Is advertising the “new dot com bubble” according to this intriguing article written on The Correspondent?
I’ve been chewing on this question since the article came out a couple of weeks ago.
Well, I’ve been chewing on variations of this question for a while, but not in the “bubble” language ascribed by Jesse Frederik and Maurits Martijn.
As I ponder this assertion, I think I would like to change the original title (and premise) to the following:
Original Title/Theme: The New Dot Com Bubble Is Here: It’s Called Online Advertising
Suggested Title/Theme: The New Dot Com Bubble Has Been Expanding: It’s Called Attribution of Sales in Online Advertising
I found myself nodding my head vigorously at much of Frederik and Martijn’s article, specifically as it called out the digital advertising community’s obsession with chasing tracked profit at the expense of incremental value and actual new customer growth.
However, here is where I disagreed with the article: the authors blamed the failure of marketers and engineers to actually demonstrate incremental value on digital advertising itself, rather than on the tactics devised from an improper understanding of attribution.
Here is what I mean, digital advertising is just advertising. It’s not the greatest thing to ever happen to marketing, and it’s not a bubble. It’s just advertising.
It’s doing what marketers have done for years, utilize a specific medium to grow a brand (and thus sales) by getting the right message in front of the right people at the right time.
Discovering that people at eBay running most of their budget into their own (exceptionally powerful) brand terms are surprised to learn they don’t see incremental value isn’t the nail in the coffin for advertising the article suggests. Ironically, PPCers in my sphere have long written on the poor eBay PPC program management evident even from the public eye (“used babies” in titles thanks to DKI, anyone?).
This is why it’s crucial to point out the thing I believe the article alludes to but doesn’t actually identify as the actual bubble: that is an improper understanding of attribution, and how that establishes misguided tactics for paid search accounts that fail to build brands and add incremental value.
In other words, I am positing that paid search advertising itself has not failed, it’s that an understanding of how to use paid search advertising as part of an integrated marketing mix for individual companies has failed. Improper use of attribution has led to an obsession with directly tracked results that over time do not build a brand and incremental sales. They simply re-target (not necessarily remarketing, btw) the same users already in the sales cycle – ad nauseam.
In this regard, I would suggest that conversion tracking is as much of a curse as it is a blessing.
When you can track what source led to a sale, you begin to think you have an understanding of how your consumers purchase, and you begin to invest more money into that source. But what if that source is only one piece of the puzzle… especially if it’s closer to the bottom of the buying funnel, meaning much contact has already been likely made by your company?
When you think you can track everything, you begin to shift your time, resources, tools, and reporting to making your trackable KPIs grow, rather than building and implementing the tactics to accomplish an actual marketing strategy within your digital channel.
If your paid search strategy focuses solely on sweeping up those bottom of funnel clicks and sales (which is what you’re tempted to do with a last-click attribution model that gives 100% of the sale credit to the last source to send you the sale), then yup… you maybe won’t see much damage (at least initially) in pausing paid search. To be clear, certainly, there are times in competitive industries, (especially with startups who don’t have more advanced marketing channels built out yet) where using paid ads to initially catch those bottom funnel users is a sound tactic.
I’m going to complicate things even further. You may be reading this and vehemently agreeing with this. “Death to last-click attribution!” you cry.
However, let me push into this even further. I believe there is no perfect attribution model, and with privacy awareness increasing and dark traffic in a continually strong space, this means you can’t really trust a more complex attribution model either.
Why did a person visit on their third visit and decide they “loved this brand and had to have one” because of “just the right” emotional experience. But, then didn’t actually purchase until their seventh visit, 12 days later? Who knows? That’s not something you can track. Attribution will always have limitations, which is why in some ways, attribution (improper or proper usage) itself may be the true dot com bubble the authors are sniffing out.
Regardless, as long as we keep chasing solely after tracked individual channel success and building digital marketing strategies (selecting keywords, ad text choices, locations, devices, audiences, demographics, etc.) without thinking beyond individual channel success, then we will continue to struggle to build brands and see incremental growth.
Only when we as paid search marketers strategize with the other channels, to build a marketing strategy targeting the right message to people at the right place in the funnel (selecting unique keywords, and channels and campaign types for those, of course) will we begin to get beyond tracked ROAS as our primary KPI and focus more on overall brand growth across all marketing channels.
If what you’re hearing scares you, because it sounds risky, well then, I think you’re picking up what I’m dropping. Real, authentic marketing that builds lasting brands has always been difficult, risky, time-consuming and expensive (with a healthy dose of luck).
I don’t think digital advertising is the dot com bubble, I think our belief that we can track everything is the dot com bubble.
Time to get back to marketing, worry less about tracked profit and build a brand.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.