Garrett Sussman: Our next guest. This one’s fun. This one is, we’re getting a little little cultural, a little little Star Wars love action. I’m joined by Bianca Anderson.

Now Bianca is the manager of organic growth at hims and hers. In 2016, believe it or not, she ate an entire bag of 100 organic lollipops in under 48 hours, which I don’t think I could pull off. She’s double jointed. Cool little fun fact. And her favorite Star Wars character is Chewbacca because she’s, like, has this affinity because he’s kinda tall. She’s kinda tall. It’s like a, you know, matching situation. So please give it up for Bianca Anderson, F Traffic: Prioritizing Conversions Over Vanity Metrics.

Bianca Anderson: Greetings. Greetings. Thank you so much for that warm welcome. And thank you so much for joining my session. The title of this track probably gives it away, but I, as an SEO, have some qualms with how we talk about success, particularly with how we talk about traffic. I don’t need to say it, but I’ll say it anyway. SEO is changing. We’re in a very transitional state, as yesterday’s talks so eloquently illustrated. AI overviews rapidly declining click through rates for informational content, especially, etc., etc. Yet I still think we often trend towards treating traffic like a leading metric of success. We’re quick to celebrate when it spikes, and we’re also quick to wallow and spiral when it dips.

But traffic does not tell even a fraction of the story, especially now. And so in this talk, I’m going to talk to you and walk you through what it looks like to deprioritize traffic as the center of your SEO universe, and instead prioritize the content on your site that is giving you business impact. To do that, I’m gonna walk through a specific framework called the heavy hitters framework. It’s one that I started to use during my time at HubSpot. It’s a way to identify the content that drives business impact at scale. But just as importantly, I think it’s also a way to convey to your team, to stakeholders, to leadership, that you’re also prioritizing value over volume.

But first, more facts about me. Hello again. Name’s Bianca. I’ve been in SEO since 2019. Started off at the Adler Planetarium in Chicago, which is where I currently reside. Went agency side for a bit, then went on to be a lead strategist, SEO strategist on the English blog at HubSpot for two and a half years. Currently, I’m a manager of organic growth at hims and hers, a telehealth tech company. You might have seen some of our commercials. Side note, I’m a huge space and sci fi fan, so perhaps a little self indulgently, this talk is space themed. 

In terms of what you can expect from this talk, we’ll first dive into the Vanity Metrics black hole, talk about how I think traffic drops trigger too much panic and why that mindset is now broken. Then we’ll go into a Copernican shift, so why should we pivot away from this way of thinking. And then we’ll go into the actual heavy hitter process, so what it looks like to identify the content that’s actually driving business results at scale. Then we’ll go into some use cases. So how can you monitor, optimize, and operationalize your heavy hitters once you’ve identified them? And then last but not least, some closing thoughts.

All right. First, the Vanity Metrics black hole. So as I said, sci fi fan lore. I want to give you some lore as to why I decided to choose this subject for the talk. And in order to do that, I’m going to get a little meta and talk about some talks I did this past fall. So in fall 2024, I spoke at a couple of conferences, Brighton SEO San Diego and Women in Tech SEO Philly. During these talks, I centered mostly on the work I did at HubSpot helping to lead the creation of an EEAT playbook, what that entailed, what that impacted in terms of our workflows, etc., etc. But there was something that was hidden in the final slides of those talks, kind of in retrospect maybe as a throwaway. But it was the need to prioritize impact over traffic volume. 

Not longer after these talks happened, HubSpot’s blog traffic took a visible hit and went viral. And I watched as the reactions unfolded in real time. Dun dun dun. Mike shared a slight different version of this graph yesterday, but, this is the one that made the rounds on the Internet. And understandably so, the industry took note. There were a lot of articles that were written and SEOs weighed in. So, I want to emphasize that a lot of the discourse that surrounded this drop I thought was very productive. Really good questions and whatnot were raised about content quality and the algorithmic shifts and the impact of AI overviews.

But as this graph started to make the rounds across the Internet, I started to see a different type of discourse emerge, one that was a little bit less productive and a little bit more smug.So what I did was I pulled some of my favorite quotes from various sources of the web Twitter, Reddit, LinkedIn. But to protect the innocent and to also remain on theme, I have given these SEOs alien alias names. So keep that in mind.

First, we have Zorblat, who said, watching the HubSpot traffic drop like a rock. And gee, can’t imagine why. Maybe flooding the internet with irrelevant content wasn’t the master stroke they thought it was. Then we have Snorgus, who said, my only surprise about the HubSpot blog traffic drop is dot dot dot, it didn’t happen sooner. Blargo says, good. I now consciously avoid clicking on anything from them. And last but not least, we have Quizzo, my absolute favorite one. You’re gonna be seeing him throughout this talk. Straight to the point, HubSpot is shit. End quote.

