn more and enroll today.

That’s Trust insights.ai slash AI Academy to enroll today.

Are you struggling to reach the right audiences? Trust Insights offers sponsorships in our newsletters, podcasts and media properties to help your brand be seen and heard by the right people, our media properties with almost 100,000 people every week from the In-Ear Insights podcast to the most timely and in the headlights newsletters.

Reach out to us today at Trust insights.ai slash contact to learn more, again, that’s Trust insights.ai slash contact.

In this week’s in ear insights, let’s talk about conversion tracking, specifically, view through versus click through conversion.

So these are two measurements, you’ll hear a lot especially when it comes to advertising.

And the big difference between them is that a click through conversion is within the space of a session, right? So you see an ad, you click on it, you go to the website, you convert that’s a click through conversion of view through conversion is you see an ad, and then you go off do something else.

And you come back later to that clients website.

And you convert.

And because of ubiquitous surveillance and tracking.

companies like Facebook and Google Ads know, hey, you did see the ad and Lady you converted.

So we’re going to treat that sort of as an assisted conversion, if you will.

And as a result, we’ll call that a view through conversion, you view the ad and eventually you convert it.

So Katie, you had some questions about this lovely tangle of of metrics and analytics.

But what’s what’s on your mind?

Katie Robbert 2:06
Well, I guess a couple of things.

And as you’re describing how a view through conversion works, the first one of many questions, the first question that comes up is, you know, is this a metric that we should be, you know, staking our businesses on staking our KPIs and our jobs, you know, on? And I guess the second question, and we can answer these in whatever order you think makes the most sense, is, isn’t a view through only as good as the tracking pixels that you have set up? So yes, Facebook can track you.

But if the company does not have a Facebook tracking pixels set up correctly, therefore, the company will never know that you saw the ad Facebook will know but they’re not going to willingly just tell you, right?

Christopher Penn 2:51
So if, yeah, we can go in reverse order.

It presumably, if you’re running Facebook ads, you should have set up a Facebook tracking pixel on your website if you hadn’t.

That’s,

Unknown Speaker 3:02
that’s a failure.

That’s why when you assume

Christopher Penn 3:08
and so yes, assuming that you’ve set up tracking pixels correctly, which by the way, if you haven’t, you’ve probably noticed your ads are spending an awful lot of money and you have no results.

That would be something to fix.

But on the first part about whether this is important.

So view through conversions are very important to advertising networks, right? It’s how Facebook and Google and all these different companies say hey, we’re valuable, keep paying us money, because we can say all these conversions were assisted by them.

The problem is abuse view through conversion benefits the advertising network, it doesn’t necessarily benefit you the marketer, because everybody can double dip.

Facebook has no idea what Google’s doing.

Google has no idea what Twitter’s doing.

Twitter has no idea what YouTube is doing.

And so everybody’s gonna say, Oh, yeah, you saw my ad.

And then the tracking pixel picked up that a conversion happened.

So one conversion, may have had three different touchpoints.

All three, Facebook, Google and Twitter in this example, will claim that one conversion as a view through conversion, even though we know from attribution analysis, you should be assigning fractional portions of that conversion to each one.

But because Facebook can only see Facebook stuff, and Google can only see Google stuff there, it’s going to claim the full value of conversion.

Katie Robbert 4:21
Well, and you know, it’s, that’s a helpful explanation, because I feel like I’ve been hearing the term view through conversion a lot.

And to me, my first thought, and of course, I had to do a little bit of research to figure out what it was it sounds more to me like an impression.

So it’s, somebody saw something so you know, Facebook, for example, can show you ads based on other videos that you’ve watched and so you know, I was telling you Christmas morning that one of the things that I will do is go through and see why am I seeing this ad because Facebook will give you some level of information, you know, a little bit of the targeting and then it’ll you’ll see things like, it’s looking for customers who have watched this video before or something like that.

And of course, my first thought is, well, I’ve never seen this video.

So okay.

And so to me, it’s a bit of a janky metric in the sense of it’s not a true conversion, you didn’t really interact with the thing, per se, like you didn’t click on it, or take an action or fill out a form.

