Most marketers know the
basics of attribution models, but did you know there are seven of them?
Let’s explore each of them on this episode of Closing Time.
Thanks for tuning in to Closing Time, the show for go to Market Leaders.
I’m Val Riley, head of marketing for Unbounce, Insightly and LeadsRx.
And today I’m joined by my colleague. Larry Todd,
who is the leader of the LeadsRx team here at Unbounce.
Larry, welcome back to the show.
Hey. Thanks for having me, Val.
So we had you on once to talk about marketing attribution.
And then I did went down
some rabbit holes and found out that there are seven kinds of attribution.
Lots of marketing automation platforms
come with attribution models that are first touch or last touch.
I’ve got 20 years of experience under my belt.
Even I didn’t know that there were seven models.
Are there some that are more recent editions?
Yeah.
You know, LeadsRx has operated with seven different attribution models
for for quite some time and a lot of them are rules based models.
Some more widely known than others.
But we do have a proprietary algorithmic model which is more data driven.
And that is a recent edition.
Okay, good.
Well, let’s educate those marketers out there.
Just like I got educated and start with the first two,
because they are the most common and probably the ones that most marketers
have heard of.
First touch and last touch.
So what’s going on there?
Yeah.
You know, pretty simple. Self-explanatory.
The first touch, attribution model gives all the credit of a conversion
to the very first touchpoint that a user engages with.
And so it’s really good
at calling out top of the funnel tactics that grab somebodies attention.
And as long as you maintain visibility through the customer journey,
you can understand, you know, where they actually ended up
converting, downstream, but you’re still giving all of the credit
to that very first touch point that grabbed their attention
and really got them to know your brand, right.
Last touches the inverse.
So what drove the user or consumer
to to take one, you know, conversion final action.
There could have been a number of engagements top and middle of the funnel,
but you’re focused on what actually drove that final action to either
get them to purchase your product, sign up for your demo request,
a, you know, a content piece or whatever that might be.
Yeah.
So that’s a big argument within the marketing team.
Usually, like the paid acquisition folks want that first touch model to get credit,
and then your content marketing people or your email marketing people are like,
no, it’s the last touch. So.
And you can’t please everybody, but let’s keep going
because there are models that do, hopefully please everybody.
The next one you talk about is the closer touch attribution.
What is that and what are the use cases for it?
This is probably the least known.
And, consequently, actually probably the the least used,
but it’s it’s used primarily for heavy DR, heavy direct response.
And so if you’re trying to measure what drove somebody to take a final action,
but more importantly, what drove somebody to take a final action immediately.
So closer touch gives is it.
It’s like last touch.
It gives credit to the last touch.
But only if a conversion happens within a predefined period of time.
You set by by the organization typically two minutes, handful of minutes.
And it’s really like take action, buy our product, receive 20% off.
Right. Take action.
Now you get a buy one get one free.
And you’re really just trying to see how many
consumers took that action immediately.
As, as you had intended it and then capturing those conversions.
So it’s it’s typically a view that marketers use,
but not the view that they use to allocate marketing budget, if that makes sense.
Absolutely. Yeah.
Okay, let’s move on.
We’re up to number four, and this is any touch.
So what is any touch?
And what are some of the pitfalls?
Maybe associated with using that model?
Everybody gets a trophy.
It’s a participation credit. Right.
If you touch the consumer and they end up converting, you get full credit.
So you know you play that out and you start to see some of the issues
associated with that. Right?
Somebody touches Facebook, somebody touches Google.
They convert both ad servers are saying, hey, we brought your customer.
We brought you revenue. Everybody’s claiming credit.
There’s double and triple and quadruple counting. Right.
But it is a good view to just say, hey, we’re going to give credit
to every single, ad or tactic or channel that drove a result.
Right.
it gives you full visibility to, to track,
engagement across every single tactic on the way to a conversion.
So that’s the one that keeps
every single person on the marketing team happy is what I’m hearing.
Yeah, right.
Everybody gets credit, right?
Everybody gets, a trophy.
So, yeah, you just, you start to get, some challenges when,
when you use any touch, you’re going to end up with conversion
results that far exceed what actually happened in real life.
As long as you’re okay with that and know what the model is intended for,
it works out great. Yeah.
It keeps piece on the marketing team, too.
Exactly.
Okay, so the next three are all multi-touch models,
but they’re separated into linear, algorithmic, and weighted. So
let’s talk about each of those.
Yeah, a couple of rules based models there, and then one data driven model.
And when I say rules based, there’s just a predefined set of rules
that the that the attribution model is going to leverage when assigning credit.
The linear multi-touch is the most fair, right.
It’s going to take a conversion and split up the credit
equally amongst every single touchpoint that’s involved.
You have five touch that drove a conversion.
