Removing Ads On Website Improved Its Ad Earnings?!

Why Removing Ads On Website Improved Its Ad Earnings?!

It seems crazy, right? The idea that your website would actually earn more revenue from showing fewer ads. What if I told you that this would also result in lower RPMs and CPMs _BUT_ higher overall ad earnings. Absurd? Maybe not.

This is the concept John Cole and myself recently explored at Google in New York during _Pubtelligence,_ a Google-hosted event designed just for digital publishers.  We took a deep dive into the economics of digital publishing and **discovered two things that influenced website ad revenue more than ALMOST ANY OTHER FACTORs.**
**These factors were visitor engagement and visitor behavior**.

We learned that these can both be manipulated and influenced to produce greater ad revenue and better user experiences; however, many publishers may be approaching these subjects in the opposite way that they should be.

Below, I’ll highlight how to improve visitor engagement to increase the value of ads on website spaces. I’ll also show you how to influence visitor behavior to enhance overall session earnings for every individual visitor.

Publishers need to please a lot of parties with online ads

The advertising ecosystem has never been filled with a more difficult group of parties to please. That includes platforms — like Google and Facebook — advertisers, and visitors.

Advertisers are leveraging new technology and sophisticated data to divide and segment visitors to improve the way they buy publisher inventory; forcing publishers to account for things like viewability and other metrics.

Finally, visitors expect more from their web experience than ever before. They expect cross-device synchronicity and are expecting richer mobile experiences. Adapting to these trends has become no easy feat.

Publishers are “over-monetizing” and it’s costing them money

**Second**, visitor attention is what advertisers are buying when they bid on your ad space programmatically. Since they are bidding on auctions inside of exchanges, they may not bid on certain ad space unless it is getting a certain level of attention (click-through-rate, viewability, campaign performance).

Improperly accounting for both of these factors means that you could see shorter sessions and less competition among advertisers bidding on your ad space in programmatic exchanges.

Below, I’ll give you an example of how over-monetization of web pages takes place.

Understanding ads on websites and how advertisers bid for them

Most digital advertising that takes place on publisher pages happens programmatically. This occurs in large exchanges where advertisers bid for certain audiences they want.

the image on the left is the large dataset used inside of Google DFP’s real-time bidding protocol; where advertisers buy website ad space based on historical data

These bids are placed based on historical data. They look at specific locations on certain pages that perform according to rules that they have put forth to help them achieve campaign goals.

This is **an example of how publishers “optimizing” around CPMs and RPMs can find themselves quickly losing revenue** if they don’t know how to measure things like [EPMV](https://blog.ezoic.com/why-ecpm-vs-rpm-is-skewing-true-earnings/) (earnings per thousand visitors) or authentic user engagement.

The advertiser here is a jiu jitsu uniform company. They may be doing some form of retargeting via an advertising platform like Google AdWords.

My jiu jitsu website may be a place where these ads would potentially show; meaning that this advertiser would put Google ads on my website via Google AdSense or Google’s Ad Exchange.

They may know that this 300×250 ad gets an average a 5% click-through-rate (CTR); which means it would need to generate 20 clicks to reach their campaign goal.

This means that — programmatically — the protocols inside of the auction will help them seek out ad space that will help them meet these goals through targeting and historical ad space data.

As a publisher, I may be taking a close look at my CPMs when trying to optimize my overall revenue.

As soon as I learn that I have a $50 CPM on my site, I may look to capitalize on this page. Perhaps this is a particularly high CPM for my website.

A common way that ad ops and webmasters will approach optimizing this information is to conduct an A/B test to learn about the financial impact of including other ads on the page.

Unfortunately, ads have a strong impact on other ads — and a dramatic impact on visitor behavior. **Over time, those ads will dilute each other**; meaning that 1% CTR that ad space had will likely degrade. This means that the historical bidding data will change and advertisers will have bids automatically adjusted to compensate.
In this case, for simplicity, we’ll say the ads split the CTR.

Yikes! **This is why you often hear about publishers complaining of ad rates dropping when they are actually on the rise**. The addition of more and more ads on a page means that ad rates may be declining on a particular publishers site.

This typically takes place over long periods of time, so this erosion of ad value over time is not always attributed to the addition of ads or technology from 6 months ago.

Example of how bounce rate is affected by ad density on a single page

- [Ads also impact visitor behavior](https://blog.ezoic.com/user-engagement-imapcts-ad-revenue/)
- We know that ad density alone can affect objective UX metrics
- This image is the impact of ads on bounce rate for a single page of a single site on specific visitor segments

This means that even a simple 12.5% increase in bounce rate could cause a drop in the average pageviews per visit; resulting in less overall pageviews from each session.

In our example, the increased ad density actually caused a decrease in avg. pageviews per visit. This led to overall less session earnings than before.

This is what leads to over-monetization and it can be avoided by taking a long-term look at how ads and other technology impact visitor behavior and ad inventory value.

To understand the trends behind both of these things combines — accounting for seasonality and traffic variability — EPMV, or session revenue, is the best way to keep a _true north_.

The impact of engagement of website ads and their value

I deeply discussed [a case study we did on actual page engagement and ad value here](https://blog.ezoic.com/make-website-ads-valuable-advertisers/).

Basically, there is a direct correlation between the value of ad space to advertisers and actual visitor engagement — time spent engaged in content; **_NOT_** waiting for pages to loading, scrolling quickly, navigating in the menu, or browsing in another tab.

You can [learn about actual visitor engagement time here](https://blog.ezoic.com/engagement-time-important-top-publishers/).

Viewability, CTR, and campaign performance are exactly what advertisers are bidding on in programmatic auctions. **By improving engagement, you are making your ad space more valuable in these auctions: driving up prices and competition for your inventory.**

Tests that can help you earn more website revenue

Rather than optimizing for CPMs and RPMs, you need to be looking at EPMV and session revenue by landing page to ensure long-term website revenue growth.

If we return to our example from the beginning; of a page that was “over-monetized” — with too many ads, videos, widgets, and other technology — we can actually test our way out of this vicious cycle.

In this case, we are going to test that auto-play video that hovers as visitors scroll on desktop devices. That video may play an ad which helps us generate a higher RPM than we had without the video.
But, when we look at the ad’s impact on avg. pageviews per visit of visitors landing on the page, we see that they visit fewer pages overall than the visitors that don’t see the video.

This has a big impact on overall session revenue; meaning that the video is actually **COSTING THEM MONEY AND ANNOYING USERS**.

Visitors affect your ability to make money from ads in two ways

Wrapping it all up

- Many publishers are over-monetizing their sites — or at least overloading their users.
- This cannot be seen by looking at RPMs or CPMs.
- Optimizing for RPMs and CPMs leads many publishers down this path.
- Understanding and running tests that account for how changes affect EPMV (or session revenue) will help prevent these risks.
- Optimizing for EPMV is also the best way to ensure you keep revenue per visitor going up over time.