How to Grow with SEO by Ethan Smith of Graphite

Ethan, CEO of Graphite, pulls back the curtain on how he builds and executes an SEO strategy.

In Ethan’s words, “SEO is magical”.

Here are our biggest takeaways from Ethan’s talk:

• Topics are the new keywords. Identify the topic then corral the keywords that map to the topic. The total search volume of the keywords that map to the topic is your total traffic opportunity.

• Match the page type on the SERP along with the elements on the page (e.g., tables, product reviews, videos, maps).

• SEM traffic is an authority signal and passes topical authority to your site which can boost your organic rankings.

• Start with conversion content where your topical authority is the highest then move up the funnel and span out.

• Build intent over time with SEO.

Watch the full webinar

Read the transcript

[00:00:00] Ethan: I'm going to walk through our latest overview of how we approach SEO and there are a few different pieces to that. So the first is. Assessing what the opportunity is and understanding where we should focus. The second is given where we should focus. More of the opportunity is how we build an SEO strategy.

[00:00:20] Then once we have an SEO strategy, how to build a content strategy for those pages once we acquire the users, how to convert or engage them, and then last is how we measure success. So go through each of those a little bit about us as you mentioned Travis, we rebuild SEO and content strategies for companies like Netflix, Robinhood, and MasterClass.

[00:00:41] And so we've tested our model over and over again and refined it. And to your question about intuition versus data, you know, the more data we have, the more cycles we have, the better our model is a better or intuition. That's all I'll share with you our evolved strategies are based on many different cycles.

[00:01:02] So starting with opportunity. I think the first thing. I want to highlight that that's surprising to people. It's just the the magnitude of channel of SEL. And it's a lot of times, not only the number one channel, but larger than every other acquisition channel combined. So you'll look at companies like TripAdvisor, Pinterest, and Reddit Pinterest grew early largely S.

[00:01:26] TripAdvisor still 80% SEO, right? It's over half SEO. Ticketmaster's a huge percentage of SEO. So the channel relative to other channels is huge. And I think it's a lot of times under-resourced relative to other channels. And so it's a huge opportunity. Focusing on a few companies that we've worked with it's, it's over half of the tropics.

[00:01:46] So better up, for example, 80% of all the traffic is the SEO MasterClass. It's around 60 to 70%. So it's a huge channel for many of the companies that we've worked with. However, it grows over time. So with paid advertising, which is what people are most familiar with, you get the results immediately. And so you're able to measure the impact very well with ads and precisely.

[00:02:11] Whereas with SEO, it's harder to measure because there's a different impact. However, there's this fat tail. So an investment that you made today may or may not get results next week, but it grows over time, and in year two and beyond it, it recurs. And there's no variable cost for the investment on that page that you made a year ago.

[00:02:28] So that article on nerd wallet, you don't necessarily need. Created a new article years later, that same article can, can not only derive results five years later but also increase over time. So it's a, it's a highly appreciating asset versus a paid where you get all the results in the first few. And we'll see that with actual examples.

[00:02:49] So this is an example of the MasterClass where the chart looks very similar to the one that I just showed you. So investments made at point G don't necessarily get returned to the next month, but they get returns years later, and you have this huge fat tail of results for the investments that you made early on.

[00:03:05] So I'm gonna use Brexit as an example, just because we haven't worked with them. And I. Speak to them without any confidential information. And I dug around Brex and Babel large opportunities to breakfast, a small business bank, or a tech bank. But for those who aren't familiar with them, they have a lot of branded searches.

[00:03:23] Half of their traffic, probably over half of the traffic is branded search and there are a lot of upsides, for them to rank them. So I'll use them as an example. And one of the big things with Brexit and companies submitted Brooks is. The opportunity is to target the persona. So there are people who are looking for a small business bank, and then there's anyone who might want to use a small business bank, which is going to be 10 X, the size of people who know exactly what they want.

[00:03:50] And so targeting the persona and targeting the director of finance is where the large opportunity to increase the dress market is for breakfast with SEO. And one important thing with SEO is that. This is the marketing rule of seven, where it takes seven interactions before you're ready to convert. And even I would say it's more so for people who've subscribed to the MasterClass, they've probably heard of it at least 20 or 30 times before free, typically before they're ready to subscribe.

[00:04:21] So the job of SEO is to increase the number of interactions with the brand. So that. Qualify and prime them to subscribe later, versus get someone to convert immediately in that session, which is what we would usually measure with pay. So we're building the intent over time with SEO and persona-based targeting is, is much, much larger.

[00:04:43] So again, the director of finance, there's 10 X, the number of quarries of the director of finance as there is that person's typing in. I want to buy I want to switch my banks and. We want to target that persona throughout their journey. So at the end of the journey people might type in a small business bank, but at the beginning of their journey, they're typing in things like corporate rate, Texas, or how to set up an options plan.

[00:05:07] So maybe they're not ready to switch banks yet, but they're exploring how to set up their financial plan. And eventually, they'll be ready to switch banks. And so if we can target them throughout that journey, then when they are ready to switch banks, they're going to think of breath. And add skew transactional.

[00:05:23] So Google is more and more taking over the result page with ads and maps and you know, rich modules and that's especially skewed with transactional query. So for better up, if he's typing executive coaching, the entire pages are ads. Whereas if you type in career aspirations, there are no. And so back to this user journey at the end of the journey, it's all ads and at the beginning of the journey, it's organic.