Each one is going to get one fifth or 20% credit,
irrespective of if that was in the beginning of the customer journey,
the middle of the end, how much time elapsed through the customer journey?
How short or long the customer journey was?
If you played a role, you share credit with everybody else.
The good thing about that is if you have 100 conversions,
we’re still going to display 100 conversions.
It’s just those 100
conversions will be credited across all the touchpoints that played a role.
the weighted multi-touch attribution model is another rules based model.
And what we do is we give 40% credit
to the first 40%, credit to the last,
and then the remaining 20% is split between all of the interior touch points.
And so it’s a way to give credit to every touch point involved.
But what you’re doing is you’re calling out the importance
of the touch point that grabbed that customer’s attention,
got them to know who you are,
got them to care, got them to take some sort of digital action.
And then the final, which is got them to actually make
a purchase decision, swipe a credit card, you know, sign up
for your webinar or whatever it is that you’re trying to measure.
But we are not just throwing all of those interior touch points
to the wayside because they get they did keep the line taut.
If it were right,
they kept, your brand top of mind, kept their attention and so on.
And then lastly is the algorithmic model,
The algorithmic model is a data driven model,
that learns the success and failure of your touch points over time.
And so the more impactful
those marketing touch points become, the more credit they get.
So that’s a, machine learning model that updates every single night.
But the most up to date information on all the marketing campaigns that you have.
So I’m in b2b SaaS.
So for me, it feels like the weighted model makes the most sense
because I think, okay,. I want to know what the first touch was,
and I want to know what the last touch was.
Those should get the most credit and the stuff in the middle.
But, the algorithmic model seems like
it’s something that more marketers should be trying.
Is that something that’s newer to the market?
Not not necessarily newer to the market.
I think, you know, data driven, models have existed, you know, for a while
now, Google, pioneered a data driven model
of their own, at some point.
But I think that, linear multi-touch
is the most, widely used,
especially with organizations that are new to attribution.
It’s easy to understand.
It’s easy to explain to both the internal and external stakeholders.
Non marketers can understand it as well, fairly simply.
It doesn’t, create any internal rifts between teams because everybody’s
getting some portion of credit but an equal portion of credit.
And the algorithmic model will start off as a linear
model until it starts to learn, right, your programs.
And so when you start to
have a little time under your belt, we see, transitions from,
linear multi-touch to algorithmic pretty consistently.
Got it.
Larry, do you recommend specific models based on business type?
Like, for instance,. B2B customer versus a B2C customer
might lean towards one model of attribution versus another.
think that in all instances, multi-touch
will outweigh, from a benefits perspective, a single touch.
I think, the way consumers are engaging brands, whether it’s direct
to consumer, e-commerce,. B2B has become more complex.
And so pigeonholing first touch and last touch is becoming more and more outdated.
I feel, for B2B SaaS, your longer tail customer journey
as you’re more researched, the sorts of engagements, right?
Higher dollar value purchases, multiple
stakeholders, potentially linear, multi-touch or algorithmic,
start to shine in those instances with actually direct
to consumer and e-commerce, we, we see weighted shine a little bit more,
which is, hey, what grabbed your attention to care about my,
you know, pair of shoes or my handbag or my piece of jewelry or whatnot?
I’m going to think about it.. I’m going to deliberate.
I’m going to maybe slightly engage with pieces of content or ads on the way.
But then when it’s time for me to actually buy
what drove me back into the store
front, as it were, the digital store front to make that purchase.
And so weighted plays really well in that instance.
I do find it funny as a consumer, if I’m.
If I’m considering buying something, a piece of clothing
or a pair of shoes or some kind of makeup that I see, and I.
I might see it on Instagram.. I might see a TikTok about it.
And because I’m a marketer, I think, okay, before I click the purchase,
which influencer or channel do I want the credit for this purchase to go to?
So when you know how the sausage is made a little bit,
you have a little bit more insight into, you know, who’s
going to actually get the credit for the purchase,
but feels like with some of these models, it would be evenly distributed.
Right.
Or, you know, like a lot of consumers,
you open up another tab and go in via Google search and then,
that’s the challenge that marketers have been dealing with a long time,
which is paid search, organic search and direct
traffic, getting credit for everything that, that marketing is bringing in.
If you could only see my spreadsheets,
you would see that that is the biggest bucket.
And oh, why don’t we just do more of that?
And people don’t understand that that that Google bucket
or that direct bucket is being fed
by all the multitude of things we’re doing as a marketing team.
So you’re speaking to my pain point.
And I think I think that’s
the takeaway message with multi-touch attribution is let’s try to increase
visibility to the entire customer journey.
We know marketers are working
tirelessly to bring folks, you know, to the fore to engage.
