Post updated on October 18, 2023.

Let’s talk about the elephant in the internet: Google’s September Helpful Content Update.

We turned to the data available to us across more than 10,000 sites to try to determine the volume of impact and potential causes for why some sites were hit by this update.

It’s important to note that we’re bringing you this data now — as opposed to two weeks ago — because we needed to allow some of the dust to settle. Running these numbers while the update was ongoing would have given us incomplete results and it takes time to analyze the data and pull it into something shareable. 

Let’s get to it, shall we? Warning: Numbers ahead.

How Many Sites Were Affected by the HCU?

Of the 10,302 sites represented by Mediavine ad management — as of the writing of this article — 607 were identified as having been negatively impacted by Google’s Helpful Content Update.

That’s 5.8% of sites. 

We also discovered 1,170 sites which saw positive increases to their Google referral traffic coming out of the HCU.

We know being part of that 5% feels huge to those whose sites were significantly impacted. But knowing whether or not your traffic decline is due to the HCU or is an organic decline over time helps us — and you — plan the next steps.

So, how are we defining “significantly impacted”? Glad you asked.

Our HCU Methodology

We started by looking at traffic across Mediavine publisher sites from the start of Q3 (so, July 1). Once we had start and end dates for the Helpful Content Update, we excluded those dates from the report. This was important to get clean data since the rollout was progressive across sites.

We also looked at other sources of referral traffic and identified sites whose overall traffic from all sources had been on a steady decline since before the start of the HCU. The indication in those cases is that, while Google referral traffic may have gone down, the update wasn’t the primary cause. 

Then we looked at the percentages of traffic increases and declines and grouped them like this:

• No Change
• Increase (greater than 5%)
• Decline between -5% and -20%
• Decline between -21% and -50%
• Decline less than -50%

In this case, “less” means more decline since we’re moving to the left on the number line.

(Confusing, yes, but our data analysts assured me this is accurate. And trust me, I asked if they were sure.)

From there, we filtered out sites making under $100 per week in gross revenue, not because these sites aren’t important, but because comparing a 50% decrease in revenue on a site making $100 is vastly different from a 50% decrease on a site making $10,000.

Both are 50%, and yes, both publisher groups matter tremendously. But the data sets can’t be compared fairly.

Now let’s talk about revenue, because a drop in traffic should equal a drop in revenue, too, right?

How Has the HCU Affected Revenue?

Across our publisher sites, we’ve seen around an 11% decline in referral traffic from Google, but we have not seen a corresponding decrease in revenue. Revenue remains at a net 0% change, most likely due to the HCU timing.

Since it coincided with the end of Q3 and the beginning of Q4, advertisers were already primed to increase their spend, and what the data shows is that many sites which lost Google referral traffic gained traffic from other sources, like Pinterest or Facebook, or even from other websites.

In fact, 40 of the 607 sites mentioned above experienced a positive gross revenue change despite a decline in traffic coming from Google.

Which Niches Were Affected by the HCU?

One of the questions we’re asked most often whenever we’re asked about data is “which niches are ____.” Insert anything from “most profitable” to “impacted” in that blank.

It may look as if lifestyle niches were most impacted, but that’s because lifestyle categories dominate Mediavine’s inventory. These are the results we expect to see here.

With that said, the breakdowns for the top 10 niches affected — positively and negatively — are as follows:

There are times when we can say, definitively, that some niches are hit harder than others.

Travel sites following a pandemic, for instance.

This is not one of those times. 

The top three niches across both positive and negative impact groups were the same and represent three of the four most high traffic niches on the internet.

There’s not a lot of evidence to support a claim that sites in certain niches were impacted more heavily than others once we acknowledge that the most heavily impacted niches are also the most abundant.

Are Higher Ad Density Settings Related to the HCU?

Based on the data available to us across more than 10,000 sites, we do not have any reason to believe that sites running more Mediavine ads — or fewer — were more heavily impacted by the Helpful Content Update. 

The data about ad density and the HCU is pretty agnostic.

Sites running all configurations of ad density settings were impacted, and sites running recommended density settings were impacted at the same rate (12%) across both Desktop and Mobile settings — even while representing 54% and 93% of publishers respectively.

Let’s talk about ad density for a second. 

Ad density, according to the Coalition for Better Ads, is about density rather than quantity. This means that the number of ads served on your site should be relative to the length of your content.