[00:05:53] And so organic and SEO has this huge opportunity to target this awareness evaluation piece with no competition. It's kind of like unpurchased ad inventory. That's just a Greenfield for someone to target. So that's kind of how I think about SEL and the workplace. So when we're, then we want to figure out what's the, what's the opportunity and where, where are the largest opportunities within that?

[00:06:17] there are a couple of ways to do forecasting. The traditional way is to gather many different keywords. So gather all the different keywords that somebody might search. For a small business bank and then some, some of the volume and the issue with that is that we don't know the position or the click-through rate by position.

[00:06:34] And so then you have this very wide confidence interval where you're either way too smaller, way too large with your forecast and it's completely off usually. So I tend not I tend to not forecast based on the keyword level, from a bottom, bottom-up perspective. And rather. Forecasts. So instead what we do is gather many different competitors and this is using similar web data, which tends to be more accurate.

[00:07:00] And we will look at organic traffic and then organic traffic. And then we will look at brand versus non-brand. And then we'll calculate out what's the total non-brand SEO opportunity. So that's how we came up with that 500 K that I mentioned, but this tends to be a lot more accurate because it factors in the position and click-through rate that I mentioned.

[00:07:27] So a lot of that noise gets Gets handled in this top-down approach. So as a tip of the bad, we do full custody. I'll say it. Quick

[00:07:37] Travis: question, Ethan. What do you think about prioritizing conversion verse evaluation, verse awareness? When you're first getting started, a lot of people would say conversion is the way to go, but curious to hear your thoughts

[00:07:49] Ethan: That's exactly what we do.

[00:07:50] We would start with conversion. Then we would go up the funnel and with Brex which, and I'll touch on this in a bit, but Brex will have more topical authority for the features that they offer. Just like the clear scope is going to have more, more topical authority for a content analysis versus SEO in general.

[00:08:06] So we want to start at the bottom of the funnel and then go up the funnel. Partly because the conversion and intent will be higher. And also because we will rank better because we'll have more topical authority. So that's typically how we prioritize. Excellent. So we're, we're forecasting based on competitors.

[00:08:26] And one important nuance. We'll look at competitors who have similar product offerings, and then we'll look at competitors who are competing for the people and the searches that we want to compete for, or that we want to acquire but don't necessarily have the same product. So these are similar products to ramp and Expensify.

[00:08:43] And then there are companies like nav or Investopedia or tide. Who does not have the same product offering, but they're acquiring the users in searches that we want to acquire. And we want to size that as well. So we'll look at those two different groups and then sum them up. And then that's the opportunity.

[00:08:59] So that's where this 500 K comes from. But again, to summarize, we're looking at not only product competitors, but people acquiring the audience that we want to acquire. We're doing persona-based, not just a feature, a product offering based. And we're doing top-down based on external benchmarks versus bottom-up keyword-based forecasting.

[00:09:24] So that is how we forecast. And now we figured out where the different opportunities are for breakfast, and now we want to build an SEO strategy. So I'll jump into that. And let me just see if there are any questions I think Bernard

[00:09:36] Travis: had another question kind of back to the beginning. It's like a, do you recommend pairing your organic strategy with SCM to generate more impressions?

[00:09:46] Ethan: Definitely. Then, benefits number one is the same effort to build the page for SEO and SEM. If, if we target both, that's great. So if we build a page about how to set up an options plan, we can buy ads for that. We can also do it organically. So there are efficiencies of cost. The other is that STM will call.

[00:10:04] Us to rank better in organic. So SEM and non SEO traffic are authority signals. So if we're buying a bunch of ads around the options plan, we're going to have more topical authority around the options plan, which means we'll rank organically for that. So for both of those reasons, it's good to pair the effort for that.

[00:10:21] So I'll jump into how to build an SEO strategy. So whenever building a strategy, we want to start with the intent to be. And Google talks a lot about intent and the way Google talks about it. This is somewhat vague. So make, make good content, build a good product, which is true, but we want to be a little bit more precise and use data for that.

[00:10:44] So the three pieces. There are four pieces there. Number one is topical authority. So again, if you're buying a bunch of ads on the options plan, you're going to build topical authority. The second is what is the right page type. So if you're, if you're searching for how to set up an options plan, what, what page type and functionality do you want for that?

[00:11:02] The, third is what's the content strategy and the last is the engagement strategy. So if you, if you publish an article about. How to set up an options plan. Did you get the answer to that? Did you learn, are you able to act on that? So these are all feeding on, you know, what the intent of the user is, and this is one of the big, different big changes in SEO over the past 15 years or so.

[00:11:22] So in the past, we would have a hundred million. This is when Bernard and I first met. So let's gather a hundred million different search quarries for finance and have exact match pages for every single one of them and spun texts and scraped everyone else's website and chop up the sentences. That was the old strategy.

[00:11:38] Obviously burner wouldn't do that, but other people would do that 15 years ago, but the algorithms evolved, and the algorithm is optimizing for the outcome of the query. So you typed in how to, how to set up an options plan. What were the outcome and the inputs for the outcome? Did you have the right page type functionality that should have the right content strategy?

[00:11:56] Did you have the right engagement strategy? And we can break that down in this data. So. And work and we're ranked based on those inputs and we're ranked relative to the next best competitors. So MasterClass is not ranked against Wikipedia for negotiating tips. We're ranked against the other competitors who are ranking.

