Welcome to a new part in our series, A Debate on Standards, as we open up a space to discuss the problem that a lack of standards in mobile attribution poses. In this post, let's hear from TUNE's Director of Product Marketing Brian Marcus on his views on mobile attribution space. He also joined us recently in a webinar to talk about the challenges of lack of mobile attribution standards. In case you missed the webinar, make sure to check out the recap at the end of this interview!
Q: To start off, can you tell us where you see the greatest challenges in the current mobile advertising ecosystem when it comes to attribution?
A: Similar to many other advertising ecosystems, attribution in the mobile advertising ecosystem can be tricky because mobile device “measured outcomes” may only represent a small slice of the complete customer’s journey. For instance, many users will click on one device, install on another, and use on the web throughout their lifetime. Without a full view of the journey, or a people-centric approach to attribution, we aren't giving the full credit across the partners. We need to be able to reconcile all channels and devices in a comprehensive way to ensure that the parties that influenced the action are recognized.
Similarly, most of the digital world subscribes to a last-click model, which isn't always the best path. With a model like “last-click” there are trade-offs, between simplicity with sophistication. On the plus side, the “last click” model is simple and clear, so that Publishers and Networks can optimize easily against it. However, last-click model penalizes those that drive upper funnel engagement. Today, the majority of advertisers default to last-click, as it is hard to get anyone to agree on much else. In effect, we remain in a last-click paradigm because it is the cleanest and simplest means to draw the line.
Q: There’s a lot of talk about working with the right attribution models, but many over-simplify the approach. How should advertisers choose the right model?
A: The best models should accurately measure the touchpoints that influence an action that meets the advertiser’s goals. At the same time, a great attribution model should be supported by the right incentives to align the ecosystem. Meaning, in the end, all parties must derive value from the overall arrangement. If you move away from last click, you have to think about how that change will impact those that are closest to driving the “action” or “outcome”. For example, if you use multi-touch attribution and spread an incentive across both upper and lower funnel partners, you have to be careful not to spread the incentive so thin that it does little to motivate any of the partners that drive the value.
So you have to consider whether your rewards are motivating enough to compete in the mix, as you attempt to level the playing field. Publishers have choices as well. Splitting up a single incentive 12 different ways may be more democratic, but it may not be more motivating… something to contemplate as you design attribution and incentive models.
We also need to account for the knock on effects of some channels in a more productive manner. That is, some channels may shy away from last click, but their value remains strong as part of the overall picture.
One strategy to help marketers understand the value of their channels is through lift testing (or incrementality testing). Meaning, can you hold a control group back from seeing your ad and measure their performance versus those that did see the ad -- regardless of where in the click-stream the partner contributed most frequently, you can still see the overall value of the partner. Doing these types of tests across partners can help you to determine the best attribution model going forward.
Q: How do you address the issue of attribution fraud?
A: The inherent problem with fraud prevention is that most third-party solutions have been built for single players, and not for the larger ecosystem. To dial down fraud successfully, it takes a village — and all sides need to truly understand the data, align incentives, and operate with a high level of transparency. Today’s third-party tools that attempt to tackle fraud do so from only one angle, with only half the picture, and often within a black box, which only exacerbates the situation.
Why? These third-party solutions only reinforce the problem, as fraud relies on opacity, conjecture, and distrust to survive. More concerning is that as mobile marketers are faced with evolving systems and new form factors, you can be sure the fraud picture will grow murkier. Without significant change, everyone loses.
Without proper precision and granularity, high-quality traffic is discarded with the bad. Heard the expression, “Throwing the baby out with the bath water”? That is largely the effect of using blunt instruments on your most precious relationships. Fraud prevention requires a more thoughtful set of precision tools to help each side of the relationship fulfill the jobs that they aspire to do. This means drilling below the ad partner, geographic, and campaign levels. Best practice means marketers take the time to isolate the issue, build trusting relationships with their partners, communicate rules of the road, and collaborate together to distill the good from the bad.
We believe in 4 steps to addressing fraud:
- Identify: Use sophisticated and granular reporting to find the true bad actors (advertising partners, publishers or sub-publishers) that are behind driving the fraud. Reporting should be down to the sub-publisher and sub-campaign levels.
- Act: Once identified, build traffic filters / tools to prevent and control fraud.
- Align: Work together with your partners. Tell them the reasons for your decisions and provide the rules of the road, upfront.
- Automate: Systematically build fraud pattern matching into your measurement systems and detect patterns as they emerge.
Q: What trends are we seeing in 2017 and what are the current limitations of mobile attribution that must be fixed?
A: We believe a device-centric approach to mobile marketing leaves far too much opportunity on the table for marketers and partners alike. Moving mobile marketing toward a people-centric approach will allow marketers to see the true impact of their marketing efforts, allow more marketing to become performance marketing, and ultimately show a truer value of ad partners.
Similarly, mobile is about location or knowing where a user is in the time of a “mobile moment.” But mobile marketing has long been siloed away from being location-centric, and we believe that should change. Knowing where your users are, what they’re doing, and how they’re behaving at any given point-in-time matters. Allowing our mobile marketing to take these data points into consideration will grow both top of funnel acquisition as well as down funnel engagement.
Also, understanding synergies between Organic and Paid Marketing, as well as Acquisition and Engagement marketing will yield better ROI and Customer Lifetime Value across your customer journeys. These components all work better together and should be unified.
Lastly, fraud continues to be a massive thorn in our industry’s side. We must continue to work together in a transparent manner to help all parties beat fraudsters as discussed above.
Q: How important is harmonization of attribution measurement across different channels?
A: Pretty darn important… but truth is, harmony across channels depends on the right measurements and the right incentives. Harmonization requires a new paradigm so that channel and device do not get in the way of good marketing orchestration. The real win is to center measurement around people. People-centricity means that we are able to track a customer’s journey regardless of the devices or marketing channel a user leverages in making his / her decisions.
When you know a journey at a customer level, you can get much clearer on the influence that marketing plays along the way. As patterns emerge, you are in a better position to lean in on those paths that are best for your users and pull away from those paths that drive lower quality traffic. Inherent in tracking people is the ability to leverage location data. A marketer needs the context of a customer journey to best serve that customer and location becomes a critical capability.
Q: The first impressions is/should be the most valuable, yet, the last touch point with the user is what gets attributed - how do you think this problem should be solved?
A: Our job is to create transparency for every advertiser on how their users best find and engage with their brands. For some, the first impressions may be the most valuable, but for others, it may not be. The answer largely depends on the product, the goals of the marketer, the market maturity, and the needs of the buyer.
In any case, attribution has to be about accuracy, objectivity, and transparency. Systems that measure marketing effectiveness have to be good at collecting, aggregating, normalizing and transforming the data so that both advertisers and ad partners can gain true insight into what factors lead to an event.
Next, incentives must drive the right behaviors. Ensure that you tie the incentive to the desired outcome, be it first or last click. Most critical in this equation is the Advertiser’s objective. Is it Exposure? Trial? Usage? Engagement? For each of these objectives, there are different partners that I would want to motivate and activate at specific stages in the customer journey.
Stay tuned for the next part in series in the upcoming weeks. Our readers will get to read on mobile attribution in greater detail through interviews with some of the major attribution partners in the industry on their take on topic of mobile attribution as well as standardization. Our readers will also be able to get deeper insights into the topic with planned webinars with representation from key stakeholders in the industry to start an active dialogue.