If your content is short, you’ll serve fewer ads. If your content is longer, you’ll serve ads proportionate to the length of your content.

Our recommended ad density setting for Mobile is 28% (20% on Desktop). The CBA recommends a 70/30 split between content and ads, and our recommendation for Desktop in-content ads is actually lower than CBA suggestions because it provides a better user experience.

No matter which frequency you choose in your Mediavine Dashboard, we will never run more ads than your content can support, on Desktop or Mobile.

Your readers and your revenue are our top priorities.

So, What Do We Do Now?

In short, nothing drastic.

It’s Q4.

We’re still learning about the HCU and monitoring its impact, just like the rest of the industry. Plus, there are two additional updates running concurrently right now.

Now is not the time to adjust ad density settings because ad spend is ramping up. And when you change something like the number of ads you’re serving — especially if you’re lowering your density because someone without data claims too many ads are the cause of your troubles — you’re hurting your earning potential from all the other traffic sources out there.

Google’s blog on what constitutes helpful content and how to evaluate your site is a great reference, but it requires evaluating your content objectively. Sometimes, we’re just too close to our own content to be able to evaluate it effectively.

This is where content optimization tools or mastermind peer groups can come in handy. Letting a tool or another human you trust give you unbiased feedback can eliminate some of the difficulties we experience when editing and evaluating our own content.

When updates like this happen and your site is impacted, it’s natural to want to know why. No one wants to see their hard work vanish overnight. 

Unless Google itself says “we hit your site,” and provides a reason, a lot of the evidence being presented for why some sites were impacted is anecdotal, reminding me of an important phrase in statistics: correlation does not imply causation.

We have a dedicated team who will continue to monitor and diagnose the HCU and other updates, and we’re looking deep into our data to see if there are any definitive commonalities across sites.

Update:

Regarding the exclusion of sites making under $100/week in gross revenue before the HCU: That $100 gross revenue per week refers only to revenue coming from Google traffic. Some sites are absolutely killing it at Pinterest referral traffic with almost none coming from Google.

We initially reported this in a way that indicated we had excluded all low-revenue sites, which wasn’t the case, and we apologize for the error. 

In our analysis, we controlled for declines in Google referral traffic against referral traffic from other sources.

In trying to isolate whether Google’s HCU was the cause for a site’s traffic decline, we have to exclude sites whose traffic from other sources is either increasing/holding steady or declining slightly (up to -10%) during the same time frame. Otherwise, we’re just reporting on a site’s overall traffic decline and not the Helpful Content Update specifically.

In the spirit of being thorough, we ran the data again and adjusted for publishers whose referral traffic from other sources was declining but whose rate of HCU traffic decline was at least two times greater.

That brought us to 795 publishers with a negative traffic impact. We also examined the way that Google Discover traffic reports through GA4 and found another three publishers impacted, which brings us to a total of 798, or 7.7% of Mediavine publishers.

Next steps for us? Looking at commonalities to try and determine connecting factors, keeping in mind that correlation does not mean causation.

The July 1, 2023 Google Analytics 4 migration from Universal Analytics has turned the world a bit upside down for publishers across the industry, and the effects reach beyond checking daily metrics.

Many publishers are struggling to measure year-over-year performance, learning new definitions for common terms (looking at you, Sessions and Engaged Sessions), and trying to figure out how to update and reconnect tools.

The experts at Mediavine are working alongside you to navigate this bumpy change! In this post, I’m breaking down:

• More about your earnings
• RPM
• How all of it is — and isn’t — affected by Google Analytics 4

Will GA4 Affect RPM?

GA4 will certainly affect RPM, but not in the way you might think!

To understand how your RPM will be affected, let’s first revisit RPM.

We’ve written several articles on RPM over the years, but it’s been a while, and new analytics are always a great excuse to review. 

RPM (revenue per mille) is a unit of measurement. Revenue per 1000 is what we are measuring with RPM, but from there, we have to establish per 1000 of what exactly. 

• Mediavine’s Dashboard default view gives you Session RPM or revenue per 1000 Sessions. 
• You can toggle for Page RPM, which calculates revenue per 1000 Pageviews. 
• If you want to calculate RPM for Engaged Sessions (a new GA4 metric I mentioned in the update about our Application Process) or for Users, you can plug these metrics in as traffic in the equation below. 

