Digital Advertising Metrics: The Limitations of Clicks and Viewthrough
Most digital media platforms—including search engines, social networks, and forums—rely on banner advertising to generate revenue. To determine if these campaigns are successful, advertisers track how users respond to display ads. These responses generally fall into two categories: immediate response, measured by clicks, and latent response, known as viewthrough.
While these metrics provide a snapshot of user behavior, relying on them exclusively can lead to an inaccurate understanding of a campaign's true impact.
The Problem with Clicks
Historically, the industry has leaned heavily on the Click-Through Rate (CTR), a metric provided by ad servers that measures the percentage of users who click an ad after seeing it. Ad servers also track unique impressions or reach by deduplicating browser cookies—small pieces of data stored on a user's device to identify them across sessions.
However, CTR is often a flawed indicator of success. Since the inception of the display banner, click-through rates have been steadily declining. More importantly, CTR often underestimates the overall brand impact and is susceptible to significant biases and fraudulent manipulation.
Not every beneficial ad impression results in a click. Some ads are never seen, while others are seen but not clicked, yet still prompt the user to take action later. This post-exposure behavior is what defines a viewthrough response.
[ไม่มีภาพประกอบ]
The disconnect between clicking and converting is significant. A study by the audience measurement firm Quantcast revealed that there is no correlation between users who click an ad and those who eventually make a purchase via a view-through conversion.
The Problem with Viewthrough
Viewthrough conversions track users who saw an ad (without clicking) and later converted on a website. This is achieved by the ad network setting a cookie and later matching it with a page tag on the conversion page. While useful, this system is open to manipulation.
Some unscrupulous ad networks engage in a practice called cookie-bombing. In this strategy, networks purchase cheap, "below-the-fold" ad inventory (ads placed so low on a page that they are rarely seen) specifically to drop as many cookies as possible across a wide user base. The goal is not to show the ad to a target audience, but to increase the likelihood that the network can claim credit for a conversion event that would have happened regardless of the ad.
[ไม่มีภาพประกอบ]
Key Facts
- CTR Decline: Click-through rates have been dropping since the beginning of display banner advertising.
- Viewthrough: This refers to post-exposure behavior where a user converts after seeing an ad without clicking it.
- Quantcast Finding: There is no correlation between users who click ads and those who purchase via view-through conversions.
- Cookie-Bombing: A fraudulent practice where networks buy cheap, invisible inventory to drop cookies and falsely claim conversion credit.
- Measurement Solution: Using standalone analytics packages and experimental design helps measure true incrementality.
| Metric | Response Type | Measurement Method | Primary Weakness |
|---|---|---|---|
| Click-Through Rate (CTR) | Immediate | Ad server tracking | Declining rates; susceptible to fraud |
| Viewthrough | Latent | Cookie and page tag matching | Vulnerable to cookie-bombing |
Frequently Asked Questions
What is the difference between a click and a viewthrough?
A click is an immediate response where a user interacts with an ad to visit a page. A viewthrough is a latent response where a user sees an ad but does not click, yet later completes a conversion action.
What is cookie-bombing in digital advertising?
Cookie-bombing is a tactic where ad networks buy cheap, below-the-fold ad space to distribute as many cookies as possible to users. This allows the network to claim credit for conversions that may not have been caused by the advertisement.
Why is CTR considered an unreliable measure of brand impact?
CTR often underestimates the impact of a brand because it ignores users who are influenced by an ad but do not click it immediately. It is also prone to biases and fraudulent activity.
How can advertisers avoid being misled by viewthrough data?
Advertisers are encouraged to use standalone analytics packages rather than relying solely on ad network data and to employ experimental design to measure true incrementality.
What did the Quantcast study conclude about ad clicks?
The study found that there was no correlation between the group of people who clicked on an ad and the group of people who eventually purchased through a view-through conversion.