Online Advertising isn’t a Bubble, it Needs the Magic Layer – Part 1

Last time, we mentioned we plan on commenting on related news and opinions in the analytics, marketing, and digital space. With this in mind, and after some pushing from people we respect, we think the piece from the Correspondent titled “The new dot com bubble is here: it’s called online advertising” deserves some attention and a response.

As with a lot of topics, we find ourselves agreeing with most of what is written, then (not so) completely shocked that what we consider to be the wrong conclusion is found. To get straight to the point, we do not see online advertising as a bubble. We see it as needing to find its focus. It needs the Magic Layer.

Photo by Free To Use Sounds on Unsplash

Why do we think so? Let’s jump to what we consider the critical finding from the research and where the interpretation really matters most. The graph titled: “Facebook ads are far less effective than we thought” shows a clear pattern.

Most of the effect within most studies can be attributed to the selection effect. It follows that “Is online advertising working? We simply don’t know”. Some want to draw a single conclusion from the data that advertising doesn’t work and is therefore in a bubble.  Or that we don’t know if it works.  But here is the exciting bit: we could know!  It is possible.


We see opportunity. We see that that there are at least some situations where more than half of the effect is the advertising effect, and more than half of the studies see a non-zero advertising effect. 


This is critical.


Advertising does work to some extent most of the time. We just have to isolate which advertising works, in which situation, at what levels of return. Then do more of what is working, and less of what is not.


Many only see analytics as post-execution validation, measurement, & reporting.  But this is not good enough. This is how we got to this world, where most of the effect within most of the studies can be attributed to the selection effect.

This is a call to research and analysis. To discovery and iteration. To better methodologies of working. Our role as analysts in the marketing and digital space is to do this hard work so we can unleash strategy and creative skills and professionals to have the largest impact that they can on the business.

So how do we create this new world?  We need to combine research with iteration.  We have created a 10 step process that has 5 research steps and 5 steps that power iteration.  Below are the 5 research steps, and the remaining 5 steps related to iteration will come in a future post.  A word of warning here, as with most things worth doing, this is not easy – but it is possible.  We can do this!

I have executive support and buy in to work differently

NewVantage Partners’ 2019 Big Data and AI Executive Survey found that leading executives know that people and process are the main obstacles to successful business adoption of data initiatives. Don’t let this common pitfall disrupt your efforts. Start here. Find the highest-ranking person within your organization who is willing to support this kind of initiative and recruit them. Tell them their public and explicit buy into this type of project is critical.


I have identified a swat team of people willing to work differently

This is just as important as executive support. The people who will be executing this type of project need to be selected carefully. We recommend asking for volunteers. They have to be open to doing things differently and opting in shows that willingness. Selecting membership on this team based only on who is available is a recipe for tension during execution and will slow everyone down.


I know who my users are – not just those that engage, convert, or like participating in research

Most people focus on only the small percentage of users who do what you would like. Creating different segments of users based on differences between converters is a common pitfall of Selection Bias, specifically Convenience Bias. The real power comes from understanding how to segment all of your users. Even those who do not convert, those who bounce or do not click at all, and even those that do not like participating in research. Understanding the segments that exist across all your users will show you who you are winning with, and who was not convinced. We recommend integrating quant and qual to discover your segments.


I know what my user segments want from my product or service

Once you know who your segments are, and their current behavior patterns, the next step is to discover what they need from you. Is there a specific aspect of your product or service that is most important to them? Do they prefer different channels? Do they have different timelines to conversion? The point here is that your segments will have different wants, needs, and desires. And those will mean that specific aspects of your product or service will be important to them. We recommend survey data integrated with behavioral and user interviews to really get to the bottom of what is desired.


I know how to position my product or service optimally for each of my user segments

Every strategist and creative person knows that discovering what a segment values, and positioning your product or service in that direction, are two very different things. At this point we know who our segments are, what their behaviors are, and what is important to them, and their preferences. Now comes the fun part. How can we communicate our product or service in a way that communicates how amazing we are in the way that each segment will appreciate most? We recommend coming up with 3 positioning ideas per segment that include appropriate visual design and copy.

Alright!  Now that we know what it takes from a research perspective, next time we will explore the iteration aspects needed to ensure your marketing is having it’s intended impact.  Until then, stop guessing and start researching!

Make sure you sign up for our newsletter so you can get notified when new content becomes available.  And as always, if you have a business problem and think we can help, reach out and we can discover together if it is a good fit.

Like this article? Share it!

Share on facebook
Share on Facebook
Share on twitter
Share on Twitter
Share on linkedin
Share on LinkedIn
Share on email
Send Email

don't like it? let us know why.

Want More?

Subscribe to our newsletter to get the latest news and tips delivered directly to your inbox.

Related Posts