Ryan McGonagill, VP of Content at Centerfield, breaks down how his team powers 10+ owned brands like business.com and security.org while keeping content simple, useful, and relentlessly user-first. We dig into why personalization beats flashy interactives, and how to design for real decisions—not vanity metrics.
https://page2pod.com - What does it take to run a multi-brand content machine—and decide what not to publish? Ryan McGonagill, VP of Content at Centerfield, breaks down how his team powers 10+ owned brands like business.com and security.org while keeping content simple, useful, and relentlessly user-first. We dig into why personalization beats flashy interactives, and how to design for real decisions—not vanity metrics.
You’ll hear the refresh system that keeps nothing older than a year (and why ISP deal pages may need updates every ~2 weeks), plus how seasonality (think moving season) reshapes the calendar. Ryan also unpacks verticalized expertise, the team’s “gold standard” quality bar, and why user value beats dwell-time games every time.
On the PR side, we talk about crafting hooks that surprise and earn coverage, the DEI pronoun study that drew massive press (and even an Elon Musk reply), and how Ryan thinks about links vs. referral traffic. We close with a pragmatic take on AI, where human judgment is non-negotiable and LLMs assist without replacing expertise.
Finally, Ryan shares why content structure may matter even more as AI overviews evolve—and the single best link he’s ever earned.
🧠In This Episode
• Centerfield’s model: owning acquisition end-to-end and publishing across ~10 brands, from business.com to security.org.Â
• Personalization > “cool” interactives: give users simple, relevant tools (even a basic calculator).Â
• “User first” beats dwell-time hacks—ship what helps people decide, not what inflates metrics.Â
• The refresh playbook: nothing older than a year; ISP pages reviewed ~biweekly; plan around seasonality (e.g., moving May–Sep).Â
• Verticalized writers + editors and a non-negotiable “gold standard” (third-party + firsthand expertise, real photos).Â
• How to prioritize: fortify winners before chasing breadth; align SEO (traffic/rank) with content’s profit focus.Â
• PR that lands: lead with surprising data and media-worthy angles; why links > referral traffic from research.Â
• AI in the workflow: frequent LLM upgrades, but keep human touchpoints for judgment, reviews, and accuracy.
This episode is a tactical blueprint for scaling content that _actually helps people decide_—and earns the kind of links and trust that last.
📢 Subscribe for more expert insights on content marketing and SEO!
đź’¬ Comment below: What's your biggest challenge with content personalization or AI in your workflow?
đź”— Websites & Brands:
🧪 Notable Studies/Articles Referenced:
[00:00:00] What does it really take to keep a content operation running across 10 brands, dozens of verticals and a constantly shifting SEO landscape, and more importantly, how do you decide what not to publish? Brian McGonagill is the VP of content at Centerfield, a performance marketing company that owns and operates a suite of consumer review sites like business.com and security.org.
Brian's job is deceptively complex. He leads the team responsible for content that not only ranks and converts, but also holds up the journalistic standards. Brand scrutiny and increasingly the expectations of personalization at scale. This episode unpacks the mechanics of a high performing content team in 2025.
We get into how Ryan thinks about personalization, not as a buzzword, but as a design principle. We talk about when interactive activity works and when it's just fluff. We explore the role of AI and content workflows, how content gets refreshed, and what Ryan really looks for when hiring writers and editors.
The real story here is how editorial judgment. Business goals and user empathy intersect In a world where content is increasingly commoditized. Brian's [00:01:00] challenge is staying ahead while keeping things simple. He's not chasing dwell time metrics or novelty for his own sake. Instead, he's betting that the most enduring content is the kind that actually helps people make decisions, but in immediate environment flooded with AI generated noise.
Can that approach still win stick? This one's a masterclass in making content that actually works.
Welcome to another exciting episode of the Page two podcast, as always, and joined by Joe Devita, my partner at Crime at Moving Traffic Media. And today we're welcoming Ryan McGonagill to the show. So we're excited to talk content and I guess link building in a broad sense, but maybe to. Helped set the stage.
Can you give us a 10,000 foot view of Centerfield? I know my wife currently works there, just disclaimer, and I know when she initially was interviewing I was like, what is Centerfield? So help us understand the ecosystem that you are basically in [00:02:00] charge of from a content perspective.
Sure. So it's funny, you're wife Jamie is my partner in crime at Centerfield.
