Hi everyone, welcome back to the Stacking Growth Podcast. My name is Sydney Waterfall. I'm a VP of demand here at Refine Labs and a co-host. And today I'm joined by another co-host, Judy Sheriff.
Thank you, Sydney. I'm Judy Sheriff. I'm a general manager here at Refine Labs and very excited because Sydney and I are talking about one of our favorite topics, attribution. The biggest hard word of the past decade or like really our entire career since you and I got into marketing.
I feel like attribution has always been just like hot topic. And something I was thinking about that I think is really interesting is I feel like over the course of like let's say the last 15 years, kind of come full circle with attribution. I think like pre technology or like pre heavy digital marketing, we're really focused on how can we get insights on how marketing efforts are working with like advertisements, radio, billboards, it was all qualitative. It was like surveys, customer insights, focus groups, really nothing that was super concrete and data focused back in the day.
Then I think right when you and I were getting out of college, tracking exploded with like pixels, cookies, post backs, all this fun stuff where suddenly marketers were like, yes, we can track everything and then like explode into attribution software. And I think you and I like very closely have seen the last few years, the flaws with that, how much we've been overly relying on attribution software. And like really how much we need to bring that contextual layer back in with some of the qualitative and set of quantitative. So yeah, I guess that brings us to our topic of hybrid attribution today.
We're like, let's do both. Let's use qualitative and quantitative and how do we get the best attribution across the board. Yes, I think gone are the days of being able to track every single touch point, being obsessed with tracking every single touch point, mapping every touch point to buy your journeys and revenue. Like you were never able to do that in from the beginning.
And so let's get over that. I think for the next topic that we can dig into is kind of going through the flaws of current attribution. And then we'll kind of dig into the concept of hybrid attribution, how we think about that, how we use that, how we recommend people to use that. So I'll kick it off and we'll probably just go back and forth here.
But yeah, to your point, it's come full circle. So from the past 10 years, I would say with attribution software and tech coming in, it's really been all designed around capture demand channels, direct response, even capturing conversions on non response channels, awareness channels, everything's been about conversion, conversion, conversion, which obviously you would be able to track. But now that the market is shifting, buyers have changed my own buyer behavior, my own like natural behavior has changed. There's certain things that you cannot track, like podcasts, Slack messages, communities, live events, things like that.
So I think people just need to acknowledge that and understand that like, yes, what we're using the traditional software for attribution or whatever cannot do these things, 100%. I'm like totally guilty of this when I was in past roles. I have a very type of personality. I loved the exact matching attribution.
If I could get it, like, oh, this deal came from this specific Salesforce campaign, perfect picture, easy ROI story, like so nice as a marketer to be able to have that. But you realize the longer that you're doing that and the more you're reporting, just how much you're not seeing there, how much of the full picture is missed. And then it really poorly informs future marketing decisions. And I definitely made calls in the past and not pursue certain areas because it wasn't attributable.
Or we don't know if this campaign is working because we can't see directly it's in a Salesforce campaign associated with it. Let's not run it at all. And it's not smart, but we definitely missed opportunities because of these decisions where we're relying too heavily on the attribution software. And like you said, it lends itself really nicely to capturing existing demand, direct response, anything where you can put someone into a clearly categorized Salesforce campaign and then the software can pick up like order of campaign membership.
But we know dark social podcasts, sharing LinkedIn content, even just like you and I going back and forth pre this podcast and things we were talking about, like that's mine sharing right there that never is going to get picked up on attribution software. Yeah, definitely incentivizes marketing teams to only focus on things that they can capture and things that they can do. Like even thinking creatively, like you could use direct mail, but only if it's like in a Salesforce campaign and we know who it got delivered, or you could use direct mail in a really cool creative way that is less trackable, that probably would make more of an impact, right? So it's even using quote unquote more traditional direct response channels.
It's rethinking how you use them as well, like how you deploy them. And I think the easy way to divide that in like my mind and probably other marketers minds is like, okay, we've got this to capture. We've got this to create because it's easier to kind of separate the channels that you would do and the tactics and how you would distribute different things if you can kind of separate them and categorize them that way, which just kind of leads us into like this whole concept of like hybrid attribution, right? You definitely use what you can like direct response, like of course, like Google ads, we're going to be tracking that keyword all the way down to close one.