Okay. I have some problems with these takes, but let’s get this out of the way. Yes. I used to work on the blog, extensively. I know folks that work on the blog to this day. That’s sung a little bit. You know, I’m human, perhaps. Some ego’s involved here. But I promise you this isn’t propaganda. I’m not here to defend HubSpot. The reason why I share with you these pieces of discourse is because I think they highlight a larger issue, perhaps. And that is that we still equate visibility with value. What this discourse shows is that when this traffic dropped, yes, it was a very notable one, but there was this rush to equate this traffic drop with all encompassing strategic failure on HubSpot’s end.

And what that tells me is that we’re not quite ready to evaluate SEO success in this post-AI world. So what do we do? Well, Copernican shift is what I’m proposing. Alright. Quick 8th grade-level science refresher. For centuries, people believed that the Earth was the center of the universe until my guy, Nicholas Copernicus over here, Nicholas Copernicus proved that everything actually revolved around the sun in 1543. Great year.

I think SEO is undergoing a similar shift. No one here believes traffic is the only metric that matters, but I do think for years it has been one of the easier ones to measure and report on success. Even when we know conversions and other KPIs are more important, traffic still often drives stakeholder expectations and growth models. But that reality is changing, as we saw so clearly yesterday. AI Overviews, SERP shifts, declining visibility for informational content, this all means that exponential traffic growth models that we are relying on or that we have relied on are just not sustainable. And so just as Copernicus redefined how we see the universe, we need to rethink what SEO success looks like in this new landscape and what we center it on.

But don’t just take it from me. Here are some stats. This is from Ryan Law of Ahrefs. 99.2% of the keywords that trigger AIOs are informational in their intent. And according to Melissa Rae Brown of Moz, 80% of all keywords are considered to be informational and thus are vulnerable to AIOs appearing on their SERPs. And then this is from Tracy McDonald of SEER Interactive. Even when you get that coveted spot, that citation in an AI Overview, over half of organic clicks disappear when they’re present, when an AIO is present.

So what I think this means for SEO is a few things. One, the green space, pun intended, for available traffic is shrinking. AIOs are absorbing informational clicks. Middle of the funnel and bottom of the funnel content is now the core SEO battlegrounds. And old models of exponential traffic growth are not going to hold. So I think as SEOs, we have to adjust our expectations, as these organic opportunities shrink.

But pause. I want to be really clear about what I’m saying here, and what I’m not saying. So what I’m not saying is that informational and top of the funnel content is completely obsolete and that there’s no merit in ever targeting it. I think there’s still merit, both within the context of SEO and also within the context of working cross functionally with your CRM and your social teams, etc. I’m also not saying that because AIOs exist, we should completely adopt a fully stoic view on traffic dips and never investigate as to why they’re happening. I’m not saying that. 

But what I am saying is this. We are undeniably in a period of systemic volatility. And in the midst of that, if and when traffic drops, I do not think your first question should be, what did we do wrong strategically? I think your first question in turn should be, or in its replacement should be, what matters and is that being impacted? 

And so that brings us to the heavy hitter framework, which sets out to do just that. So what are heavy hitters? Heavy hitters are the top performing 10% to 20% of content in a given library. This is identified through a formula that evaluates both traffic and conversions with a much larger emphasis on conversions. Wink, wink.

So in my time as an SEO, all of the sites I’ve worked on have followed Pareto or Pareto’s principle, not quite sure how to pronounce that. But 20% of the effort generates 80% of the results. So this framework tries to not mimic that reality and really focus on that 20%.

In terms of how you identify your heavy hitters, there’s four steps. Step one, you normalize traffic and conversion data for each URL. From there, you utilize a score that considers this normalized data while adding greater weight to conversions to reflect their higher importance. Step three, you calculate the percentiles of your weighted scores. And then step four, you assign tiers based on percentiles that select the top ten percent to twenty percent as your Library’s heavy hitters.

Alright. So we’re gonna walk through each step of this process. And to do so, we’re gonna use the fictional company I made called Nebula Health. They are a health company that creates protein packed, nutritious meals for astronauts in space to help prevent muscle deterioration in zero gravity. So astronauts like Katy Perry, for example. Yeah. Yeah. Okay. And their slogan is, Stay one giant leap ahead of health issues. So, yeah. All right. 

So we’re going to do step one, normalizing data. So like our solar system, traffic and conversions are on completely different scales. So we need a fair way to measure them against one another. So we normalize traffic and conversions so that all pages are scaled equally on a scale of 0 to 100. The formula is as follows: Traffic of URL divided by max traffic in that data set times 100. Rinse and repeat. Do it the same for submissions or whatever your key conversion metric is. So here we have a little small data set from the Nebula Health website.