We work together super closely. So Centerfield is a customer acquisition company. We will handle everything from the. Let's say PPC ad on Google all the way through closing the deal for Fortune 50 clients. As part of that, we have a brand portfolio of about 10 different brands that we own and operate, and the verticals that we serve our clients in.
And that's what I work on. So for those brands, I'm responsible for anything content related. That would be editorial, video, social, e-commerce, community, you name it. And so that's where. I spend my day to day.
That's awesome. There's so much to dive in there. I think maybe we'll start a little bit before your role at Centerfield and so you studied music at Berkeley College of Music and we actually have an upcoming interview with another gal who, um, started her career in music.
Love to hear a little bit about how you. [00:03:00] Got from music to marketing and content?
Yeah, so I had a communications job for a couple years in college. It was a work study, writing blogs, onsite content, press releases, stuff like that. I actually remember being able to make changes to the university site with no prior approval, so I hope that they fix that.
That's probably not a good idea. But my first real job after school was a PR specialist for a content marketing agency in South Florida. I actually founded on Craigslist, believe it or not, and I thought, you know, I can probably fake my way through this until I learned how to do it. And three months later I ended up moving down there for this company and I was there for about five years.
Funny story. I actually started my freelance career off of Craigslist. I was looking to buy a condo, be magical
place. Yeah,
you really can. I was in San Diego at the time. I wanted to buy a condo and I was like, how can I make some extra money for a down payment? And found some PPC jobs. There was an agency that was looking to get health and here we are.
Is there anything about, [00:04:00] so music tends to be very mathematical, right? Just in, is there anything from that background that you've been able to translate to what you're doing in marketing or a lot of content analysis and things like that? Is there anything, any relationships there? I know Joe is also big into.
Starting instruments in music as well.
Well, I love that you said music is math, because I say that to people and they're like, I don't get, no, it's not. And I'm like, no, it really is though. So I went to school for piano, and let me preface this by saying I do not enjoy performing. I was not the best pianist in the world, but I think what set me apart was the emotion I was able to put into the music I was playing.
So I like to make stuff. Sound pretty, make people feel something. I'm not necessarily trying to conquer the most technical piece. In fact, if I had a niche, it would probably be breakup songs and power balance. But if you boil it down, making people feel something is the goal with content to an extent. I don't think that you can maximize your success in this field unless you can understand [00:05:00] and even empathize with your audience on a pretty.
Deep level. So when I say understand, I don't mean knowing who's searching what and at what volume. I really mean putting yourself in the user's shoes and saying, okay, if I am X person and I'm looking for Y, would I be satisfied with this piece of content?
I read something you posted years ago about producing content that makes people feel like they're a part of it, and you described the success you had with quizzes.
This was years ago. You posted, maybe you don't even remember you published this article, but I thought there was a really good insight there. If you can get someone to engage with con, like voting for something is a really easy thing for people to do and then they feel. Part of that content and it's easier for them to maybe share you the first person that I have read that wrote about that making people feel part of the article.
It's hard to do in day-to-day life, but a good message.
Thank you. Wow. That was a long time ago. We do very thorough research, Ryan, just so you [00:06:00] just, I think that. What that ties into is people getting their opinions validated. I have an opinion, I'm taking this poll, and I don't know a thousand people agree with me.
And so I feel backed up by that and I wanna share that because I have social proof. I think that's where that ties in. And I think I'll talk about personalization probably a lot throughout this conversation, but that ties in with the two.
Yeah. Let's dive into personalization right away. I feel like that is where, if you think about AI mode and even the chatGPT experiences, now that they have a little bit of memory behind them and other LLMs, right, the those, that seems to be the way that everything is moving, right?
There are so many data points that these systems know about you, and they're able to map that history with. Recommendations or feedback. How are you thinking about applying some of that to content and research that you're doing today? You've been a, as you mentioned, you've been a big [00:07:00] proponent of personalization for a long time, but how has that evolved as maybe the tools are getting better?
And the analytics are getting better. Like how are you thinking about pulling more of that data in to make them even more personalized, is probably the best way to ask it. I love that question.
I think that there's going to be a lot of opportunity with ai, LLMs, et cetera, and dynamic personalization based on your behavior on site.
However, I don't know that we're there yet. What I also think is, so I talked about this. Years ago about how I thought the interactives were the future of content. And I actually stand by that, which is crazy considering how things evolve. However, I would swap the word interactive with personalized. I think, you know, to your point, we have so many, we're able to get deeper into analytics and we have so many more capabilities.