And we're going to be reporting on all of that like by keyword by ad group. Of course, we want that like we're going to keep doing that. But curious on your point of view there of working with both of those. Yeah, I agree.
So I think like when we talk about hybrid attribution, what we're really mentioning here is instead of just relying on the software, we want to ask people, how did you hear about us? How did you hear about the product, the service, get that qualitative feedback, definitely recommending like open text, free text or writing in as much info as that person wants to deliver. This actually naturally went against my instincts too, I'm very much like clean reporting. I want everything to be in clear bucketed categories, but you can uncover so many layers and depth that you would not be able to surface if it was a pick list that they were choosing from.
So I think like what I've seen work best is yes, use this the attribution software, attribute what you can, no, it's not going to tell the whole story. And then layer on the self-reported attribution data. So you can see for people that are coming in the attribution software says direct and organic, we need to layer deeper. What does that really mean?
We know direct traffic is just like that, is everybody that can't fall into any other category, but we know there's more to it. And so I think those particular channels are where we really uncover the most interesting insights. So you can see the attribution software might say that you know, 40% of the deals were coming in from direct traffic. But if we layer in, how did you hear about us, we see things popping up like, saw you on LinkedIn referral from the specific person's name, read about you in a Forbes article, like whatever it may be, we get that extra level of insight.
And then when we're able to report, we show here's the comparison. Here's what we're seeing from a software standpoint. This is what Salesforce is telling us. And then here's another level of marketing insights to inform so we can make really smart budget decisions that are combining both sources of data.
Yeah, I think there are two kind of separate categories, right? So you, your self-reported attribution, which is the how to hear about a skill, that is going to give you insights into what's working and what's making an impact for your create demand strategies and create demand tactics, where the capture demand, which is your more traditional, is going to say, okay, how did those people actually come and convert into high-intent conversion, which is also important to track and monitor as well. So that's kind of holistically how we think about it. I think one really important thing to note is you want to make sure that you have that qualitative field.
You want to make sure that you have another field that kind of buckets those things to make it easier to report. And, you know, shout out, like, there's a whole deep dive, there's literally two episodes, like a couple episodes back on this podcast that go really, really deep into self-reported attributions. If you want to learn more, you can check that out. But the reporting that you're going to get is you need to make sure that you have both of those fields, the qualitative field, which is the how to hear about us.
And then we normally would say, like self-reported category. You want both of those fields mapped all the way to the opportunity. So then you can report on create demand attribution and capture demand attribution and you all the way to revenue. And you get a fuller story with using both rather than using one.
Yeah, 100%. Nice story there. I think we could probably talk for like another two hours on the logistics of setting it up in the various automation platforms. But I know we go deeper into that in other episodes.
I think the most important takeaway on hybrid attribution is that it really isn't a single solution answer. We need to use all of our available data points, both quantitative and qualitative, and continually be digging deeper to get that full story of where people are hearing about us, where people converting, and how is that turning it around? Definitely. I mean, I think the how you use it is really the game changer.
So I mean, I can just speak to how some of our clients have been using it and how we've seen it be really effective. Obviously, we talk about internally how we use this data. All of our marketing is on create demand. So it's no shocker that internally our data skews heavily towards when we get the self-recorded fields.
It's like 80% of our pipeline is like, create demand podcast, LinkedIn, word of mouth. But that's going to look different for companies that probably don't do as much marketing under the create demand strategies that Refine Lobs does. However, you still need it, especially as you're transitioning from lead gen to demand gen. It's like 100% crucial to have this to prove out early on that the channels you're investing in that you no longer have your KPIs that you're normally looking at are actually working.
So we can chat through how clients use it, what we've seen, different things like that. I think that would be probably pretty helpful for people of how to actually use this. Yeah. I think where I've seen people getting the most value out of it is really understanding which social channels have been working since that seems to come in pretty commonly.
People remember what platform they saw an add on and we'll recall that there. So kind of the social channels. But on top of that, I've seen clients uncover new areas to focus on that they didn't know about until they implemented this field. Orbs is a specific example that comes to mind where a client had been picked up on Forbes and it was like organic press and like they were getting a ton of people that had seen them on this article right about it was coming in their self-reported attribution.