At its core, RPM is just a basic math equation: 

Revenue / Traffic * 1000 = RPM

What Does RPM Mean for You?

Many of you already know these details, but some of the following questions are common, even with experienced publishers. Let’s break it down and connect more of the dots.

➡️ If RPM goes up, that does not necessarily mean you made more money. 

➡️ If RPM goes down, that does not mean you made less. 

➡️ If RPM is displaying as $0 or appears as a flat line on the graph, that does not have to mean your money has flatlined as well. 

We know this can be confusing. It seems like high RPM and/or higher traffic should equal more revenue, but that’s not always the case. We’ve got a handy help article about why RPM may drop during a traffic spike that dives into more detail. But the bottom line is this:

RPM is only a useful unit of measurement when the revenue and traffic parts of the equation are correct. 

In the context of GA4, Google is now calculating Sessions differently. Here’s how:

• GA4 Sessions: A session initiates when a user either opens your app in the foreground or views a page or screen and no session is currently active (e.g., their previous session has timed out). By default, a session ends (times out) after 30 minutes of user inactivity. 

• Universal Analytics Sessions: Each time a user comes to your site. An additional session is only created for a user if they visit another page after 30 minutes of inactivity or come back to visit from a different source or referrer.

Since the Traffic part of the RPM equation is slightly different in UA vs GA4, RPM will also be somewhat different.

For some publishers, sessions are tracking a bit lower in GA4. In that case, RPM will be higher because we are dividing revenue by fewer sessions.

If your sessions in GA4 are skewing higher than in Universal Analytics, RPM will be lower because you are dividing revenue by a higher number.

If RPM Goes Up, Why Don’t You Necessarily Make More Money?

Let me explain why you don’t always make more money if your RPM increases.

Example #1: If your developer accidentally deletes your analytics code halfway through the day, and it is only replaced halfway through the next day, you’ll have two days that only report half of your traffic. All that traffic still existed — readers were still visiting your site like normal from various sources — but Google Analytics wasn’t recording it.

In this example, you’d divide revenue from ALL the traffic by half of the traffic that was tracked. As a result, RPM will look artificially high for both of those days. However, the amount of revenue stays the same regardless of what Google Analytics is tracking, because your revenue isn’t determined by Google Analytics.

Example #2: The same logic applies if RPM drops after accidentally doubling a tracking script. You didn’t receive double the traffic by tracking it twice. In this scenario, you’d divide revenue by 2X the traffic that visited your site, so RPM will look artificially low.

But Wait. How Do Advertisers Know How To Spend If I’m Not Tracking My Traffic?!

The reason is simple: Advertisers aren’t connected to your Google Analytics. 

We connect your Mediavine Dashboard to analytics so we can combine your data with details about your ad revenue pulled from the ad server. The ad server is where all of the money details live.

If you never installed analytics, advertisers could still spend on your traffic. You just wouldn’t have a way to gain meaningful information about your revenue.

You can learn about S2S (Server-to-Server) and real-time bidding in detail here, but basically, a reader visits your site in their browser on a computer or mobile device, the browser talks to a server, then the server talks to the ad exchanges.

That browser action is why we care about cookies 🍪, the impending demise of those cookies, and why we built Grow.

Reminder: Use Grow now if you aren’t already! 2024 is just around the bend!

Establishing a New Normal

What will a new normal look like? This is where the real challenge lies. 

Traffic varies day to day, month to month and year over year, so it can be challenging to gauge if today was much higher than yesterday, especially when we’re tracking a slightly altered version of today compared to yesterday. 

If you started collecting GA4 data alongside Universal Analytics data prior to July 1, 2023, you can select a date range in GA4 and then select that same date range in UA. Compare the sessions reported and then calculate the percentage of difference. 

I sampled three months from my site and found that GA4 is reporting higher than UA by between 2.8 – 6.5%. If you recalculate RPM using the formula (Revenue / Traffic * 1000 = RPM), the percentage of change should match up with the difference you just tracked in traffic. 

Using your historical data will help you compare and find opportunities as you go forward. But don’t fret if you were a late adopter or waited until the July 1 deadline to make the switch. Just know where you made the switch and think of it as a reset. 

Make the most of the shift by setting a few new measurement goals, like increasing your engaged sessions and email subscribers.  