[00:12:16] So it's not about what. Overall authority is, for example, it's our authority relative to the other competitors. And it's not about whether or not our content is good in the abstract. It's how does it compare with the other results that are ranking? And we can go through those results and measure ourselves on these dimensions of authority, design content conversion, relative to the other results that are.

[00:12:41] And the more we fulfill the intent across these dimensions, the better we'll rank. And so MasterClass is a good example where MasterClass is not had as much authority as other sites like Harvard business review, or at least they didn't when we started building, building a strategy. But we want on these other dimensions, especially on the content side and on the engagement side.

[00:13:02] So let's break that down on the design and UX side. So it's not just about having an objective or a subjective. Good. It's especially about the page type in the elements that are on the page. So a good example is if you type in best plus X for pretty much anything, like let's do the best shampoo.

[00:13:20] There are no category pages. So 10 years ago you would have category pages with a greater list of items. And now it's a listicle so you'll see all of these are listicles. So birdie allure near New York post. These are, are lists of items framed as an article with an intro and then snippets for each.

[00:13:41] And if we have a category page, we're not going to rank, even if we're Amazon, we're still not going to rank because the page type is not the right page type to fulfill the intent. The intent is to browse and learn about shampoo. It's not to buy. Yep. And so if we have the wrong UX and page type, then we will not rank for that.

[00:13:58] So step one is given a particular query what's the right page type. And what are the elements of that? If we're looking for a restaurant or a hotel we want. If we're looking at a product, we might want to buy that. So we want to make sure the page type is the right page type and the elements on the page type or the or are all the right elements.

[00:14:16] So we do that by taking the queries that we want to rank for and looking for patterns. And I'll show you an example of that. So for corporate tax rate texts, We can screenshot the different results for that and look for patterns and what are interesting patterns. And we'll just sort of screenshot this and put it in a Figma file so that we can eyeball all the different patterns.

[00:14:38] So there are two patterns with the corporate tax rates, in Texas. The first is that it's typically articles and the second is that we'll have tables or maps. So let me show you that if we go over here. Yeah. So if we go to. One here rich states, poor states. And this is an interesting one because this is not a site with mass quantities authority, but it's winning on the content and it's winning on.

[00:15:03] And you'll see a map and you'll see a table. So it's not just the corporate tax rate of Texas, but it's the corporate tax rate of all the different states, which might not be obvious to us because if you're answering the query of the corporate tax rate, Texas, you just say what it is, whereas the actual pages that are ranking or a table so that you can, and probably that's because the intent of the user is they're looking at different options.

[00:15:24] They're looking at the corporate tax rate of Texas, which is, which is pretty low. And they're looking at the tax rate of other states. And so to fulfill the intent, when you. And if we look at some of these other examples here, such as the center square, you'll see that again. So there's a.

[00:15:39] And this is not a Texas-specific page. So this is another interesting thing. We might be able to rank for corporate tax rate plus any state with a single page versus making 50, 50 different pages. So we can do the same thing for any query, but we're looking for patterns that correlate with ranking and those patterns then say to us, okay, so we should probably have an article.

[00:16:00] We should probably have a table, should probably have a map and mentioned the corporate tax rates, and corporate tax rates of other states. And now we have some base requirements for what page we want to create to rank for that.

[00:16:13] Quick

[00:16:13] Travis: question from Bernard, kind of on the UX. Are there other kinds of page types outside of the article listicle directory engineering widget that performs

[00:16:21] Ethan: well? Yeah. The main page types tend to rank our category page. So a shampoo grid of shampoo, a listicle, which is an article of a list of items framed as an article, like a, what I just showed her product page.

[00:16:38] So the product page could be a page about a camera, a page about a restaurant. It's not a hotel. So it's a page about individual item articles. And homepage. Those are the primary page types that we see the most often. And Bernard we've got got a lot of questions from Bernard which is exciting.

[00:16:54] Masterclass pages are rich with video. Yeah. So MasterClass it's a good example. And if you type in video is always better. So if you type in how to dribble a basket. For certain quarries, you'll see a lot of videos, and then for other quarters you might not, but you'll see.

[00:17:16] Video outranks. So for not all, but some of the MasterClass particles we'll have a video. When we've tested, adding with video, the click-through rate goes up adding the video, causes the rank to go up and wider video, rich snippets, and then last, most importantly, the engagement goes up. And so the primary thing that MasterClass is trying to get people to do is to watch a video.

[00:17:36] When you watch a video, not only did you get this visit, but if you watch a video and it's branded as MasterClass, the intense. Increases significantly. And so that the number one impact of video is to just make more out of the visit that you got to, to showcase that product of MasterClass. And then the second is that it causes a ranking increase and it causes a click-through rate increase in the title of the page. Yes. So we do. And I'll show you what that looks like. So we'll put it, we'll append it here. You don't have to, but it helps with put through it when you add with video because people want to watch the video. And again, we're following the intent better. People want to watch a video about how to poach an egg.

[00:18:17] And so so the result of ranks higher has better engagement, like, like the live example. I have not put that. That sounds exciting. So adding a live example, I'll try that out.

[00:18:28] Okay, so we've, we've built. We, figured out what the opportunity is we've identified or with prioritizing the different opportunities to category page, product, page listicle article homepage. And we built those. We figured out the requirements for which beyond these pages, we know that we want a map and a table for our ops or for our corporate site.