I’m looking at it. This is Q1 traffic and Q1 submissions. We’re going to start by normalizing Q1 traffic. I see here that the highest single URL, single traffic for a given URL in Q1 was for what astronauts eat in space. So that’s going to be our baseline and what we use to normalize the traffic. So all of the traffic is gonna be divided by that baseline of 24,000, and then we multiply it by 100. And then we’re left with normalized traffic that’s on a scale of 0 to 100, with what astronauts eating in space being the cap at 100 because it’s the URL that had the largest traffic in Q1.

Same thing with submissions, looking at the same data set. This bottom URL here, Nutrition Console before launch, had the highest number of submissions in Q1. So we’re going to take that and divide every single URL by that as the benchmark. And we are left with normalized traffic on a scale of 0 to 100 for both oh, sorry. We’re left with normalized numbers for Q1 for traffic and submissions, all on a scale of 0 to 100.

Step two is calculating weighted scores. So since conversions matter more than raw visits, we’re going to weigh them heavier. So this is a formula that I like to use. It’s as follows: Normalized traffic times 0.3 + normalized submissions times 0.7. These weights can be tweaked as you see fit as long as they add up to 1. So if you wanted to weigh traffic a little bit more heavily, you could do normalized traffic times 0.4 + normalized submissions times 0.6. So we’re just going to plug that in.

So we’re plugging in the normalized traffic times the selected weight of 0.3 + the normalized submissions times the selected weight of 0.7. Alright. And then you’re left with a weighted score that considers traffic and submissions with a much heavier weight on submissions.

Now we’re going to calculate percentiles using this weighted score. So percentiles are going to tell us how each page stacks up against others. If a page is in the 19th percentile, for example, it outperforms 90% of other pages. So higher percentiles equal higher business impact. So to do this, I just plug in a percent rank formula using Google Sheets, to get the percentiles of each individual weighted score. And here we have the percentiles.

So now we’re going to use these percentiles to tier our content. So, this is going to help us segment URLs based on their relative impact and not just raw numbers. This is a tiering mapping that I like to use. Tier 0 at the very top is following Pareto’s principle, 80th to 100th percentile, top 20%. Then we have tier one strong performers, 60th to 80th percentile, next 20%. Tier two middle level mid level performers, 30th to 60th percentile, middle 30%. Tier three are the low performers, that bottom 30%.

So here are the tiers, paired with the percentiles using that mapping I just showed you. And as you can see here, we have a couple of heavy hitters that have emerged in this data set. So let’s bring back the original q one and submission numbers to see how they fare with the final tiers. 

So as you can see here, even though what astronauts eat in space had the highest traffic numbers, it didn’t make the cut as a tier 0. On the flip side, Nutrition Console before lunch had the lowest traffic in Q1, but notable submissions, a disproportionate amount in this data set, so it’s a tier 0 URL.  

Okay. So now that we have identified our heavy hitters, I want to walk you guys through some use cases, what it looks like to operationalize these in your workflows. So use case number one, one of my favorites, is core systems monitoring. So when search volatility hits, like during a Google algorithm update, your heavy hitter URLs can be an excellent first place to look as these pages drive outsized value, so even small changes can have a big impact. So my recommendation is to create a dedicated header report, make this a reoccurring report that isolates the performance of these URLs.

So look at traffic submissions, rankings, all the works for these tier zero heavy hitters. Then you can compare performance against site wide trends during volatility windows or performance steps. So benefits of the heavy hitter report. Early detection, so you can quickly see if your most important pages are being impacted. Pattern recognition, understand that these URLs mirror or diverge from broader site trends, i.e., top of funnel content versus bottom funnel content. How are they performing? And then more confident reporting in general. So when leadership asks, how bad is it, you’ll have an answer pretty quickly.

Use case number two is a natural progression of use case number one, dire updates. So now that we’ve identified our heavy hitters, we’ll need to monitor them at least monthly due to their outsized impact on conversions. A dire update is triggered when a heavy hitter shows meaningful decline that may indicate SEO or SERP related issues, etcetera. Dire update criteria that I’ve used for this process. It’s basically a checklist. So if the heavy hitter URL hits these marks, then you can rest assured that it is indeed dire and should be expedited for optimizations with your editorial teams. Some of this criteria includes, of course, a month over month submissions dip that is more than x%. For that x%, recommends looking at historical data to see what you’re comfortable with as a benchmark for being the cap or the yeah, the benchmark for what you want for what is bad for dips. Yeah.

You can look at rankings that are declining for one or more of the top twenty traffic driving keywords. I used Ahrefs for that. Of course, verifying that the post is still indexed by Google, that the drop is not due to declining search demand again, you can use Ahrefs or Google Trends and that the drop is not due to a temporary spike from a recent republish.