And they didn't actually have any like effort put towards that. So it was just a way to really figure out their future strategy around PR and speaker engagements, getting some of that leadership picked up more intentionally. So I think it's maybe undervalued as an area to pick up potentially new marketing channels. I've seen similar things around like underinvestment and review sites and then those coming up.
The cap pair is the G2s of the world that people doing the research there and that not showing up in attribution software as well. Yeah, it doesn't get captured in the river or they look and then three days later they just come in direct to the website that they really heard about you from review or something. A couple other examples is to like also prove out some experiments. Right.
So we have two clients that with one we ran a Reddit experiment on and then a Twitter kind of experiment and then another where we're kind of actually pitching podcast advertising and dabbling into that, right? Like the attribution on that is garbage. So how we set this up is we want to do like one or two things at a time and we have small investment. We want to see certain things and so for us we had a small investment in Reddit.
It's like 5,000 a month relative to their budget. That was a small amount for them. Every budget is different. But like within the first three weeks we saw four self-reported attributions come through that were the right titles, the right type of companies.
So it's like validating their persona is there and we're reaching them on a new platform. So that's number one huge. Number two, we were actually pretty surprised that it came back quickly within like three to four weeks. We didn't really know what to expect and we were kind of just monitoring.
Those conversions came through. They didn't get attributed to Reddit. Reddit even in platform didn't show even though we had the pixel implemented and everything. But with the privacy laws it didn't show that there was any conversions.
And so that gives us more confidence to keep investing in there and seeing if like, okay, does this keep happening? Like what happens if we expand our targeting? What happens if we push more budget? What do we see there?
And then it also allows us to say like, great, these are awesome ICP fits demos that are coming in. But how do they convert? Are they converting to hero pipeline or not? Right?
Because at the end of the day, if they come in and they don't convert, we need a decent couple more than three examples. But we need a small sample size and we can make some really good strategic recommendations for anybody. You brought up just a good point around all the issues we've been facing with tracking over the past few years, iOS 14, privacy issues, like all of those contribute to our need to really layer in the hybrid attribution because we're losing more and more visibility. And I foresee that that's just going to continue with stricter privacy, especially around some of the GDPR things that have continued to evolve in Europe and depending on where you're based, I think, or European clients have found a lot of value from implementing this.
Yeah, 100%. The kind of last example I'll dive into is like, obviously from a channel perspective and like a marketing perspective, it's super useful, even sales or looking at partners, things like that. But how CMOs and more like executives have been using this data? So for before, we would deploy our strategy, we see, you know, organic, indirect, you know, sources going up and the revenue going up, right?
You can correlate that with like advertising spend and you can show the correlation, right? But people really don't like correlation. No, they weren't. This demo specifically came from this Facebook ad at this point in time.
Yes, they want everything attributed. So, you know, the CMOs definitely like, this is part of the strategy. This is exactly what's going to happen. What we said was going to happen happened.
This is why well, now we're actually able to go back and show them like of the percentage that's getting attributed to organic and direct. They're actually coming from the investments that we're doing through our paid advertising and also their content engine. So it's also not just around paid advertising, which I think some people get caught up on, but they also have a huge event strategy and they have a huge content engine that like fuels their create demand and fuels a lot of their other content. And now like that's getting a lot more credit as well.
And she's able to go to the board and go to like her peer CFO and the CEO and say, look, here's the full picture. And here's what the data shows. Here's what this data point shows. Like, this is how you have confidence in me managing this budget.
And now I can have a fighting chance of getting more budget to kind of fuel on the fire for that. Yeah, it quips leaders with so much more actionable insights to go off of to justify budget requests. So let's talk about some of the, I think you mentioned privacy. I mean, fun topic.
Yeah, I think it's just it's going to get harder and harder. I think, you know, the sooner you can, like, what's the downside of like implementing this? It's like the sooner you can implement this, you get another set of insights and it's getting harder and harder for marketers. So like, why would we not?