[00:18:46] Page. Now we need to connect these pages and Google finds pages via links. So backlinks are well-known, which is people linking another site and looking to us, but linking within our site is less well-known and we've seen a large correlation with MasterClass or our company or Brex linking. Linking into itself.

[00:19:08] So it was Brex or MasterClass linking to all the different pages on their site. Multiple times. It's not enough to just put the page on the site map when we need to link to it multiple times. So we've seen a strong correlation between links within the site with crawl and traffic. So the X-axis here is just linking to our, linking to our poached egg article, more within MasterClass.

[00:19:29] And then red is Kroll and greenest traffic. So basically more lengths mean more crime and more traffic. And we've seen this multiple times. So this is super important and is typically not optimized well, and here's what that can look like. So with a Thumbtack let's just show you that we have these links at the bottom here, and this helps the body find all the different pages on the site.

[00:19:53] And this causes more Tropic and more crawl. The links can be anywhere on the page, but this is a nice place to put it because you can automate this and scale it, but this is super important.

[00:20:03] Travis: And then what the internal links, how specific are you with the anchor text?

[00:20:06] Ethan: We just use the title of the page.

[00:20:09] We hypothesize that you can cycle, we hypothesized that the anchor text actually passes some sort of topical authority or they don't have great data on that. So an example of that would be, you know, we, we add the word best plumbing near being, and we rank for the word best for, I haven't tested that.

[00:20:27] So I'm not sure if that's true, but we'll just use texts that make sense. The primary keyword, the page, would be an interesting test. Does the internal link? Yes. So we're an ARDS question is what if we'd link 10,000 times as 10,000 better than 9,000? And I don't know, but I don't think. It's typically a sufficiency thing.

[00:20:48] So if you link, at least let's say 10 times or 20 times that's better than five, but it caps out. So it's more like a log curve and it's sufficient. So 10 to 20 links is sufficient. You don't need to link 10,000 times. And then I've heard that using the broad range. Yeah. I think that's true. And I would love to test that.

[00:21:07] So on the internal thing, Piece, we can also measure that and we can measure that with screaming frog. So what I will do is I'll just show you what that looks like. So we're looking at looking at ranches blogs. And I will test that by going to the robots, go to the robots, then you get the site map.

[00:21:30] So I have this here, there's a site map index. So you'll see that here. There's multiple of those. And then I will use screaming, frog, and I'll show you what that looks like really quickly. I'll let people look at a screen. Tutorial separately on YouTube. But just to show you how we can quickly gather the number of internal links, we'll load screaming, frog, and then we'll take the site map and I will show you that really quickly.

[00:21:56] So here's screaming, frog. Let me go. Here we go into the list mode and then we upload, we download the XML site map. We paste that in and then we start crawling and we have a little site map here. And now it's crawling, I don't want crumble side, but just to make sure everyone knows. So now it's going, and this will then give us a unique inlink here.

[00:22:30] And that tells us whether or not we don't have enough internal links. So I'll show you that here.

[00:22:39] So now we have this and we can see that the majority of the posts on, on branches blog, which is great, don't have enough links. So it's a branch that has published all this great content about how to grow mobile apps, but we're not linking to it enough. And so the bot can't find it, or at least it doesn't crawl it in as much as it could.

[00:22:55] And that isn't getting as much traffic as it could. So we can precisely measure whether or not. Internal links or an opportunity and it is, and it's almost always important to opportunity. Now we can assess that. So we've identified the opportunity. We prioritize the different opportunities. We figured out the right page types for you, for the by intent.

[00:23:17] We figured out the requirements of the page, and the pages we've connected to them with internal links. Now we need to build amazing content. So let's jump into that. One

[00:23:58] When Bernard and I started, we would have one page per keyword. And so we were doing keyword-level targeting and ads paid searches, the same we're targeting individual keywords or broad match keywords. Now we're targeting topics and the topic is like a concept or a single URL or a single article. So for the how to set up an options plan, that is a concept with a single article and that articles, ranking many different keywords versus one page.

[00:24:25] And here's an example of that. And this is interesting. So if we look at how to set up an options plan, actually nobody's even typing that in. So we didn't know that, but if you look at the top at the topic level it's popular. And so people are typing in things like employee stock option plan, and I might not.

[00:24:43] That to add employees. So I don't even necessarily know what the keyword should be, but I know the concept of the topic. So then we would want to find the keywords that are mapping to that. So employee stock, and you'll see that it's similar, but not quite the same. So an example, for startups specifically, or a template.

[00:25:00] And if we had an article that didn't have an example or it didn't talk about startups, it only talked about non-startups. We're not fulfilling the intent of that topic. So. It's important to understand the keywords that mapped to this topic so that we can understand all the different subtopics. And the other thing was the opportunity.

[00:25:20] So if we looked at how to set up an options plan, we wouldn't, we would never make a page about it because we would say that there's nobody searching for it. But the actual search volume for this topic is thousands of people. And when we're prioritizing this versus something else, we want to know the search point of the topic, none of the single.

[00:25:36] And we would be wildly inaccurate and make that decision. If we only looked at this one keyword, so super important the next thing is topical authority. And this ties back to you know, prioritization of conversion, keywords versus awareness keywords, but topical authorities, a new page rank.

[00:25:54] So page rank that people know about it's based on backlinks and the more backlinks more Patriot. And that's. More complex and more robust than sin. And so there, there are new inputs into authority and then authority has topical relevance. So the new inputs to authority are social shares, a branded search, anchor texts for back links.