And, you know, perhaps I’m old school in that way.

But to me a view through is more akin to a billboard, whereas a click through conversion is probably a more accurate way to measure.

But you know, tell me where I’m going wrong here.

Christopher Penn 5:40
It’s not just an impression, it is an impression that was served up to somebody who eventually did convert.

So there is there is that anchor that says, Yes, this person eventually did convert.

So Katie, if you’re out on, you know, on the web, and you see an ad for say, I don’t know.

But you know, Bed, Bath and Beyond.

And then three weeks later, you go on a Bed, Bath and God’s website, you buy something, Bed Bath and Beyond, as the advertiser will say, Ah, that view helped the conversion, and therefore that is a view through conversion.

So it is more than just an impression, because the impression would count whether or not you bought anything.

In this case, it’s a an impression linked in some way to a conversion.

Now, where this goes wrong for advertisers, for companies like ours that are advertising, is that that credit apportionment, that’s the heart and soul of attribution analysis.

Only the company and its tracking systems has the full view of the customer has the full view of where did that person come in? Did they come in through Facebook? Did they see a YouTube ad? Did they open an email? Did they come to a webinar, whatever, and eventually convert? And that’s why attribution modeling techniques like Markov chain modeling, for example, are so effective, because they can take all of those interactions, wait them appropriately, and then say, okay, yes, Facebook, you’re the traffic that you sent, or the view that you sent for that ad was valuable, but it wasn’t as valuable as you think it was, you that you’re claiming the full value, the conversion, and we’re saying you should really be claiming about, you know, a 10th of a conversion for that.

Now, where this gets challenging for marketers, is when it is purely a view through meaning there wasn’t an action taken.

So if you see the ad, Katie, it served up to you.

And you don’t click on it, you just scroll by it in your timeline.

But then you go and buy something later, only Facebook has that data about whether the impression was served or not.

And so, it is trickier in that case for a marketer to say, okay, the value of that Facebook impression would did lead to a conversion.

Again, this of strongly benefits the advertising network, Facebook, it does not benefit us as the marketer, but it is there is it’s not say is without value, because yeah, if you see the same ad 100 times as you’re swiping through Facebook, eventually you’re like, Oh, my God, I can’t stop seeing this ad like it won’t go away, you’ll at least remember the company even if you hate them.

Katie Robbert 8:09
Well, and I guess that’s my whole point slash question is, you know, it is sort of like an impression in that sense of you don’t have to take, you know, any sort of action for it to claim credit.

You know, I could, theoretically, I could send out like, $100,000 worth of ads for Trust Insights.

And, you know, put them across all the different networks and put up billboards, and never have anybody have to take an action with it, I just have to annoy them with this ad so that they see us.

And then, you know, six months down the line, somebody comes to us and says, Hey, I want to work with you.

Well, great, I’m gonna say that my ads were the thing that brought you win.

And really, they’re going to end the customer might say, Actually, it was because I saw Chris say something intelligent at a conference one time.

And so it’s like, I don’t know, it, just to me, it’s a poor way of measuring your ad campaigns.

Because to your point, Chris, these ad networks are so hungry to take credit so that you keep pumping money into them.

It just it.

It seems like a false metric to me, like, I understand what you’re saying, like it’s a valid metric.

But there are better metrics out there to be measuring your campaigns on and if you want to do purely an awareness campaign, then great view through impressions, whatever you want to call them is probably a good way to go.

But if you’re looking for someone to convert and take action, just to me a view through metric does not seem like the right way to go.

Christopher Penn 9:50
It’s half of a metric.

And that’s where ad networks really don’t want you to talk about this part, but we will because we’re not an ad company.

The second half of any kind of impression based technology, whether it’s view through conversions, brand, whatever, is that you have to establish whether or not you’ve actually got share of mind.