But I think that a mistake people make is thinking that a really complex interactive is super cool and consumers are gonna love [00:08:00] playing with it, but they won't. They want information that's relevant to them, even if it's in the form of a basic calculator. And so I think that's what you have to go for. I could probably get on a soapbox about this, but one of the reasons Apple has been so successful over time as they boil things down to the simplest possible product most of the time, and I believe that with content too.
You need to appeal to the lowest common denominator and make it as simple as possible. So with personalization, I would follow the same philosophy.
Fi 50 years of the Keep it simple stupid methodology still rings
true. Absolutely. I gave feedback on an article with our content team, actually probably, I don't know, four or five days ago, and I was like, I don't think that, let's say someone in Nebraska who.
Has a GED and does not use the internet very often, would understand this, and that's who needs to understand that.
So yeah, I guess a part of that is always keeping the audience top of mind first, and then you [00:09:00] can fill that relationship in. Is there a conflict between, so one of the things that we learned from the Google antitrust trials is that there is.
Actually, uh, a dwell time metric can be applied to ranking. Is there a conflict with maybe removing, let me ask this different, is there a perceived conflict with removing some of the more like interactive type of, I don't know, assets that would theoretically. Diminished dwell time, right? So again, you're trying to get the person that information as quickly as possible, but in some cases that means they get it and then they leave.
Do you see any conflicts with that, or how do you think about that and as it relates to content?
I think that my customer is the user. It's not Google. So, yes, I do pay attention to dwell time, bounce rate, engagement rate, et cetera. However, my main focus is this content, giving the user what they want, and I truly feel like that will lead us where we need to be.
[00:10:00] And so. I'm not going to add a component or like an interactive just to increase time on page unless I feel like it's going to give the user more information that they need to make a decision. I think at the end of the day, if you create content that you would wanna read and it's helpful for users, regardless of the metrics and the weight that the algorithm gives them, you're gonna succeed.
And I think that we've seen that over the past five years.
Good answer. I wanted to dive into, so if you look at the portfolio of sites that sit sits under center field, there are themes certainly across the sites, but in some cases they're also very different. How do you approach managing a content team where there they have to be experts in so many different areas?
Are there some commonalities that you. Build into the process that sort of allow for that? Or are you really looking for real deep individual expertise on each topic?
So first I do look at what are the [00:11:00] commonalities between the brand. Most of our brands review products and services. We offer advice, and so there are gonna be commonalities between those, regardless of whether it's a point of sale system for business or a home security system for your house.
At Centerfield, we also have what I call a gold standard, and so we have a set of criteria for content that has to be met, and it applies to all brands, and there is no wiggle room. Whatsoever. You, for example, you need third party expertise. You need firsthand expertise, you need photos, you need to make the reader feel like they bought it themselves and they've gone through the process, right?
But after that, I think you adapt it based on the brand. There's gonna be stylistic differences, CMS, limitations, brand guidelines. But I think the most important part is you never deviate from the standard for any reason. So I don't want any piece of content [00:12:00] published that is not our absolute best that we can accomplish, given the time and resources that we had available.
In the beginning of this interview. You said 10 brands. I thought it was more closer to 15. Is it 10 different sites that you're producing content for?
So my team focuses on. 10 different sites. We do own more brands than that, but some of them are not organic plays, and so the content team does not focus on those.
So when you have a great content writer, do you ask him or her to produce content across the sites, or do you try to find what they're really good at and keep them in that lane of, you're my insurance specialist, or you're my. Security specialist. How do you keep people, how do you keep a writer engaged
for a long time?
We verticalize and we have people who are experts in business finance. We have people who are experts in physical home security, different experts in digital or cybersecurity, and so we do verticalize pretty distinctly, writers and [00:13:00] editors as well.
So maybe just to dive into one, to help people understand how.
Complex. Your job is in insurance. You have the insurance vertical. You have to produce a lot of content for this vertical. You've got a group of writers. How often do you task them with bringing new ideas? What is that? What is that like? Is it like a newsroom where you're like, all right, who's got the new ideas today?
That one's good, that one's bad. Pursue that. Gimme more on that. What's that like?
I'm always open to ideas from writers, and some are more collaborative than others. I will say. However, because the writers are doing most of the heads down work, our editors on our team, we call them content managers, they're really thinking about strategy and ideation all the time, and so they're typically the ones that are coming to the table with new ideas based on one, what's going on in the space, or two feedback that we're getting from our.