[00:26:15] If we're buying a bunch of ads on the options plan page, that that gives us authority. So that. More inputs than backlinks and the inputs are applied to a particular topical area. So if we've got a bunch of backlinks and branded search about options plan, but not about credit cards, we're going to rank for options plan.

[00:26:32] We're not going to rank for credit cards. And so not only are we looking at our authority overall, but we're looking at where it is. And and this is an example with Brex. So branded search people type in the brand and then they type in a phrase. So branded search plus Brex plus credit cards. And now we have more authority for anything containing credit card.

[00:26:52] We have a more topical authority for keywords containing corporate. And even if we're just ranking for something and we have good engagement, like vision statements, we're going to rank better for, for that as well. Same with LLC. And you'll see. Brex ranking for a bunch of LLC-type quarries. So their topical authority are all of these words.

[00:27:13] So the simple way to assess where you're going to go for you is just go and Google search console and look what you're ranking for. And that's where you have top. And again, we are ranked relative to other companies. So Brex is ranked when we're, when they are competing on LLC tax benefits, we're not competing with Wikipedia necessarily.

[00:27:33] We're competing with the other results that are ranking for LLC tax benefits. So it's all relative.

[00:27:42] So massive POS when MasterClass started, they did not have authority for food or for photography. They had authority for the names of the celebrities. So they had authority for Gordon Ramsey or Anna Lee, the webs, but not for pizza. And when we. When we started, we did topics that were adjacent to that.

[00:28:01] So rather than talking about pizza and salad, we talked about beef Wellington, or we talked about salmon, which Gordon Ramsey is known for. So we had topical relevance for Gordon Ramsey, and then we have topical relevance for topics that are adjacent to Gordon Ramsey, which are things that he is known for versus food in general.

[00:28:19] And so we started with beef, Wellington, a salmon burger, and then we expanded out from there. So back to the example, back to the funnel and the question about where to start, we want to start with, you know, what we're known for, but then we can expand out from that halo. And we can prioritize topics based on opportunity and not just on volume.

[00:28:38] So we, we talked about. The topics being the sum of the volume for all the different keywords. So for gross revenue it's or let's take the types of LLC. So it's the sum of all these, and then it's also where we're topically authoritative. So Brex for example is again, authoritative for LLC. So we pulled out if we were to work with Brex, And we used our, our strategy.

[00:29:03] These are the top, these are the first topics we were focused on. And again, this is based on what's popular, but also Brex specifically is going to outrank for these types of topics faster and an LLC in particular, we want to focus on. So this is how we would do an initial topic structure for Brex and Bernard as a ways to generate more of these types of branded searches.

[00:29:25] Yeah. So how do we, how do we grow more branded search and the way that we do. More branded searches is brand marketing in general. So if we buy a TV ad like Ticketmaster or sorry TripAdvisor we'll do TV ads saying go to Google and type in you know, you're traveling to Santa Monica type in Santa Monica hotels trip.

[00:29:45] Which is a great way to get people to type in TripAdvisor and a city name. And every time somebody types in TripAdvisor, Santa Monica, they rank higher for anything containing Santa Monica and hotels. So brand marketing, TV ads, display ads, YouTube ads, MasterClass buys a ton of YouTube ads. And then when people watch that they go to Google and type in a MasterClass Gordon Ramsay.

[00:30:08] So there's a bunch of different ways to, to grow branded search, but in general, it's.

[00:30:11] So we, we identified the hope, any advice on taking advantage of an exact match domain? Yeah. So exact match domain. Good question and example, there would be let's do chase center, chase center seating chart. And SeatGeek is number one. So this is a good example. So seat geek has the word seats in the domain.

[00:30:38] And what used to happen is that because a SeatGeek has seat in it and the quarry is seating. We will rank for that because it's in the domain and that's actually not true anymore. However We will be because the word seat is in the name. People will type in CP CP yeah, promo code. And we are now getting relevant for seat as a result of this branded search and anchor text and share.

[00:31:03] So if you link to CPQ, it's like CP chase center so that there's not a direct transference of exact match domain. Per seat geek, but there is a secondary or indirect benefit of branded search shares, the anchor text that all contain seats. And so you'll see, seek geek ranks really well for seating charts, even though they don't have anything special about seating charts, relative to Ticketmaster, and the ticket Mac master outranks or anything containing tickets.

[00:31:32] So there's a secondary benefit of tokens in your domain, honey ranks for things containing audience. So. For a period of time ranked honey as a coupon site and they ranked for Honeybaked ham, but they didn't have coupons for Honeybaked ham because honey was in honors was in the quarry.

[00:31:46] So tokens are words in the domain that can impact authority. The indirect signals. And how do you, how do you con talk relate topical authority? So the way that you calculate topical authority, the best way to calculate it is to just look at your search console and like your top quarries. And that's where you have topical authority.

[00:32:03] That's a shorthand for how to measure it. The secondary way to do that is to look at anchor text for backlinks. Look at which pages are shared the most, and then a branded search. Or actually, let's do clear scope and I will look at keyword discovery and I'll type it Brex. And then I'll look at all the different words and phrases that co-occur with Brex.

[00:32:25] So Brett, like credit card here or. Or a business card valuation. So all of these things are where we have a topical authority for Brex. So this is how we can measure it. And then we can measure this relative to others. So how many people type in ramp credit card or Silicon valley bank credit card, and then how do we compare relative to that?

