Table of Contents
Online advertising appears to be the ideal fusion of data and creativity. Advertisers can test out different ad styles and see results quickly. Digital advertising is almost entirely quantifiable. “Data-driven marketing” is what it is, unless the stats are entirely made up, which is becoming more and more likely.
I recently had a conversation with a business associate who had spent years working in advertising. He got wind of disparities in reports from multiple sources, including Google, and got in touch with them to fix the issue. They advised him not to check into it if the difference was less than 20%. Although there is a significant error bar in your advertising statistics, it is insignificant in comparison to what other people are discovering.
A few really depressing numbers
The Association of National Advertisers (ANA) reports that just 36% of advertising spending on demand-side platforms really reached the advertiser’s target demographic. A significant portion of the money was spent on advertising that were unviewable or only displayed to bots, according to the ANA research on the programmatic supply chain.
According to Bob Hoffman, the author of “Adscam,” the figures are much lower. According to his estimation, only 3% of digital ad spending is related to real human ad views.
It should be noted that statistics are provided at every stage of the advertising campaign. Everything is very “data-driven,” but the information may be false. What is happening?
The well-known statement, “Half the money I spend on advertising is wasted; the trouble is, I don’t know which half,” is credited to John Wanamaker. Ad executives could confirm in Wanamaker’s day that their advertisement was indeed on the billboard, in the magazine, or broadcast on the radio. You are unable to verify that your advertisement is displayed because online advertising is personalized for each user.
How could the numbers be as dismal as the ANA indicates? Let’s go back and investigate the potential causes of these stark disparities in data, concentrating on a few important parameters.
Perspectives and feelings
Advertisers want the right people to view their commercials in the right situations. There are several reasons why any one of those components could go wrong. These are a few of the main offenders.
sensations that are not visible. These happen when your prospect visits the appropriate page, but your advertisement is “displayed” in a part of the page that the prospect never sees. Advertisements can still be recorded as “viewed” even though the visitor never saw the ad, despite efforts to prevent this.
The advertisement is not on the correct website. Your advertisement is really shown on affiliate websites that don’t draw in your target audience, even though you want it to appear on a well-known media website. This can be the result of confusion on the back end or human error during the setup of the advertisement, but it does happen. The advertisement appears, but it may not be seen by the intended target or in the appropriate context.
deception. Ad fraud is one of the biggest problems with online advertising and encompasses the following:
- Using bots or automated scripts to create fake impressions.
- “Ad stacking,” where multiple ads are placed on top of each other in a single spot so that only the topmost is visible.
- Pixel stuffing, where a large ad is stuffed into a 1×1 frame.
I understand your thoughts. “I’ve known my representative for years, and he is an honest guy.” Okay, but even if your representative is a saint, he is but one part of a much more intricate system that nobody truly comprehends from beginning to end. Both good and evil actors are part of this system; examples include nations whose primary source of foreign exchange earnings is various forms of cybercrime.
Reporting is skewed by the possibility of counting repeat impressions for the same user. Adtech can and should, once more, make up for this, but it’s not a given.
Added to all of this are issues with geographical insignificance, disparate impression counting standards across various platforms, and standard human mistake or technology glitches.
What is the total of all of this? Although Hoffman tries to explain everything in his book, I believe the truth is that only God knows.
The click-related issues
A visitor to the website may be impacted by an advertisement just by viewing it, and many advertisers want “view through conversions.” Because they direct your prospect to the page of your choice, clicks are internet gold. Hopefully, anyway. Issues still exist.
Click fraud enabled by bots, inadvertent clicks, redirect loops, proxy server traffic, and simple dishonesty can all taint click data.
The benefit of clicks is that they direct users to your page, where you are in control of handling the traffic analysis. You can better determine how many people were drawn to your advertisement by filtering your traffic if you have bot detection technologies.
Error is still possible. Clicks and page views will not match if the user exits a page before the tracking code activates or if the tracking code fails.
The difficulties with attribution
In marketing, the “rule of seven” is a well-known idea. According to the theory, a person must hear or see a marketing message at least seven times before acting upon it.
Which of those seven touchpoints—if true—should be credited with the sale? Though trite, this query is essential to internet marketing.
In his piece “Measuring the invisible: The truth about marketing attribution,” Tim Parkin refers to marketing attribution as “the fog of marketing” in part because of this. Technical concerns can lead to inaccurate attribution data in addition to these conceptual challenges with attribution.
Various advertising networks track conversions in different ways, which causes disparities. A conversion during a designated window following an ad view or click may be tracked by one system. For instance, a conversion can be defined as occurring within 30 days of seeing or clicking on an advertisement. Last-click attribution or variants on that theme are employed by other systems.
Increasing what works and decreasing what doesn’t is the goal of optimizing ad spend. How do you proceed if a prospect needs to view seven advertisements before acting?
Do you think advertisements are real?
Does this mean we should give up on advertising if the majority of the data is garbage, which may very well be the case, and if attribution is somewhat of a fantasy, which also appears to be the case?
Without a doubt. Even after learning which advertisements were effective, Wanamaker kept purchasing them. Furthermore, commercials have a significant impact on our shopping decisions, as we all know from personal experience. Promotion is effective.