Insurance segment team on what clients are looking for, what consumers are looking for.
[00:14:00] So you can't plan too far ahead of time because you're trying to stay a little bit real time with the news of the month or whatever, what you're trying to stay relevant. So you can't plan, here's our content plan for the next six months.
You don't plan that far ahead of time, I would imagine.
So we will. But it's very flexible. And let's take a security site, for example. We run security.org, and so they cover home security, but also VPNs, identity theft, data removal services, et cetera. And so there are. Certain pieces of content that need to be updated on a specific cadence, regardless of what's happening in the news cycle, but there might be more demand for a service that pushes it up on the editorial calendar.
For example, this year we had data removal services plan throughout the year, but the demand was so high that we pushed it up into H one.
I'd love to talk about the [00:15:00] refresh cadence. Again, I think we're getting more and more data out of sources from LLMs and things like that, and a lot of it, large percentage, like 75% plus are within the last two years, some cases a year.
I guess. Are there criteria that you establish to inform those pieces that are needed to be regularly create, regularly updated? Is it like a content type for example, like for review or? Is it just content that your competitors are regularly refreshing? Like how do you land on what that refresh cadence is?
Um, it doesn't have to necessarily be centerfield. It could be prior experience as well, but
so at Centerfield, nothing is older than one year. Everything is refre refreshed within a year. I think if for no other reason, then the user might have a bias if it's older than a year, even if it's current information.
So that's one reason. Otherwise, it depends on. The article, like you said. So for example, if we're talking about business.com [00:16:00] and it is 10 things to make sure that you address with your direct report in a performance review, once a year is fine, but if you're dealing with a large. Provider of internet, like at t, T-Mobile, Verizon, whoever you need to be looking at, probably every two weeks because their deals are gonna change, their product's gonna change, and you wanna be on top of it.
And so we will go through and categorize every single article and determine at which cadence it needs to be updated.
Even in the review example, do you try to line that up with, I don't know, like end of year, right? Like when. Reviews are happening. Do you try to think about it from a, any sort of seasonality perspective at all, or?
Yeah, we do take into account seasonality, so that's a great example. Another one might be moving, so when you move, you need. New utilities, you need internet, et cetera. And so most people do that between May and September. And so that's gonna get updated in [00:17:00] April and then again throughout until maybe October.
But we'll still keep an eye on it because you never know what's gonna change. Different promotions, everything's on the table.
I'm still trying to wrap my head around how you. Your team produces as much as it does, and I have to follow up on an earlier question to really understand how you prioritize all this content.
So there's probably no shortage of ideas if you've got a great team under you, but you've gotta make the call on what to prioritize, what to get done today, this week, this month, whatever. So you maybe have two buckets. You have existing content you can refresh, you've got new content that you want to create for the new content piece.
Do, how do you organize that in your head to prioritize and maybe a follow up if I can get you beyond the record here. Do you have a few? Do you have different KPIs for different types of articles once you decide to move forward with it?
Great questions. Okay. In terms of prioritization, we look at a lot of different things.
So have our competitors overtaken us? [00:18:00] If so, why are we number one, which we are frequently, even outranking brands. And if so, why do we think that is? How do we fortify it, uh, et cetera? How do we produce the content that we do? I hope this isn't a cop out, but I've worked really hard to put a. A really great team in place, and so I have the utmost trust in the managers on my team, the writers, the editors, to make quality content.
Like I don't have to, I don't have to worry about that. I do not look at every article. I spot check every once in a while, but I hired experts, let them do their job in terms of new versus existing content. If there's a new space that seems to be hot, meaning like consumers are showing increasing interest, like data removal again, then we'll hop on that.
But we don't really create content just because we think it sounds like a cool idea. Our sites are pretty [00:19:00] well established, and so we're more so focused on. Fortifying and enhancing the pages that are already working for us than trying to go super wide in new categories. And honestly, I think that might dilute our authority in terms of KPIs.
So we have a really synergistic and close relationship with our SEO team, which I love because I don't want my team saying they gave us a bad strategy and I don't want the SEO team saying they gave us bad content. It's a really great relief and so I think it's fair to. Judge their success on traffic and ranking, whereas my personal metric is profit.