[00:32:45] And that's where we can cut the top 40. Have you considered doing an end grim analysis? Yes. So that's exactly what we'll do. So we'll take search console, we'll do an Ngram analysis. So an anagram analysis is where we break this up into words and phrases. So the word card occurs as many times and business and, and valuation.

[00:33:02] And so, yeah, we'll do a grim analysis to attach that.

[00:33:09] Cool. So we've, we've figured out where the opportunity is we've figured out the right page types for the different intent clusters, we've sized them. We've prioritized which pages we want to rank create in what order. And now we want to make some amazing content and Google says, make good content, which is, which is great.

[00:33:31] And so I'll write an article and I'll say that, I think that is. And nobody, nobody finds it and it doesn't rank that well, and good content is not subjective in this case. It's algorithmically. And ultimately what the bot is looking for. It's a few different things, but it's especially looking for comprehensiveness.

[00:33:50] And let me show you an example with butter lettuce. So with butter lettuce, for example, a MasterClass ranks for butter, lettuce, and the reason why they ranked for butter lettuce is because it's among other reasons it's comprehensive. So if you look at butter, lettuce, and the topic level, it's a ranking for hundreds of different cubes.

[00:34:11] And it's ranking for things of the concepts of them better, less, like, is it healthy or how does it compare with bib or romaine lettuce? Is it good for you? What are the health benefits would have been nutritional values? So people are clearly interested in the health aspect of butterflies. And so when we're figuring out the what content.

[00:34:31] On the on the article, we want to look at the different teams. So for stock options, again employees versus startups versus examples versus incentives, you know, different types of stock. And we want to make sure that. Figuring out the right content outline. And so whenever we're publishing an article, we want to figure out what that outline is.

[00:34:48] There's a few different inputs. So one is what, what the different themes are within the keywords that we're targeting. We can also run a clear scope report and look for themes for that. So here for butter lettuce, you know, we're looking at full lakes, people mentioned full light when when they're writing about butter, lettuce in the health benefit.

[00:35:06] So these are all inputs to tell us what themes are people interested in when they're searching for butter, lettuce, or stock options when, and this is again a key way where we can outperform others. So part of the reason why MasterClass French butter, lettuce over a food network is that food network has gaps in its content.

[00:35:22] It's not comprehensive. And so it's not fulfilling the intent of that whole keyword cluster and a big competitive advantage, as well as that. I think most people only target the, the head term query. So we are only looking at butter lettuce. We're not looking at the health benefits or how it compares with bib lettuce.

[00:35:41] So if we have a great clear scope score for butter lettuce, but we do not for For healthiness. So here are the MasterClass article, for example, has a great clear scope score for butter lettuce, but it has a B minus for helping us. We're not fulfilling the antenna that we, we won't rank as well. So we'll try to typically do two to five different clear scope reports and we'll look for themes.

[00:36:03] So let's get one about healthiness. Let's get one, contrasting it with it or romaine lettuce, and then the head. We only look at butter lettuce, then that we're missing out on, on, you know, all, all these other themes. And and you'll see again, a MasterClass is actually underperforming on the healthiness.

[00:36:20] So if we added, you know, if we optimize for that health theme, we would, we would rank better for that. And and most people don't do that. So I always recommend two to five, roughly clear scope reports, such that we cover all of the different themes of the keywords in that. What do you think about splitting out of subtopic intent?

[00:36:38] Yeah. So health benefits for butterflies. So should that be a separate article or should we add that to the article about butterflies and a good way to look at that is to see if the results are the same for, for two quarries. So are the results for butter, lettuce and is better? Let us help or let's do.

[00:36:57] Butter lettuce and health benefits, butter, lettuce, butter, lettuce, and health benefits. So basically are these all the same results or are they a different results? And you'll see here. Yeah. Most of these are not mentioning health benefits and we see Healthline basmati everyday health here. And it looks actually, they are, they're not overlapping.

[00:37:25] Yeah. So actually in the Ms cases, interesting, we might actually want to break this out into separate article, but if all these results were better, let us and health benefits a better list. If all of these were the same, then that's saying that a similar article can fulfill the intent of these two different ones.

[00:37:39] If they're different. That means that Google thinks that we need two separate articles. So that's how we think about whether or not we have two articles and whether or not we combine it all in the same article. Optimizing a single article. Yes. So question from Olivia is optimizing a single article for multiple keywords.

[00:37:57] Yes. In many keywords. So this article about butter, lettuce, we're targeting 300 different keywords. The options plan, we're targeting hundreds of different keywords. So we're usually targeting at least a hundred keywords per topic or per page. Not, not just one.

[00:38:14] What if there. Or two to three. Yeah. So, so if the overlap isn't 10 results overlapping, it's one result or it's two to three results. How do we think about merging and splitting? And I guess I would say pick a number and then test it and see. So if there's two to three, that's, that's a decent amount of overlap.

[00:38:34] So let's just write one article about butter, lettuce, and see if it ranks for health benefits. And then if it does great. And if it doesn't, then maybe we split it. So I think we would use intuition to this. Whether it's 10 results or two to three, and then test it. And in the case of MasterClass, we actually are ranking for health benefits.

[00:38:52] So in this case you know, maybe we don't need separate. And you do clear scope of port for all of those keywords. So should we run a clear, should we run 300 different clear scope reports? That would be great, but we would need Bernard to give a, I think we need a discount in order to run 300 reports.