So let’s bring as much visibility and transparency of that whole customer
journey as possible so you can do more of what’s working well, right?
And how do you talk to marketers about, considering an external marketing
attribution solution versus the Intel that marketers
are getting from vendors like Google and Meta?
Yeah, it’s it’s a pretty easy sort of talk track.
And I think that by the time
that we’re speaking with marketers, a lot of times
they have a sense of the challenges that they’re dealing with.
And those challenges are born of a bunch of different bias
data sources being their, their source of truth for a long time.
So it’s, you know, easy to understand why
Google and Facebook and all of the other ad serving platforms,
would like to put their best foot forward with their metrics.
They have some skin in the game.
They’d like to get more ad dollars from you. Right.
And so, looking into a third party, impartial multi-touch
attribution solution, is is the way to go for a lot of folks
that are looking for a clear,
unbiased perspective into what’s working well and what’s not working well.
So, so a third party, partial multi-touch attribution is just going to be
the third party referee, the measurement partner tell you how things are going,
what’s working well, what’s not, without having any skin in the game at all.
And I think the big question is you told us about these seven models.
Do marketers have to choose, or can they find a vendor that offers
all seven and they can potentially
analyze all seven and come up with, you know, what they feel
like is a bigger picture for their marketing dollars?
Yeah. I mean, this is where my bias comes into play.
Knowing LeadsRx very well.
I know that we have, you know, a multitude of of different models at your disposal
that you can toggle back and forth between so you don’t have to choose,
typically from day one, what is going to be your best measurement,
solution and sort of continue on with that.
You have the ability to look at the data, look at the
different models, see what impact, those models have on your brand.
The use cases of attribution are pretty wide and varied.
We work a lot directly with brands, but then we also work a lot with agencies,
and agencies
support a multitude of different customers under their agency umbrella.
So a different model could work for a different kind of business.
Under their under their purview. Right.
So it all just, depends on what model works best for your business.
How you can more clearly, most clearly paint
the picture for your internal and external stakeholders.
Right. Place a lot into it as well.
So if you’re looking to aggregate a whole bunch of data for the VP or CMO, right?
Using a a model that is going to give credit in a clear,
defined way across all of your tactics is a really good way to go.
If I am somebody focused primarily on paid search,
I may want to have a different view that gives me
a little bit more of a refined view into my particular channel.
Right.
And what is driving last touch conversions from that perspective.
But you don’t have to choose one or the other from the outset.
You have all of them available to you, and you can take a look
into each one of them, as you go and refine your strategy.
I imagine you’ve probably seen in the course of your career with LeadsRx,
a few aha moments where folks were perhaps only using first touch or last touch,
and then all of a sudden, were exposed to data and thought, wow,
you know, these programs, these channels that I’ve been investing in that I haven’t
thought there is a significant amount of ROI,
but suddenly using a linear multitouch, model, all of a sudden
they can say, oh, well, we actually are really benefiting from these channels.
Like those aha moments must be, pretty significant.
Yeah, it’s it’s interesting because there’s a couple
of different views that marketers tend
to come into the conversation which with which is,
one being, hey, I would really like some increased visibility.
I don’t have a sense, really, of
of what’s working within within my marketing mix.
All I know
is that the bottom of the funnel are this are the same three kind of channels.
And I’d really just like to learn more about what else is going on.
The other is, marketers that have a predefined notion
of what they already feel is going on, and they’re looking for some validation.
Right.
And they’re just looking for an attribution to prove out what they
what they feel they already know.
And so the aha moments, come a lot on the,
on the side of the marketer who comes in with just more questions and answers,
doesn’t have a lot of visibility into top and middle of the funnel
and really doesn’t have any sense of, of kind of what’s going on.
Prior to and prior to the last touch
and, some of the aha moments on the other side
come from those predefined notions being proven wrong.
Right?
So, it is a couple of different, different use cases, which is seek
to understand or seek to validate something that I already think I already.
Yeah, that I already think I know,
but moments on on both sides of the coin there.
Awesome.
I feel like we could talk for hours
about this, Larry, but that’s probably a good place to stop.
Where can folks learn more about, to connect
with you and learn more about marketing attribution?
Yeah. You lead.com.
There’s a ton of, resources there.
Blog content.
You can go, hit us up there and request a demo of our product.
We’d love to learn
more about the challenges your you’re looking to solve.
And,
and give you a tour of the product and talk about the value
that we could potentially bring to your organization.
Larry, thanks so much for joining us.
My pleasure.. Thanks for having me again. Awesome.
And thanks so much to all of you for joining us.
Remember, you want to like LeadsRx on LinkedIn
to keep up with all the latest on marketing attribution.
And if you want to get this episode
delivered right to your inbox, click the link in the show notes.
We will see you next week.