I know people will make the argument like, we don't want to add another form to our field, adds more friction to the buyer journey. Like with two fields is all we want on our submission form. And I hear you, there's been a lot of studies around that. But we've actually done some tests with clients directly on conversion rates on their form with and without like kind of pre and post.
And we did not see that it significantly hurt conversion rates in any way. There was no significant data from the test that we ran. The data we did collect was a lot deeper insights into organic and direct channels into areas we weren't previously measuring. But we did not find that people were less likely to fill out a form because we asked for this question.
I think there could be circumstances where forms can be done really poorly. Maybe there's a lot of other bad fields that are already really long and there could be other areas to improve the form. But if you're just adding this in as one additional data point, the test we ran don't support that it's going to hurt your conversion rate. Yeah.
I mean, the best case scenario is your replace a field. Like that's unnecessary. Right. But we have had people say, oh, we tried this and it decreased it by 15%.
It's like, okay, well, what's that relative? But I would also argue that if you're gamey, if you have a lot of junk trash coming into your site, you're like performance marketing and you just have a bunch of form fills, I look at, like, any of you out of field or something, you might see potentially a drop. But if it's truly a high intent form, again, don't put this on all of your forms. This is meant for your high intent conversions only.
But I would argue you should look at your website to form fill to meeting booked as well. Like don't just stop at like, okay, we've implemented this and we saw, you know, flat conversion rate or like not a statistically significant dropping conversion rate, which we have seen with most of our clients, like we haven't seen it be statistically significant or even like buried that much. But I'm sure there's someone out there that will probably implement this and they'll probably see a dip. My advice to you is when you're setting up the test and you're going to roll this out and test it, look at like the traffic, the form fill rate, and then the form fill rate to the meeting booked.
And then look at that also after you implement that, like take a little bit wider lens on it because that is also very insightful. Like you're optimizing for quality, not just the form conversion point. And it's a good indicator quality. If someone was paying attention to your business, you're brand, they remember what they heard about you.
They have something to fill in like, they are higher qualified than some junk lead coming in through, you know, your low intent, non-branded pay search. Yeah, totally agree. I mean, we can kind of, you know, wrap up with some last comments here, but I would say the biggest unlock for me personally with like hybrid attribution is getting qualitative insights without having to do mass surveys or like, you know, trying to get the sales team to always ask the question on the call. Like, your sales team should still ask how they hear about us on the calls, but you have a data point that is your customer is telling you versus someone internally filling out that's always going to be better.
So that was a big unlock. And then also for me, like having access to hybrid attribution and looking at it both ways really gives you as like a marketing leader, a picture of how this transition from like a lead gen to demand gen is actually working. If you're not seeing, you know, something come through and self-reported attribution, like you're doing something wrong. And so that's the key insight to then taking like, okay, we need to fix our strategy on how we're actually trying to do this, right?
I think those are the two biggest unlocks for me personally of like thinking about this and working with a hybrid attribution model. What about you, Judy? Yeah, I totally agree with you. I feel like that was a very distinct summary of all the benefits.
But I think if we're going to take away kind of listeners or take away two immediate actions from this is if you haven't implemented it yet, why? What's your blocker and can you remove that because it's going to be really actionable for you? And then once you get implemented, make sure you're doing something with that data, make sure you're incorporating it into your MBRs or, you know, your quarterly reporting, surface that up. So leaders at your company see this and make sure you're really championing all of the activities you're doing that are showing up in self-reported that wouldn't be captured by the attribution software or, you know, vice versa, but that you're showcasing the full complete picture of marketing's impact.
So yeah, take a look at your forms if you're not doing it yet. It's our number one recommendation for this month. Yes, I love when people post on LinkedIn that they've implemented this. I'm like, hey, yeah, your guys are doing so good.
Go, go. Yes, agree. Yeah, so if you listen to this and then you implement it, post about it, tag us. We want to hear.
Yeah, we'd love to hear your insights. We love to interview people too that have implemented this and have insights that they learned on their business. We'd love to have you on the podcast. You know, people get tired of always talking to me and Judy probably.
Yes, come be our guests. Awesome. Well, thanks so much, Judy, for chatting and we'll be back soon with another episode. Thanks, Sydney.