Even though I'm not over product, for example, or engineering, I'm still working very closely with them and trying to influence what I can so that our profit is as high as possible. So that's probably a very long-winded way of saying everything rolls up into one goal.
I did wanna quickly circle back to the refresh.
Cadence because I think the question we get from [00:20:00] clients all the time and just generally is like, what constitute constitutes a refresh? I'll give you an example. So I write for Search Engine Land, search Engine Journal. They regularly reach out and say, Hey, can you do an update for this article? And they provide a list of criteria.
So one of them is they want the article to be roughly about 30%. Updated, and one of the ways that they suggest to get there is rewrite the intro, rewrite the closing, and that'll get you most of the way there, depending on the length of the article. And then you go in and you make sure data points are updated and things like that.
I don't know where they got that 30% number from. Seems reasonable, like you're touching a third of the article. Roughly, do you have any guardrails for what constitutes an update, whether from Google directly or just what you've learned over your time in content? Are there like, pardon fast rules or guidelines around any of
that stuff?
I wish I knew from Google directly, but I don't. Anecdotally we will consider anything between [00:21:00] 30% and 50% content refresh. Again, my directive to the team is here's two articles. Review both. If one is completely up to date, you have nothing to change, don't change anything. And you can add a little subtext that says This was reviewed on X date.
Everything is accurate. We'll review it again, three months. Another one might need a 50% refresh and that's also fine, but I don't know. I could be wrong, but I don't personally believe that. Revising a paragraph, for example, for the sake of revisions and not adding anything new is gonna do anything for Google.
Makes sense. I wanted to talk about two articles recently that you've been involved with that got a lot of press and links. One was around the DEI pronoun study and another was fusing marketing interns to research the job market, which was like one of the smartest things that I've never thought of. So like how do you ideate?
[00:22:00] These sorts of things. Is it something that you're hearing from, I don't know, the sales folks, and you're like, oh, this would make sense to build into something maybe longer form, or is it like a personal interest and that sort of takes you down a path where, oh, this would be a smart thing to do. I'd love to just get inside your head around the ideation of some of these ideas and try to understand it.
I will try to make this as actionable as possible. Well, first off, I think whenever you're trying to get press coverage, you're thinking, how do I please the media and how do I please my audience or my target market at the same time? Ideally, you can marry the two and it can be done. However, if I have to choose one or the other, I'm picking the media.
Because I want the press coverage, I want the links, and I think that if I get those, I will surface higher in the circs. People will go to the site, the onsite content will do its job. I am not trying to get referral traffic from research like this in terms of how. I [00:23:00] come up with the ideas, I can't take all the credit of course.
Part of it is I've just been doing this a long time and I have a gut feeling for what the media would be interested in and what is old hat and might not work. And sometimes I go with that gut feeling and it's successful. Like this DEI pronoun study and sometimes I go with it and I have to like crawl back with my tail between my legs and tell my boss, oh, this didn't really work.
But with the pronoun study that's been a pronouns in the trans community have been a topic of contention in the media for a long time. And at the time that I did that study, it was during COVID, the job market was really tough. And so I was thinking about how one impacted the other. Um, and so I, I took a shot and it landed and I'm happy for that.
And that's how that came together.
How often do you have to sell? An article to, you've gotta, you've gotta get the news to notice you [00:24:00] and to do that, you've gotta make a lot of requests. Hey, would you read this? Are you interested in covering that? That's part of your job too, right? The pr you're trying to, you're reaching out to news organizations to see if they're interested in covering.
I don't know that anyone we've interviewed has that responsibility. I wonder if you could just talk a little bit about how much time that takes. It feels like maybe some could be automated, but it feels like a lot of. Rejection you've gotta deal with and just like time investment. No.
Yeah, I, so when I started in this industry, I actually was the person pitching, and at the agency I was at, we did personalized pitching.
So every journalist got accustomed. Pitch, which I think is a great method that takes a lot of time, like 30 minutes a pitch. Maybe you get out 10 a day with your research and everything. There are agencies that will take a different approach. It's one pitch and they have a huge media list. I think that both work, I know that I would do.
Get some disagreement on that, but I've seen it. They both work. What's most important is the [00:25:00] story, and so if you have a really salient story, I believe you'll get coverage even if you have a very mediocre pitch. If you have an amazing pitch that's super personalized and your story's boring, you're outta luck, which is why I laugh at all of these agencies and stuff.