If you’re interested to see how 2023 revenue is tracking for publishers across Mediavine, check out our latest installment of Behind the Numbers With Brad: Planning for Q3 and Q4. 

As always, we’ll figure this out together.

The switch from Universal Analytics to GA4 is well underway, and as we all learn about these new metrics, adjust perspectives on how we measure success on our sites, and recalibrate to account for the differences (we see you lovers of real-time analytics), here at Mediavine we also have to shift our application process a bit. 

All of the components that we look for in a quality site are still the same, but in GA4, some of the terminology has changed. So! I’m here today to walk you through what makes a great Mediavine site and to help demystify some of the new metrics.

Understanding how readers use your content (all details you can glean from analytics) is the best way to ensure the sustainable growth and success of your business. 

What Mediavine looks for in an application

It’s widely known that Mediavine has what is perhaps the most stringent vetting process around. This keeps our traffic quality high and is one of the reasons that Mediavine boasts higher CPMs than all of our competitors.

Advertisers know that spending on Mediavine sites is a great return on their investment, and maintaining that reputation starts with our Application and Onboarding process. 

Minimum Traffic Requirements

The minimum traffic required to apply for Mediavine Ad Management is 50,000 sessions per month, based on the last 30 days. This hasn’t changed. 

What has changed is where you find that metric.

When you log in to your GA4 account, this is the default view:

To find sessions, toggle open the Event Count metric and select Sessions from that menu. Then you can adjust the date range to reflect the last 30 days.

Are we evaluating Sessions? Or Engaged Sessions? 

GA4 offers two separate Sessions metrics: Sessions and Engaged Sessions. Both offer really valuable insights into how readers are using your site.

 In Analytics, a Session initiates when a user either opens your app in the foreground or views a page or screen and no session is currently active (e.g. their previous session has timed out).

An Engaged Session is a session that lasts longer than 10 seconds, has a conversion event, or has at least 2 pageviews or screenviews.

When you apply to Mediavine using your GA4 account, we are evaluating Sessions because it most closely mirrors sessions in Universal Analytics. 

A Session in Universal Analytics is defined by Google as a group of user interactions with your website that take place within a given time frame. A single session can contain multiple page views, events, social interactions and e-commerce transactions. 

GA4 defines a Session event the same way, and just like Universal Analytics, a session times out after 30 minutes of inactivity. 

Traffic is just one piece of the puzzle. 

A minimum of 50,000 sessions per month is the very first, most basic requirement. 

Remember the quality I mentioned? Let’s take a look at what that means for a site applying to Mediavine.

• Original Content! This is an important place to start. We are looking for well-written, well-organized content that is original and belongs to the site owner. Original photography or licensed stock photos are great additions. The internet is a visual place. We DO check for proper attribution and permissions for images. 
• Brand safe content. Our SVP of Sales and Revenue wrote a great post a few years ago about keyword anti-targeting. Advertisers are looking for family-friendly content because serving their ads on a website is effectively like product placement. We steer clear of content related to violence, extremism, and topics that are overly salacious, just to give a few examples. 
• Users and Pageviews. The numbers have to make sense. We are looking at the number of users and pageviews relative to sessions. Most lifestyle websites average about 1.2 pages per session. Depending on the content, that metric can move up or down, but if it’s too high that can indicate a reliance on things like slideshows that artificially inflate pageviews and are not great for engagement (or your wallet). 
• Bounce Rate. This is one metric that does calculate a little differently in GA4. In Universal analytics, a typical bounce rate for a lifestyle site is between 75-90%, especially with a high volume of search traffic. In GA4, bounce rate is the inverse of that engaged sessions metric. So if your bounce rate is 66%, engaged sessions are at 34%.
• Countries of Origin. Traffic from the US monetizes at a higher rate than traffic from other countries. But that is not to say that we don’t accept international traffic — WE DO! What we are looking for here is a breakdown in locations that will be good for revenue. Canada, the UK, Australia and the Netherlands all also tend to be high on the list for advertiser spend. If you are already a Mediavine publisher you can check out the Country report in your dashboard to get an idea about how advertisers spend in different countries around the world. Our focus here is to be sure that any site we launch is actually going to be happy with revenue and RPM. 
• Average Session Duration. Are readers engaged with the content? If not can we see an easy answer as to why? 
• Devices and Browsers. Most sites are approximately 75%+ mobile traffic. That can vary a little based on niche, so we are checking to be sure that the details here all work together with the other puzzle pieces and that the dots all connect related to the content on the site. 
• Varied traffic sources. The majority of Mediavine sites have Google as their top traffic source. Y’all are winning the SEO game. High Google traffic is not required though; social traffic, as well as direct and email traffic, are all also valuable sources. Having a variety of sources lets us know more about how you have built your readership, and it allows us to follow your readers from the beginning of their journey so that we can offer insights into how we can better monetize a site. 
• Top posts and pages. Having an eye on the most-trafficked content on your site is always going to help you monetize more effectively. We read your posts! I always know we have a great application when I start bookmarking things or when I get lost in the reading and forget that I am supposed to be doing work.

