Solving The Problem Of Unattributed Conversions With Rockerbox
Whether you call it direct, dark, or unknown traffic it can be a black hole along the path to conversion that obscures the impact of their marketing channels. Throughout this webinar Ron and Matt discuss the why behind dark traffic and how brands can shed light on the complex issue by combining an attribution platform with post-purchase surveys.
What is direct/dark/unknown traffic and what are the reasons it appears?
We often use direct just because that's what's shown in Google Analytics. But direct traffic is very much just traffic has no indication of where it came from. So could be a ton of reasons. It could be someone just typing in the URL of your site and going there directly. It could be UTM is dropped off based on some kind of browser issue. It could be just an untagged link in an email. There's a whole slew of different reasons why you might have direct traffic and I'm sure we'll get into this but direct traffic it's not helpful in the sense of understanding where customers are coming from and allowing you to better allocate media spend in your strategy on the marketing side.
Why is it a problem for brands?
It's a problem because you want to make sure that if you're spending somewhere that you, as much as possible, know that it's driving results. If you just launch you know billboards in Williamsburg, like you have everywhere over here. And you see a sudden spike in direct traffic. You want to kind of go with those billboards did something. So like it's a natural consequence of your marketing. But that's where, you know, can you use promo codes can use vanity URLs, can you use post-purchase surveys like what tools can you use to try to like help them uncover that what could be a problem, “Oh, my God, more people are coming to the site directly what the hell's going on?” To actually turn that kind of back in marketing. I know you know what, those billboards are doing something we should be investing more and more there. So it's a problem and so much as like you're kind of paralyzed from taking action or knowing what's going on. But I think tools like Rockerbox and Fairing if we do our jobs well should mitigate that as much as possible for you.
What are the different roles that attribution and post-purchase survey platforms play in demystifying it and why is it important to supplement attribution data with customer-level data?
So just to take a step back on the attribution side, there's really two key questions. One could argue there's more if you're trying to measure returning customers on the attribution side, but it's really (1) “How did you hear about us?” and (2) if you wanted to understand the lag to purchase time, it could be, “How long ago did you hear about us?” But for this context, we're really going to just talk about how did you hear about us surveys ... [which] do a very good job at just understanding a holistic overview of all of your channels that are driving customers to purchase. So we've mentioned influencers, podcasts, and … paid media, kind of strategies and tactics you can implement. But what surveys are going to do is help capture Word of Mouth in some of these other kind of ways in which your product can see virality and Word of Mouth can literally be someone walking down the street with a pair of Allbirds kind of thing. So it's going to do a much better job of capturing some of those top-of-funnel channels and provide a more … holistic overview of what's happening.
What do you do when you get conflicting information on a source?
Yeah, … it's very often on channels where you'd expect it. So you think about podcasts, influencers, TikTok is getting better. But TikTok a year ago, was essentially zero attribution was attributed to TikTok. And so you started asking questions and we've, anecdotally, I’d say hundreds of customers who are now spending on TikTok, and at first didn't think it was working. So I see the word like conflicting and conflicting is often like a negative word. But I think in this sense, it's it's a good thing in the sense of, okay, we're learning more. So, to Ron's point, like we definitely echo the same thing it’s really all about triangulation and creating that focal point. So if you think of like some kind of multi-touch kind of click-based attribution model, which is very much a modeling approach, you have your post-purchase surveys, even maybe your media mix modeling if you can get to that scale. So it's really all about that triangulation to help you tell that better story. So I think the conflicting information is at least with our customers, it's not usually like “Oh, like damn, we have conflicting information.” It's usually like a smile on their face. Like, “Oh, I had no idea this customer came from this channel.” And now we've unearthed that. So that's typically how we think about conflicting is more or less a positive.
What can brands start doing differently once they’ve got a better grip on direct/unattributed traffic?
So the biggest thing is just help diversifying the media strategy. And I think that's the that's the key kind of learning here. If you're overly leveraged in let's say, Facebook, for example, we saw what happens with whether it's Apple making an update or just CPM is going crazy. So that's the that's the one positive core outcome here is improving the diversification of your traffic once you better understand all of it. So we we often argued the brands that create their own kind of opinion or model etc, on attribution or just even just like reporting and whatnot, will build a better competitive advantage.
Is direct/dark/unknown traffic something we can ever totally solve? If now, why is that not such a bad thing?
I don't think any of this stuff is solvable. Like … one of our customers says, “Attribution is that journey with no destination.” And when I heard that I laughed, because it's very true. Listen, we're trying to solve something that's inherently unsolvable. And that can be very uncomfortable. Unless you're just okay with that. And I think the whole goal of measurement and frankly, marketing overall is just to continue to get better and continue to take something and find incremental ways to just get more out of each dollar. I think it's how do we minimize that? How do we minimize imperfections, and do a better job of understanding what's working?