Because these are dire and they’re so important, I think speed and efficiency are so crucial. And one way that I like to ensure that is by trying to assign dip buckets to these, dire updates, to give a language to the editorial team so that they can quickly know what’s wrong with the URL. So this is a few examples of DIP buckets. It could be SERP intent misalignment. The resulting optimization would be to align the content with the updated search intent for those target keywords. Perhaps a SERP feature was added. Optimization would be to try to optimize for that new SERP feature if possible.

Cannibalization could be another dip bucket. So consolidating or differentiating overlapping content to reduce competition. Lack of media. So you can send that in a different pipeline to try and get images or whatever onto the article to make it more competitive and user friendly. Not exhaustive, but just to give you an idea of some of the buckets that you can use to try and make this process as expedited and efficient as possible.

So, yes, why use performance dip buckets? When something breaks on the International Space Station, astronauts don’t guess, they get Katy Perry. Okay. Sorry. I I was I don’t wanna I don’t wanna overdo that joke. That was the last one, I promise. When something breaks on the International Space Station, astronauts don’t guess. They isolate the issue and fix the right components. So DIP buckets do the same for content. They turn performance drops into targeted fixable problems that your editorial teams can act on even faster. So instead of just saying this post is down 15% month over month, you can give them a specific issue, like a serpent issue. And, your editorial team now has a starting point, not just a red flag.

Another one of the use cases I like for heavy hitters in this framework is identifying your brown dwarfs. So one of my favorite celestial objects are brown dwarfs. These are objects that are too big to be a planet, but not quite massive enough to elicit, to start nuclear fusion and have the brown dwarf become a star. But in this use case, brown dwarfs are your tier 1 to 3 pages that have high CVR. So these are pages with high conversion potential, but they don’t have a ton of traffic coming to them. So they might be good opportunities to optimize and become your next heavy hitter.

So I’m bringing back the Q1 submissions and traffic data for Nebula Health. And we’re going to do a quick CVR percentile formula for the CVR of these URLs, again, using that percent rank formula in Google Sheets that I shared earlier. So now we have the CVR percentile of these URLs. This is the mapping I like to use for what I like to call CVR rank. Again, we’re following Pareto’s principle. 80th to 100th percentile high performing CVR, 50th to 80th percentile standard CVR, less than 50 percentile underperforming CVR.

So looking at this data set, I see there’s a Tier two URL that has a CVR percentile of 67%, which puts it in the standard CVR range. It’s Tier two. There could be an opportunity here to better make use of that strong stronger CVR. So I’m in Ahrefs looking at this very real, spread of nebulahealth dot com slash blog slash prevent muscle loss in 0G. And I see that How to Prevent Muscle Loss in 0G is currently ranking in position number seven. So perhaps it’s worthwhile to optimize, try to get that in the top three positions, and then, take advantage of that higher CVR in the process and get more people to the page. So even though the CVR is falling within the standard page, this page is still significantly outperforming others in its tier of tier 2. It’s converting well, just not at scale. And so it may just be underrated because of low not because of low value, but because of low visibility. So a small boost in visibility could be an optimization or a revamp, could elevate it into a tier 1 or tier 0.

Another use case for this framework is deep space content audits. So when your content is tiered by business impact, you gain clarity. And so utilizing these tiers is going to help you pinpoint which URLs to defend, which ones to throw out, and which ones are your brown dwarfs.

So, here’s a table that maps out the tiering structure against what it could mean within the context of a content audit. Tier 0 URLs, that’s your bread and butter. They contribute to probably 80% of your business’s pipeline. So we’re going to monitor, protect, keep fresh, treat these as VIPs because they are. Rising suns, the tier 1, strong performance, optimize, promote, test. Tier 2, stable satellites, moderate value, consider updates or retargeting. We get to tier 3, fading moons, the bottom performers. This is where you want to get a closer look and see if it’s time to prune, to consolidate, to repurpose, recycle, or retire. And then there’s the brown dwarfs, which are their own thing. These are URLs that are not a tier 0. But they have a high CVR or a pretty good CVR. And so it could be worthwhile to search for them, optimize, link to, etcetera, to try and get them higher ranked in this tiering system.

Ah, closing thoughts. Protect what performs. Not all pages are created equal. Your heavy hitters should be monitored like the core infrastructure because when they drop, so does your pipeline. So guard them with the urgency that they deserve. And lead the shift. Again, we’re in a very transient transitional time. Traffic is probably going to drop. And like Mike said, the traffic that has dropped, I don’t think it’s coming back. And so for SEOs, I don’t think this is just about execution. It’s about education. We set the tone for what success means, what gets measured, and what gets resourced. And I hope that this heavy hitter framework model is one of the many tools you could use to try to add to your arsenal to accomplish just that.

And lastly, please let’s all try to be a little less like Quizzle and a little bit more like Nicholas Copernicus, who on his deathbed famously said, fuck traffic and go get some conversions. Thank you.