List in their job descriptions. For PR pros, you must have a Rolodex or something. It really doesn't matter. It's all about the story. It's all about the data. So it depends. Sometimes you can pitch three people and they all cover your story and it syndicates like crazy. Sometimes you pitch a hundred and you're outta luck.
At the end of the day, it all comes back to knowing your audience, right? Like they're, they know their audience best. If the pitch is weak in relation to who their audience is, that they're not gonna pick it up. I've, I like e even just in the short time we've been doing this podcast and going back through and like editing the videos and pushing them out for the audience, I think one of the toughest things to [00:26:00] land on is like, what is the hook?
Like, what is that? Because you may have great data and that story or that hook may be hidden in there, but I think you know your talent. One of them is probably being able to pluck out what the right hook is. Do you have any tips on, and I know it's probably very tough 'cause they're all different subjects and things like that, but are there any specific things that you would like when you see it, you're like, aha, that's one that will resonate regardless of audience.
Does that make sense?
Yes, absolutely. What's gonna surprise people and or what is going to maybe contradict a deeply held belief? And so I don't mean that to be incendiary, but the data is the data, right? That's why we lead with it. And so if the data says one thing, and it's a contradiction to what a subset of people have believed for a long time, I'm gonna throw that out there.
It's objective. It's not my opinion, and you guys do with it what you wish, but I think that and surprise are the [00:27:00] most, inciting is not the right word, but conversation sparking emotion.
Well, you're basically looking for that conflict.
You're looking for conversation. Conflict a lot of times is the impetus for that.
I
would think it's business.com that creates the most opportunity for that, for you hope. Maybe I'm wrong, maybe it's just 'cause I'm an avid business news reader, but it feels like you could come up with something every day to get that community engaged.
You could. It depends though. savings.com is a good example.
Starting four or five years ago, we began this franchise on again because of COVID. How are. Parents supporting their adult children financially. And that is gonna be a hot topic because you have parents who are like, I have to take care of my kid. They can't help it. This is what I have to do. And you have other people who aren't empathetically saying, cut 'em off, let them spread their wings, blah, blah, blah.
Everyone argues about the [00:28:00] amounts that people are spending, and so that. That's another one. I would say anything to do with money is probably gonna spark that conversation.
Like what the getting, getting the reactions from the studies are probably the goal, as you mentioned, like the links and things like that.
Just quickly touching on the DEI pronounce study again. You got a comment from El Elon Musk and you were talking about business conflict and things like that. Was that like exciting or was that like, like him in that conversation is weird. What was
that like? It was, okay, so the PR person in me was thrilled to get exposure from Elon Musk, however.
It set off a cascade of articles on publications that lean a certain way politically, and they were so brutal to me. However, at the end of the day, they were giving me coverage and making me look good to my boss. So thank you very much. I'm very happy with that. I don't really care. I do [00:29:00] care, but I can't control what people say about the research that I'm putting out.
All that I can do is make sure that it's objective, it is statistically significant, it's ethically done, et cetera. And then people are gonna do what they want with it. And I hope you talk about it.
I'm curious the transition away from Fractl to centerfield. Certainly at Centerfield you still have like internal clients, right?
Probably a very different conversation than what you were having at Fractl. Is there for folks who are trying to pitch these sorts of things to clients, did you have any common things that you would utilize to try to sell in some of these ideas or were they pretty well received from clients generally in, in the name of getting links and, and exposure.
It depended. So when I was agency side client tolerance dictated how far I would push the envelope. So some clients would say, you can do literally anything that you want as long as you get links. Enterprise level clients, that was not [00:30:00] the case. They were not okay with that. You had a lot of guardrails between their like teams that you didn't even really work with, like pr, legal brand, and it was a little constricting.
However, it goes without saying that you can get good results with no controversy. It's all about the data. So if your conservative client has interesting internal or proprietary data, that's a gold mine. And so that was the approach that I would take with them. So it still had a competitive edge, but it wasn't, I never wanted to be sensational.
Got it. That makes sense. I've always been curious about this as well and would love to get your thoughts or maybe even your approach on it. So you drive a ton of links into a study or maybe, uh, an infographic or something like that. And is there ever a point in time where you then take that URL and redirect it about to a service page or to a product page where you're, I dunno, maybe more directly taking all that link equity and pointing it to a page.
[00:31:00] Has that ever been part of your process?
Rarely. Typically we'll go through and find natural opportunities for internal linking and it'll flow down that way, but very rarely would we redirect. I actually can't think of an instance where we've done that, with the exception of we have these franchises that we do every year, and so we'll use the same URL every.