That means doing things like on your forums, asking how did you hear about us and having, you know, different campaigns, things listed there running surveys in your target market saying, Are you aware of Trust Insights? What is your intent to purchase from TrustInsights.ai in the next 90 days, and then of course, looking at things like branded organic search, if you run that hundred thousand dollar campaign, Katie, and we don’t see searches for Trust Insights go up at all that we know they just wasted a whole bunch of money right now we should have just got done, you know, you know, drugs and people in illegal professions.

The second half that measurement is so important of did this achieve our goal of getting into somebody’s head? And the only way we find that out? Is by those activity based metrics that indicate Yes, I’m searching for you and thinking about you.

I’ve heard of you.

And you will never ever ever see Facebook say yes, run a survey to verify that our impressions are actually having an impact on the consumer.

Because what you’ll find out more often than not, is people like what I think about a Katy beside clients, and ourselves.

What’s the last digital advertiser you remember?

Katie Robbert 11:30
I couldn’t tell you, it was probably something Amazon or pet related, given like the last few things that I may have done.

But even then I couldn’t tell you outright that that was like, if I had to bet my life on it, I would not

Christopher Penn 11:47
write exactly the last Adam any kind that ever call by name and the brand was Ryan Reynolds doing stuff for his MIT mobile stuff.

And it wasn’t an ad, I went to his YouTube channel and watch the ads that he has on his YouTube channel.

Because damn funny.

They’re worth watching.

Even you know, and they’re unpaid impressions, because I, you know, he’s just, he’s just enormously funny and very talented to come at at relentlessly Schilling in hilarious ways, the products and services he makes.

And so I know for a fact that companies are spending exorbitant sums of money to have me scroll through their stuff on Facebook, on Instagram on Google on YouTube.

And I can’t remember any of them, not a single one.

So if we if they were doing an unaided recall survey, which by the way, is sort of the gold standard for this anything impression based.

They’ve all failed, they’ve all wasted their money, because I can’t remember any of them.

If you asked me, hey, was the last brand of gaming hardware that because I watch a lot of gaming stuff with let’s let’s brand a gaming hardware.

I like I don’t remember.

I mean, I know the brands that I like, but I don’t think I saw ads for them.

And so, for marketers, you know, this kind of flips nicely to something that we were talking about earlier as well, I was just Oh, he was asking us like, you know, how do you What advice do you give marketers who are trying to do less tracking and tracing and surveillance and stuff, but the reality is, the tracking is really important.

But so is understanding what’s actually working.

So view through conversions are not a great metric without being paired with the other half, which is the the actively asking people questions.

So you don’t need surveillance and tracking for that you can just ask people, in fact, should always be asking people, have you heard of us? How did you hear about us? What made you come in today?

Katie Robbert 13:34
Well, and let me ask you this question.

This is more of the technical side of things.

So how long does someone have to be viewing an ad in their social media stream in order for it to be counted? As a view through like, Is it one of those like, I’m scrolling, I’m scrolling and scrolling.

And somehow it managed to fly by while I was doing scrolling? Or do I actually have to sit and watch this video? It doesn’t have to be a video.

You know it? Can it just be a static image that I happen to like pass by for two seconds? Yeah, so

Christopher Penn 14:08
it depends on the network.

Facebook, for example, if the ad is rendered, meaning it shows up at all, it counts as an impression.

If it’s a video, you have to watch it for three seconds for it to count.

So those are sort of the two things.

So yeah.

If you’re scrolling and an ad, physically renders on your device, it counts as an impression.

Even if you went by so fast.

You have no idea what it was the fact that Facebook software had to pull the ad from the ad server and serve it up to you, then tells advertisers Yes, we served you that that we serve that impression.

And if you did accidentally happen to convert later on.

Facebook could say yes, we did our job.

You get the view.

We got the view through compression, even though you’re like I went by so fast.

I have no idea what it was.

Katie Robbert 14:52
See, and you’re I guess you’re just further reiterating my concern about view through being A metric, I’m not going to say it’s not a valid metric.

But it being a metric that you should, you know, base your success of your campaign around.

Everything you’re describing to me just continues to reiterate that a view through metric is problematic.

Now, so let’s say, you know, we had a client who said, I’m gonna stake my whole reputation on view through campaigns.