[00:39:10] So I would, I would, I would run reports for teams, so we don't need to run a report for both. Is butter, lettuce, healthy and health benefits of butter, lettuce. It's going to be similar or like butter, lettuce versus. Great butter letter. You know, what is by the lettuces, same concept. So we don't need to run a clear scope of port for every single keyword.

[00:39:27] We want to run it for the primary themes of, of keywords. We'll be wondering for.

[00:39:34] Cool. So we have we looked at the opportunity. We built an amazing SEO strategy amazing premium content. And now we want to make the most use out of the visit. We're getting all this traffic now. We want to have it be valuable and it's common for someone to type in this and go here. And get the answer and then leave and never remembered that they've visited the spruce.

[00:39:59] And so the spruce got that visit, but there was no value to that because the user never really that they didn't convert, they didn't buy anything. And they didn't remember where they read the article, they read it and left and that's pretty common. So we don't want that to happen with lot of the visit to be valuable.

[00:40:15] And SEO builds intense. Somebody might, might be able to transact, but they might not be, they may not be ready to transact. They might be wanting to learn something. And so we want to build the intent of the user. And, you know, again, back to this rule of seven framework, it takes multiple times for someone to see MasterClass or somebody to see better up before the.

[00:40:36] So the job of SEO is to build that intent over time. And again, back to back to this, some quarries they're ready to convert some they're not so corporate rates Texas, they're not ready to switch banks yet. And that's okay. So if I don't sign up for. On this page, that's actually fine. As long as I built the intent of the user.

[00:40:55] So I have different goals depending on where you are in the journey. If you're at the end of the journey, I want you to subscribe or buy something at the beginning of the journey. I want you to see our brand and remember it and associated with that keyword that we have. And we build that over time and there's different ways to do that.

[00:41:08] So the typical way to do that on an article is to have a big pop-up and subscribe to my newsletter and that doesn't convert well. And it's not what the user doesn't want a newsletter. They, they wanted to. Figure out what the w w what the corporate tax rate is for Texas. If they want to do this set up an options plan.

[00:41:25] So a newsletter is not fulfilling their intent but a video is so there's, there's different ways that we can build the intent of users. So with MasterClass, we have videos and we'll push users to watch a video trailers. And on the salmon article, we'll get somebody to watch a trailer of the Gordon Ramsey class and a MasterClass.

[00:41:45] This person who did not maybe it hadn't heard of MasterClass. They just wanted to understand how to cook salmon. They watched this trailer. Now they remember MasterClass and they associate MasterClass with cooking, and they're not ready to subscribe to master class today, but when they are ready to learn more about cooking, they think of MasterClass because they've seen this article and six other articles, we built their intent and theirs, their awareness and association with MasterClass.

[00:42:09] And and, and cooking. So video is super effective. And even if the video isn't about, even if this, we don't have the video about Sam, and even if we just have a video in general back cooking, we can still get you to watch that video. And same with preparation Texas, maybe we don't have a video about corporate tax rates.

[00:42:25] We just have a video about how Brett's is a great bank. You can still have that on the page and build the entire user.

[00:42:30] Quizzes. So quizzes works really well. Well, today's webinar. I'll be going on YouTube. I don't know. Yes.

[00:42:38] Ethan: So quizzes work really well. And I am looking for a golf driver and I come here and this site is asking me questions about what I want, or am I looking for women's or men's club?

[00:42:53] What is my height? And this works well because my intent is to get a, get a golf driver that works for me. And these questions help this. Understand what I need. And so it helps me. So it's fulfilling my intent versus the newsletter, which does not fulfill my intent. And so I answer a question here, and then I go into a quiz.

[00:43:14] This, this is super effective and it builds the intent of the user. It shows doesn't tell what curated does security as a site that matches golfing experts or snowboarding. And they recommend what, what products for for, for users. And so we're showing what the product and what the value proposition is versus a big newsletter, like sign up curated has experts, you know, telling you how great we are.

[00:43:38] We're showing you how great we are and it's relevant to the intent that you came in on versus disruptive. So quizzes can work really well. Images work well, too. So one of my favorite examples of five-year plan, so a five-year plan, some people know that they want executive coaching, some people don't and they're just looking to improve in their careers.

[00:44:00] So they type in five-year plan and they read this article and then these images are really engaging and we have this better up logo here. And this is not in your face disruptive, but you're associating the brand a better up. With a, five-year improving in your career, building a fiber. So when you are ready to get a coach, do you think a better up other things are widgets like a mortgage calculator HubSpot has templates and you download them, you get a PDF, you get an email with something you know, better up could have a five-year plan spreadsheet that you download.

[00:44:30] So all of these things build in 10 and better than. Better than the, the email newsletter generic. Pop-up I know we only have eight more minutes, so I'm gonna jump over to measurement. And Kenzie had a question. How do we manage that? Both food type Q is X software to be approach. I'm not actually sure.

[00:44:48] Kenzie sorry about that. I don't fully understand. So how we're measuring this again? As SEO drives impact. So if we don't get results in one month, that's okay. And if we need results in one month, we should not work on SEO. We should, we should work on paid or other channels that are going to get us results faster, but if we can make longer term bets, then we should focus on SEL.