Because publishers are gonna reference that. We would redirect that.
Can I maybe try to make a little segue from manual verse automated into ai? I have a competitive question for you, but I'll start with an anecdote between Jon and I. Jon is. Pretty good at understanding when someone sends content that's created with ai.
We're on a lot of the same emails, and he'll slack me and say, Hey, did you see this email from X, Y, or Z? Clearly, that person used chatGPT, like he has an A knack for understanding. When people don't customize at all, right? They just copy and paste in [00:32:00] your, it sounds just from this interview, you do a lot of competitive research before you decide what to what content to produce.
Are you noticing any of your competitors, you don't have to call an out by name if you don't want, but are you noticing some. Are using some automation and that maybe you're like, wow, that's, I don't know. It's interesting. It's, or any opinions on the com, your competitive set, leveraging AI in ways that you think maybe are not ethical or something you wouldn't copy, you just don't feel comfortable with it?
I think it's still the Wild West. So you know, we have some major competitors that have made it very clear they are anti ai. They don't use it at all. I think that they are hanging their shingle on that, which we have others who I think go too far in the opposite direction, where I'm like you. I'm like, okay.
Like everybody knows that. Yeah, like maybe a few less M dashes. Right. I try to put. In the shoes [00:33:00] of the consumer, because I think as marketers we're hyper vigilant and hyper aware of stuff like this, whereas the consumer might not be, and they still might get what they want. A sidebar. I think that sometimes people see, for example, this content was AI assisted and it immediately introduces a bias.
Whereas if it, if they were told a freelancer wrote this a hundred percent top to bottom, and it's the same words. They would not have that bias. My perspective is you have to identify where in your process human touchpoints are critical. You are unyielding about those and also your standards. And I do think that AI can fill in the rest.
The same level of quality, if not an enhanced level of quality.
I thought the experiment you guys did on business.com where you're comparing the writing outputs across LLMs was really interesting was were there any like aha moments that you guys had, maybe yourself or even the team internally around what [00:34:00] those moments might be from that study where.
Could come in and maybe fill that gap or get to an end product faster. Yeah, talk, maybe talk about that study as well a little bit and maybe some of the findings.
I think that that study gave a great analysis. I also think now it's too old. I think that every six to eight weeks there's an LLM update.
It's a drastic improvement over the last version, and so I'll go a little anecdotal. Two things. Again, I believe the AI can. Create content that's better than what a human would produce. I also think that there have to be human touch points placed very strategically throughout the content creation process because we need human judgment.
For example, I would not have my team ever publish any content from any source that was not informed, read and edited in full by humid. I think if we are talking about specific human touch points, if I'm. If I'm a writer and I'm doing a product review, [00:35:00] I need to review the product and it needs to be my words, my experience not filtered through ai.
I need to get unfiltered expert quotes straight from sources. I want first person photography. I want first person opinions. And what can be filled in by AI is maybe more informational content, for example, that could be pulled from a vendor site like security camera specs, that doesn't have to be written by a human as long as it's accurate.
Fact check by I
I would imagine in your role and probably the team as well, you guys are, I don't know, like for me, for example, even if it's not work related, I'm still interested in the tools and things like that are coming out more of just like how fast this industry is moving. Are there any tools that you've personally played with that?
You're like, wow, this is, I could see potential here. Good
question. For me personally, I use Claude and Chat, chatGPT the most, which I know is not groundbreaking, but I think it's more about figuring out how to use them. I will use them for very different things. My [00:36:00] team has gotten more in the weeds. So Make is a platform.
It's been really amazing for process automation, project management, et cetera. We're not using anything that nobody else has heard of for day-to-day tasks. We're using cloud. We're just using it in a way that really works for and benefits us.
I was on a, I was on a panel with a bunch of recent graduates, I think two weeks ago, and one of the questions was for up and coming grads, what would you recommend?
Or how would you think about their skills shifting? What should they be doing? And somebody said, oh, use LLMs. And I was like, that is true. Like really learn how to use them and like prompt engineering. Basically create a project or a custom GPT and get the output and then evaluate that output and then go back and re-edit that prompt, right, and continue to get it better and better over time.