And we said, that’s a terrible idea, do this instead, what would be the this instead?

Christopher Penn 15:32
That would say the same thing we say often, which is for every 25, for every dollar, you’re going to spend on ads, bucket, 25 cents of it for measurement.

And so we can run those measurement campaigns to go with it and say, Are you having an impression? Did you see let’s make up a client, a coffee shop? Well, you know, did you see a Oh, Katie’s coffee shop? advertising in the last seven days? In fact, Twitter does this a lot.

You’ll see this in your Twitter feed.

If you’re looking in just your home feed, they’ll have these little brand series a pop up that say, Have you seen an ad for X ray Katie’s coffee shop in the last seven days? Yes or no? And it’s a you there’s short surveys, like one or two questions.

And everyone’s like, I have no idea if I have or not.

But at least that’s an advertiser, which is doing the right thing of saying, okay, we’re running the ads, we’re also running the verification metric say, Mm hmm.

People at least remember seeing our ads, even though yes or no, at, you know, with, I forget what the last count was, but something insane, like, people see between five and 10,000 advertising impressions a day.

You know, from simple stuff, like a one t shirt, am I wearing to you know, actual ads being served on devices? Of course, people aren’t going to remember unless something they already have an affinity with.

Katie Robbert 16:50
So well, who’s running those? Did you see those ad? Like, is that something that me the owner of Katie’s coffee shop can ask Facebook to run or Twitter to run on my behalf? To see if people are not only seeing my ads, but remembering or is that something that the ad network runs and that I the business owner, or whoever never get the data?

Christopher Penn 17:14
It depends a twitter at least on Twitter.

Twitter’s are run by Twitter itself, I think to calibrate their advertising.

To my knowledge, I don’t believe any of these networks share that information unless you run a campaign specifically designed for that, you know, Google consumer surveys, for example, would be an example of where you have to commission a an additional campaign in order to measure the effectiveness your other campaigns.

Katie Robbert 17:36
But can you run a Google consumer survey within a social network?

Christopher Penn 17:41
No, you have to do it within the social networks themselves.

Katie Robbert 17:45
So like, so there’s also another disconnect there, because you may not be reaching the same audience.

You know, people who are searching for your brand might not be the same people who are seeing the ads on their Facebook feed.

And you know, it just again, it’s concerning when I hear that view through, or similar type metrics are the things that people are staking their reputation on.

And so I guess the whole point of this conversation is to deconstruct what goes into a view through metric and why it’s, yes, it is a metric.

But it may not be the best metric, if your goal is conversions, because there’s flaws with it.

There’s flaws with any metric, and there’s flaws with setups, but this one to me, just sort of struck a chord of, ah, let’s not use that as the basis for your success in this campaign.

Christopher Penn 18:39
I want to throw an additional wrench into it.

Sure, having on the ad blocking technology, some ad blockers will intercept at, at the at the server level, meaning that the ad blocker will prevent the browser from even asking for the code for an ad say, Nope, we’re not even going to, we’re going to block that request from going outbound.

So the never reaches the advertiser.

Other ad blockers, let all the data come in from the source and then just cut the ads out.

It’s like imagine get a newspaper and saying just give us a pair of scissors just cutting out the actual ads.

And we’ve been throwing them away.

In that second scenario, the advertiser has paid for the ad.

And then that because the network says I rendered it, even though the consumer never ever saw it, because the ad blocking code rewrote the web page that the consumer saw.

And so at that second category of ad blockers can actually be even more problematic, because the network is saying, Yeah, we serve the ad, the customer got a view through conversions valid and the consumers like and see any ads.

And you don’t know and there’s no way to tell, except that you know that you’re paying, you know, through the nose for no cost per impression advertising.

And nobody’s seeing even those networks are saying that they have.

So one of the things we tell people when it comes to analytics and metrics is do not use any of the impression basically You cannot rely on the impressions being served.

Focus on traffic, focus on conversions.

You know, even if nobody converts on the site in that session, if they at least got to your site, then Markov chain modeling can detect that traffic the footprints in your site, and then eventually attribute that later on.

