Advertise • The Podcast • Our Social Media

Today's News:

ADS • Cheap “Made for Ads” Sites Still Pulling $

PINTEREST • Filtering by Body Type Rolls Out in US

GOOGLE ADS • Why Is Google Showing Competitors?!

ADS • Marketing Mix Modeling Tool from Google

YOUTUBE • #1 Creator Shares Thumbnail Secrets

Trust

Consistency

Visual Clarity

Testing

GOOGLE • New ‘Core Web Metric’ Coming Tomorrow

ADS • Cheap “Made for Ads” Sites Still Pulling $

Advertisers are still funneling millions into clickbait sites, defying industry efforts to cut back on such spending.

This from a new report by Adalytics, which found that inventory on so-called made-for-advertising (MFA) sites, continues to infiltrate the ad market.

The investigation, prompted by a Fortune 500 company's request, uncovered that it had inadvertently spent at least $10 million on MFA websites in the latter half of 2023, despite efforts to avoid such expenditures.

They also found that H&R Block delivered more than 2,100 ads to a single user on an MFA site within just one hour, and Comcast paid an astonishing $2,628 CPM to reach a single consumer on another MFA site.

Adalytics's findings highlight what many media buyers know is a widespread problem, implicating major ad companies and brands in the purchasing of MFA inventory, whether intentionally or not.

The report follows the Association of National Advertisers' discovery last year that 15% of programmatic ad dollars end up on MFA websites, totaling around $10 billion.

Efforts to address the issue have seen some ad-tech vendors pledging to remove MFA inventory from their offerings. But recent analysis found that brands, including Molson Coors and State Farm, continued to place ads on these sites early this year. State Farm countered by stating they do not endorse advertising on MFAs, aligning with brand guidelines.

Ad vendors such as OpenX and Pubmatic have taken steps to limit MFA transactions, with OpenX offering clients the option to exclude MFA inventory.

PINTEREST • Filtering by Body Type Rolls Out in US

Pinterest’s search tool to filter by different body sizes and shapes has now rolled out in the U.S. 

This is a great feature especially for brands that provide clothes for a variety of body types, though for now it’s only available for women’s clothes and wedding planning. The company says they’ll add men’s clothes later in the year.

It also can remember a preference. If someone filters their search results by a specific type twice, it will remember that for future searches.

Pinterest says people who used the filters showed a 66% higher engagement rate per session.

They have a patent on the tech which can identify various body types across the 3.5 billion images on the platform.

Pinterest has leaned hard into serving diverse people — their filters include the ability to search by skin-tone or hair styles.

GOOGLE ADS • Why Is Google Showing Competitors?!

SERoundtable has uncovered a slightly frustrating apparent addition to Google business profiles… in some cases, your profile will now also show ads for your competitors.

This was spotted by a user named Anthony Higman, who posted a screenshot on social media.

SERoundtable notes that this has been seen before with Google local listings and branded knowledge panels and LSA messaging.

So far, no response from Google.

ADS • Marketing Mix Modeling Tool from Google

Google has launched a new open source marketing mix modeling tool called “Meridian.”

Marketing mix modeling helps media buyers more accurately track how their ad campaigns perform across multiple channels. At best it’s an estimate, but the good models are reasonable at giving you enough feedback to know what direction to send your campaigns in.

YOUTUBE • #1 Creator Shares Thumbnail Secrets

Ask any YouTuber what the most important part of their setup is, and chances are they’ll tell you it’s the thumbnail image.

Some creators, like the wildly popular Mr. Beast, spend an enormous amount of energy on those — up to $10,000 on a design, and sometimes testing up to 20 variations for each video.

Last week, YouTube’s creator channel did an interview with Chucky Appleby, the guy behind Mr. Beast’s thumbnail designs.

The full interview is about 45 minutes long, but here’s the the four key points.

Trust

The first is about audience trust. If you consistently use clickbaity thumbnails that don’t accurately represent the content, over time people will learn that it’s not worth the click.

And it’s not just delivering the promised content, you have to do it fast — Appleby says the design of your thumbnail should focus on what you’ll deliver in the video’s introduction, rather than what many people do, and visually promise a big ending.

Consistency

Second, and I think most of us know this, consistency. If you look at Mr. Beast’s thumbnails, they have one element in them always — him, looking directly into the camera, with some kind of emotional expression on his face, usually happiness, but sometimes fear or boredom.

Visual Clarity

Your thumbnail has to communicate as much of the video’s main point as possible, but without a ton of clutter.

Testing

Finally, something hopefully you don’t need reminding about: testing. While not all marketers will have $10k to throw at testing thumbnails, you should still try at least a couple of variations.

Mr. Beast’s team often tests 50 combinations of title and image — switching everything up from changing his shirt color, adding a logo to his shirt, moving where his face appears on the image, and so on.

GOOGLE • New ‘Core Web Metric’ Coming Tomorrow

For the nerds out there, Google will be changing one of the metrics in its Core Web Vitals tomorrow.

Core Web Vitals are a set of web measurements that Google wants people who run web sites to improve — things like how fast a page loads, how soon a user can interact with it, and so on.

So First Input Delay is going out, and it will be replaced by with Interaction to Next Paint.

That new metric will observe the latency of all click, tap, and keyboard interactions with a page, and will report the longest duration, ignoring outliers. A low INP means the page is consistently able to respond quickly to the vast majority of user interactions. So it’s very similar to First Input Delay, but just takes a few more interactions into account.

Google has already added INP to its reporting in Search Console; most experts say it should have no effect on your Google search ranking.

¹ Some links in this newsletter may provide affiliate revenue to us.