None of these puzzle pieces will yield an approved application on their own. Our Applications and Onboarding Team are experts in figuring out how readers use a site, how well that site will perform, and if the traffic is sustainable. 

GA4 doesn’t change any of this, and as we all move forward into this new world of metrics and reporting, we look forward to offering you all more insights into how to use your analytics to continue to grow your business.

Video ads are an important part of the monetization strategy for your website, but the options can be endless, and some options can completely destroy your user’s experience. So what’s the best way to increase revenue and RPM, while also maintaining an experience on your site that draws the reader in and allows for more engagement and time on page?

We have answers!

“I want to level up my video strategy.”

“What is the best practice for embedding videos?”

“How should I incorporate YouTube videos in my blog posts?”

“I’d like to add video to increase RPM but I am worried about user experience.”

Hi, my name is Heather Tullos 👋 and y’all probably know me because a big part of my job is moderating our Facebook groups, and these are all recent questions you’ve asked about video. I’m here today to provide answers, starting with an announcement we’ve spent the last month or so preparing you for:

Today is the day we kill sunset say goodbye to autoplay

We’ve killed off a few things here at Mediavine — pageviews in favor of sessions, AMP in favor of Core Web Vitals, and now we are taking out autoplay instream videos in favor of a better user experience.

In 2021, we rolled out the Universal Player as a solution for site owners who want to earn those high video CPMs, but might not have yet ventured into producing their own video content. What we learned almost immediately is that the Universal Player, when set up properly, very quickly outpaces our previously recommended autoplay instream settings.

Before we get too far into the details, let’s cover a few important definitions

I’ll be mentioning instream and outstream a lot in this article, so let me explain what those terms mean.

• Instream Video = Video ads that play before or during video content
• Outstream Video = Video ads served outside a traditional video player

Autoplayed video content with an ad that runs before it has always been classified as instream.

The point of an autoplayed instream ad has never been user experience — it was always just money, as recently discussed by our CEO, Eric Hochberger in AdWeek..

New IAB Standards

The Interactive Advertising Bureau (IAB, the organization that develops industry standards) has set forth guidelines that now require that these autoplayed muted ads are classified as outstream. This shift in standards is based on establishing a better user experience.

So let’s talk about what that means for you. Because while we all obviously want a better experience for our readers, we also need a paycheck.

More Money. More Time on Page. Better UX.

Effective today, we are killing autoplay. This means that if you were running autoplayed instream video on your site, those videos will no longer play automatically for the reader. They will be click-to-play.

The Universal Player — that small outstream player that we rolled out in 2021 — is going to do the heavy lifting where your video revenue is concerned. Here’s how that works:

When we look at the numbers, most autoplayed instream videos average fewer than 1 impression per session. In fact, on most sites that number drops to 0.5 impressions per pageview. That means that readers are closing out the autoplayed instream player before they ever even see the video content you created.

In contrast though, the Universal Player averages 2-4 impressions per pageview. It’s 10 px smaller and tucked out of the way in that recommended Bottom Left position on mobile, so it just chugs along, refreshing every 30 seconds as long as there’s outstream inventory.

It works if a reader carefully scrolls your content and reads every word, but it also works if they scroll really quickly or use a jump link to hop down the page. Since it DOES work so well, there are not actually that many Mediavine sites that are still relying on an autoplay strategy.

For those of you that will just be making the shift today, I wanted to be sure to provide answers to common questions, and to help you be sure you are set up for video success.