It’s like, yeah, you can’t see what’s happening outside your coffee shop.

But at least once somebody sets foot inside the door of your shop, you can watch them and see what they do.

And yeah, maybe they walk over to the scones display casing Oh, I don’t know about garlic and kale scones, and they walk out at least, you know, at least you know that you got them in the door, and you can figure out where they came from.

So that would be our counsel for any kind of impression measurement is instead focus on the traffic and focus on where’s the traffic coming from even if they didn’t convert, because eventually, if you’re doing attribution analysis with modern methods like Markov chain Monte, it will be taken into account later on.

Katie Robbert 20:59
So it sounds like the takeaway is, number one, have a plan because you want to know what the heck the goal of your campaign is, figure out what kinds of metrics you want to be trusting, and how you want to put your analysis together, make sure that your infrastructure is set up in such a way that you can track people across these different networks that includes those tracking pixels, that you set up through Tag Manager, or whatever your system is, be aware of ad blocking system.

So basically, think of like a giant goalie net, catching all of those ads, and they never get to you in the stands.

It’s that big net in front of you and like, yes, somebody’s seeing them.

But it’s just a bot and the bot doesn’t care.

And so it sounds like there’s a lot that goes into measuring campaigns and just picking a metric, because it’s what the ad network tells you you should be looking at is probably not the most accurate way to be measuring because it’s not in your best interest.

It’s in the ad networks best interest.

And if the ad network is saying, yes, you’re doing great, give me more money.

That’s probably something fishy about it.

Christopher Penn 22:05
Exactly.

And to circle back on Oz’s question about the advice we would give to marketers, this, that also answers a lot of these questions about advertising effectiveness.

It’s really straightforward.

Talk to your customers, talk to them as real people, person to person pick up the phone shoot, to text them, whatever, and say, Hey, Tim, what you’re looking for? What brought you in today? How can you How can we better serve you, you’ll be amazed at the quality of information, you can get people to volunteer, as long as you know, you’re not a jerk about it.

And you will get a really good sense of what’s actually working and what’s not.

So please just talk to people don’t rely on technology to do a human’s job.

And I know that sounds absolutely crazy having me say that, but it’s

Katie Robbert 22:48
well, and you know, if you’ve seen any versions of our customer journey or customer experience, you know that evangelism is that last phase, because then you have that loyalty of the customer, they’re telling you exactly what they want.

And then they’re also helping you do the selling because someone is, you know, a consumer is likely to trust someone else who is just like them in their experience, versus a company saying you must buy my thing, you need this thing.

And so definitely be you know, use your customers in a smart way.

Ask them what they want, have them do word of mouth marketing for you, you know, pumping lots and lots of money into an ad network is okay.

It depends on what your goal is.

But nothing replaces actually talking to your customers and finding out what they want.

Christopher Penn 23:39
And speaking of talking to customers, if you’d like to talk to other folks in the analytics space, about the challenges you’re having with measurement and talking to customers, pop it over to our free slack group go to Trust insights.ai slash analytics for marketers and join over 1400 other marketers about talking about analytics and and all things related to measurement.

Thanks for listening to today’s show.

If you’ve got questions about it, pop on over to Trust insights.ai slash ti podcast where you can find this episode many others and find all your favorite ways to subscribe.

Thanks for watching and listening.

We’ll talk to you soon take care.

One helps solving your company’s data analytics and digital marketing problems.

This is Trust insights.ai today and let us know how we can help you


Need help with your marketing data and analytics?

You might also enjoy:

Get unique data, analysis, and perspectives on analytics, insights, machine learning, marketing, and AI in the weekly Trust Insights newsletter, Data in the Headlights. Subscribe now for free; new issues every Wednesday!

Click here to subscribe now »

Want to learn more about data, analytics, and insights? Subscribe to In-Ear Insights, the Trust Insights podcast, with new 10-minute or less episodes every week.

Leave a Reply

Your email address will not be published. Required fields are marked *

Pin It on Pinterest

Share This