[00:45:10] So typically we'll recommend have a portfolio approach. So you have some long-term bets on SEO, have some short-term on, on ads, have some midterm on social and PR and have a portfolio so that we're not purely dependent on buying. Forever. And if we don't get results in three months, don't panic. So this is, this is another mistake.

[00:45:30] We, we have a great strategy for it. We build all this great content and we're not getting results, you know, in three months. That's okay. As long as we see leading indicators that over time, we're going to see a chart like the green, and we can measure that over time and we can measure ads, SEO, like ads based on ours.

[00:45:49] And we can build an ROI model like this one. So we made an investment and it didn't pay back in month one, but it gets payback over time. And each article or page that we publish it'll increase in traffic over time. So we can basically look at the payback. So if we get a payback, let's say six to nine months on SEO, like we wrote an article.

[00:46:09] Month nine, we got enough visits and conversions or intent from that. Now we know that we should double down and if we're not, if it's taking five years to get payback on each article, then we should spend less money. And same with that. So if we're getting a back, we're getting great payback. We should buy more ads than if we're not getting paid back, we should buy fewer ads.

[00:46:27] So this, we should measure SEO in the same way that we do ads based on, based on our own payback. And again, it's gonna it's going to get paid back over time. And this actually happens. So with SEO, you know, this is not linear. This is a curve, with MasterClass. So it grows over time and it compounds and similar with better up.

[00:46:47] So you'll see that this is sort of curving up as well, and it's compounding and it's multiplying. And even if we didn't get great results at month three, if we had stopped, then we would not have this really nice curve. That probably kicked in around month nine or, or month 12. So this curving actually happens.

[00:47:05] And last is that SEO's not, not mysterious, it's magical. And, and by that, what I mean is that it's mysterious in the sense that Google doesn't give you the algorithm. You have to figure out the algorithm with data and testing it. And. But if you figure it out, it's, it's magical. So if you can get a curve like this with if you can get 80% of your traffic in a curve like this, that's pretty great.

[00:47:24] You know, to do that with email or to do that with any other channel to do that with PR would be impossible. You can't get charts like this via anything other than ads, really by and large. So SEO is no mention of building backlinks. So backlinks backlinks gets into how do we actually grow our authority?

[00:47:44] And there are different inputs of back links getting viral content. That's shared a lot getting more branded search, via brand marketing. All of these things can be effective. I didn't want to go too much into backlink building just cause there's a lot of. Information online, Brian Dean has a good class on backlink building.

[00:48:00] So I didn't, I don't have a unique things to offer on the background building side, but back building backlinks is great. And the way to measure that, the way that I measure that as referring domains and then looking at the anchor texts and seeing what themes the anchor text is around, so that we've not only built links, but we build topical stories and on that.

[00:48:15] And how do you determine which topics work for each stage of the funnel? Yeah. So how do we know whether or not a topic like corporate rate tax? How do I actually know that that's the awareness stage versus the conversion stage, but I would say that it's intuition based. You can also just look at this, you know, the, the conversion rate, but it it's intuition based about where, you know, corporate tax rate, Texas, where in the journey that is we use intuition is Amazon eating into SEO.

[00:48:43] For e-commerce I mean, any site that ranks really well is eating into the market share of other sites and Amazon ranks really well. So the, the answer to your question is yes, but Amazon, you know, Google's not going to rank Amazon 10 times for every single query. So it's not, you know, their, their market share will be.

[00:49:01] Because they're a single company. But you know, the more, any company ranks, the more they're getting in the market share Bernard asked what services do we offer so we can build all this for companies that want our help. So our company what I'm describing to you is what we do. So we build out strategies to build all this and to produce content, to To get charts like that better chart.

[00:49:21] Cool. Any other questions? Any tips on separating reporting impact of link-building authority? Yeah. So Sam, you asked. You asked about how do we actually measure the impact of link building. And it's really hard. There's a lot of noise. So we've got to link to the New York times. What's the impact of that.

[00:49:38] It's really hard to measure that, but the way that you would measure that would be that you silo link building, like we're just going to do corporate tax rate, Texas link building. We're going to get five links about corporate tax rate in Texas, and we'll see the impact. That's the only real way that you can do that.

[00:49:52] But in reality, People could be linking to all kinds of different things. It's hard to do controlled experiments for link-building. It's easier on the branded search side, so you can precisely see on a per day basis, your brand. You can do that in search console, and then you can you'll see the impact within a day or within one to two weeks a branded search.

[00:50:11] So branded search is a lot easier to, to measure an attribute than link building. So link building, you kind of have a faith-based approach to some degree, or you do a purely controlled study, like what it described. Thank you, Jeff. Do you actually need backlinks to rank for keyword? So no, you do not need backlinks.

[00:50:26] Inputs to authority are backlinks, branded search shares, not SEO, traffic. These are all beneficial. If you add no links that you have the other stuff you, you could rank, but not to say you don't have that backlinks. And Bernard asks, do you subscribe to the idea that links bolster a specific page types?

[00:50:45] Yeah, so we got a bunch of links to our listicle page, but not to our category page. So we're all w will our listicle page rank really well on our category page will not rank well, and I don't agree with that. I don't think that that's true. I think that it transfers authority to the domain overall, and then also like.

[00:51:03] Let's say that we had one page that did really well that'll transfer page rank to other parts of the site, and it might be a logarithmic fall off of the authority transferred. But I don't think that it's purely at the page type level. I think that it's at the page and link level. Not more so than a particular.