WHAT ABOUT ALL OF MY VIDEOS?!!!

If you’ve spent time and money creating useful video companions to your blog posts, you are still in great shape. Video is an excellent way for readers to learn more and have an additional visual companion to the content on your site. It speaks to the different ways people like to learn, and when they choose to click on a video they are engaged.

Engaged readers pay you more in the form of higher CPMs and longer time on a site. Not only is that reader seeing the ad in its entirety, but they are also sticking around for your content instead of looking for ways to exit.

Video is also still great for SEO. Google loves it, and videos can give you additional opportunities for how you appear in search results.

And let’s not forget all the ways you can creatively reappropriate video content across social media.

How to set up Mediavine video for the highest revenue, RPM and ROI

Since we are removing Autoplay Settings for you today, including Featured Video, there are fewer buttons for you to worry about, and this set up is really simple.

Check on Your Optimize Placement Settings

Without videos being autoplayed, these settings are no longer needed. However, before you disable them permanently, be sure that your videos are embedded in the correct place in your posts. Find out more details about correctly embed, because if a video is NOT autoplaying, we want to be sure that it IS where your reader can find it and click to play.

Optimize placement settings can be important if you have not been consistent with video embeds across your site (maybe they are in the first screenview, maybe they are not in the easiest place to see in most posts, etc).

To check your video placements, disable Optimize Placement on desktop and mobile, and save. Then go to the Top Videos section of your dashboard and click View on a few of the top videos. Scroll through the post or use your jump links. If the videos are where they should be, you can leave Optimize Placement off.

Embed Your Videos

This one is important! If you have video content it should be embedded into the corresponding blog post.

➡️ If your post has a how-to or recipe card, be sure to embed IN THE CARD.

➡️ If your post does not have a card, embed in the blog post.

❗️ It is not necessary (and is not best practice) to embed in both the card and the post. If you have a card and embed there, that is all you need to do as far as video schema goes.

Be sure to fill out your video details.

This is still important for advertiser spend as well as SEO. Advertisers don’t watch your videos so they use the associated URL and keywords etc to understand more about how to spend.

If you have YouTube Videos…

You can embed your YouTube videos (the recommendations for embeds are the same! In the card if you have one, in the post if you don’t) if that’s your preference, but there are a few things you will want to keep in mind.

• If a reader clicks-to-play your YouTube video, you won’t be able to monetize that view on your blog (you CAN with the Mediavine embed).
• YouTube embeds can be slow. They often turn up in pagespeed reports so it’s important to keep an eye out.
• If you want to grow your YouTube presence via your website, you can also connect your YouTube channel via Mediavine video.

Video has changed so many times over the years, and it will likely change again (but hopefully not for awhile). This current strategy based on IAB standards and best practices is really striking an excellent balance and we think it opens up so many more creative opportunities for how to approach video content on your site. Have questions? You know where to find us!

You can bookmark this help article and are always welcome to reach out!

What’s My Goal?

• REVENUE / TRAFFIC x 1000

• IMPRESSIONS x CPM / 1000

• What does weekend support look like? Is it just “limited” support on the weekend? What does that mean? Weekends tend to be the highest traffic days for sites in any niche, but weekends are also when bloggers get a TON of work done. So will someone be there to help you in case you need it?
• How will you support me? Do you offer email support? Phone support? Do humans work there and will they answer me?

• Can I see the number of impressions served? How about viewability? CPM? For each unit?
• Is the data offered net or gross? Are the earnings in my dashboard finalized? (is that what I will actually be paid?)
• Do you offer page-level reporting or country-level data?
• Do you offer viable solutions on things like CCPA and GDPR to help me with my data responsibilities in reference to ads?
• How will you help me use the data provided to grow my business?

How are You Going to Help Me Reach My Goals?

Knowing the “Unknowables”

SEO Resources from the Best in the Business

#PublisherGoals

Investing in the Future, and the Publishing Ecosystem

We Are Independently Owned & Operated

Peace of Mind

What is a Heat Map?

How Heat Maps Can Help You Optimize

What I Learned

How You Can Use Heat Maps to Optimize

Rebranding is bad for your business.

But I don’t feel like my name fits me anymore!

A post shared by New Kids On The Block (@nkotb) on Aug 24, 2017 at 2:16pm PDT

What about my ads?!

Long Term Effects