I think that is the. I, I think that will be the differentiator between someone who's like, yeah, I use G ChatGPT, versus someone who like really. Uses it [00:37:00] and can figure out how to get the best output from it. It's going back and revisiting those prompts and refining them over time, where I think you really get the unlock from the LLMs generally speaking.
I, I have a good question from Jon statement I got before I forget it. Is there a skillset that has changed, like who you're hiring? Do you, is the person you looked for 18 months ago different than who you'd look for today? Maybe just based on that familiarity. In use of ai,
when I am looking for people, I'm looking for someone who is comfortable with ambiguity, someone who has a figure it out attitude.
And I don't even mean that like you have to solve every problem. I want you to figure out, can I solve this? And if so, how? And if not, why. And that's okay. Someone who's empathetic, we're writing for people. You have to put your yourself into a lot of different shoes. Jon, who's my boss, he used to work at Amazon, and so he introduced me to Amazon Leadership Principles, which I actually am really into.
And so a [00:38:00] few of those that I look at are learn and be curious. You really need curious people and people who. Assume positive intent, right? Because in our industry it's not science. It's a blend of art and science. And you might have differing opinions, but I want people who think that everyone's coming from the right place.
'cause I believe that insist on the highest standards. I want a bias for action. I want people who will disagree with me and people who will deliver. So that's what I look for. We have people of all backgrounds. We have journalism, statistics, speech pathology, and that was the case 10 years ago. I won a well-balanced team over a team that all knows Excel at an expert level or three years of technical SEO experience.
I have people who are great at. Creating reporting dashboards or working with the engineering team on our brands that have a more complex backend. I also have people that are not good at that, but they're [00:39:00] incredibly creative, and so I think that if you have a team that's well balanced like that, you can play to everyone's strengths.
You can avoid their weaknesses. To an extent and that makes them happy, is also important.
God, that positivity is so important, isn't it? Like it, I feel like that's rarely mentioned, but it makes all the difference.
There's gonna be so many roadblocks, right? And so I, I want someone who's gonna hustle and push through even if we don't get the end result.
Like maybe it's okay. This was a totally. This is a failure of an idea. Okay. But we tried. We got there. Let's pivot. No big. He learned something from failure too.
Exactly. Yeah. Teachable moments come out of that as well. Well, before we wrap up, I'd love to get your thoughts on what the future of content might look like.
So can you foresee any significant changes to. Maybe how you approach content or maybe the impact AI is gonna have on content production in the future.
So one, consumers want a personalized experience. [00:40:00] Two, I'm not a tech SEO expert, but I know that ChatGPT, BT, and Claude do not crawl a page the same way that Google does for traditional SERPs.
And so it's a little too early to tell, but content structure could have a relatively outsized impact compared to before. I think it's been the same answer. All along though, there may be different formats, different ways of surfacing answers, make content that you wanna read. That's actually a really big question I
didn't, I didn't think to ask, which is how do you think about outlining content?
So on, on our side, we're thinking about like semantic structure, right? Heading tag order, and. Writing for the web versus editorial, like traditional editorial, which is like big blocks of content, right? We're thinking about bulleted lists and key highlights and things like that, right? That is more scannable.
How do you guys think about performing an outline before you get into content? Or maybe how have you thought about it in the past?
We do the same. We'll work really closely with SEO, and so it [00:41:00] will. Be structured in a way that lends itself well to obviously surfacing as a result on Google, but feature snippets, AI overviews, et cetera.
Well, for a minute like that, while also keeping in mind is this detrimental to the user, it never is, frankly, that's how we go about it.
So we've been asking this question to wrap things up. If you go to google.com 12 months from now, what do you expect that experience to be? It's changing so quickly, but.
Can you foresee any significant experience change when you go to visit google.com?
The optimist in me hopes that their AI overviews and AI search is more accurate because if so, if they can get there in 12 months, I think that it will be a completely personalized. If not, I think it will be. Some variation on what it is now.
Very good. And last thing before we let you go, what's the best link you've ever gotten?
Oh man, that [00:42:00] is so difficult.
You've gotten so many. I was just curious if one like really stood out. I think New York Times,
you can't
beat that. That's the holy grail, right? Because you have to pass so, so many thresholds to be included there.
I don't even know how we did it. I'm grateful for it. I don't even think I can take credit. Maybe it was an act of job. I have no idea.
Love it. Ryan McGonagill, everyone. Thanks again for joining the Page two podcast and if you enjoyed the show, please remember to subscribe. Great. And. We'll see you next time.
Bye-bye.