Excited to have everyone. Looks like we have a good turnout. So I taught the Reforge course before, which is focused on process and frameworks. And this is pretty different.
So this is if we had a roadmap for 2024 for any given company, what would that roadmap be? So this will be very tactical and practical. And the intention of this is that you can basically take this and create your 2024 roadmap out of this. So I'll basically go through the stuff that we've seen has the highest impact across a bunch of different projects and tell you how to do that. And then, Chris, if we have questions, just add them to the chat.
Right. And then we'll go through them at the end. And the Q&A section would be best. Great. Awesome.
So in general, SEO can be a huge channel for. What we've seen is when you have a fully optimized SEO channel, it could be on average 60% of all your traffic is coming from SEO. So it's a huge channel.
Paid is getting more difficult. Cost to acquire is going up. So it's important to unlock other channels. And SEO and organic is a really compelling one and generally performs quite well when fully optimized. Here are a few examples.
you know this is us you have traffic but but again in short it can be a huge channel uh it does take time it doesn't need to take years but it definitely takes some time but given proper investment and and time it could be a huge channel and we are recording chris and we'll share that yep yeah so we'll share and we'll also share the slides at the end as well yep so there's the 80 20. principle of 20% of what you do has 80% of the impact. And we actually see an even more disproportionate distribution where 5% of the SEO work produces almost all of the impact. So an example of this is that it's common to have 100 pieces of content and five articles are driving almost all of your traffic, or you do a whole bunch of SEO work. and there were only one or two things of a long list of stuff that actually drove the impact.
The only real way to know what that 5% is, is to try a bunch of stuff to do analysis to see what correlated with impact, then to reproduce that and to keep on reproducing these tests to see what's driving that impact. What I'll share with you is we've worked on a bunch of projects at Graphite and these are the... specific initiatives that have caused the most impact. And it's very important to prioritize properly to avoid the SEO death spiral.
So the SEO death spiral is where you hire an agency or a director of SEO. They come up with a very large roadmap and lots of data and lots of ideas. You launch some of those ideas and you get no impact. And then engineers start losing trust and then fewer things launch and then everyone... sort of becomes disillusioned with the agency or the person.
And then you get a new agency and a new person the next year. And then you keep repeating this cycle of big list of stuff that seems like it's probably correct, launch stuff, and nothing happens. And so this is the SDR death spiral. And this is essentially prioritization.
So most SEO work is stuff that's true, but zero impact. And so we only want to do the stuff that's impactful. So this is the stuff that's impactful. This is a short list.
Obviously, it depends on the company. But in general, this tends to be the list of stuff that drives the most impact. And I'll go through each one by one.
So technical SEO, I'll start with technical SEO is where people spend, where people waste the most amount of time launching stuff that are again, true, but low impact. Generally speaking, the only two things that actually matter in technical SEO are your internal links and whether or not Google can read the page. There are other special cases where there are technical SEO things other than this list.
But generally speaking, these are the only two things that are impactful for technical SEO. So this is an example of a technical SEO audit. So a lot of times when people are starting out, you get an audit and you get a list of stuff. This is a common list of stuff. And there's literally one thing on this list that's impactful and everything else is zero impact.
So examples are things like HTTP, HTTPS, missing alt tags. You know, you should have an alt tag, but if you add an alt tag, your traffic doesn't go up. Page speed is probably the thing that people spend the most time on that does not drive impact.
Usually, sometimes it does, but usually it doesn't. Yeah, pages have H1 tags. So essentially, this is the SEO tech audit and almost all of it's a waste of time.
It's true, but zero impact. And so. We want to be especially careful with, again, not especially starting off a project with low impact work.
So let's go through the two things that actually matter and have the most impact. So the first thing is cross-linking. And cross-linking has very high impact. So this is a study that we did where we correlated the number of internal links to a given page and traffic.
So what this means is that I have 100 different pages. for each of my pages, how many times am I linking to myself within my own site? So links that are internal. And more is better.
Obviously, there are diminishing returns. So 1,000 versus 100 doesn't really matter. But five links versus one link versus 10 links has a very large impact. It's not enough to just put your content in a site map or link to it from one other place in the site.
You want many different paths for Google to be able to traverse your whole site to crawl sufficiently and to get full traffic. So we usually aim for five to 10 links within a site for any page that we care about. Here's a study.
So again, the way to know what the 5% is of this long list of stuff is to do rigorous analyses. Generally, you don't do rigorous analyses in SEO, which is why you have so much work that actually doesn't drive impact. But for each of these things, we've done a rigorous analysis to validate what we're saying.
So this is a specific example. where we tested this across several different sites and basically found this is using an algorithm that we have, but our algorithm versus current algorithm, we see 42% more impressions. It's pretty substantial, a better algorithm versus less good algorithm.
Usually when we're adding internal links, we're putting them either in the body or at the bottom of the page. It could be either. so this is it's typically easier to add them to the bottom page but you can also add them to the body of the page like this but yeah in general aim for five to ten links if you have one or none or two then you're not going to do nearly as well uh let me show you also so i want everyone to be able to reproduce all of this when you're creating your 2024 roadmap so the way that we you can quickly do this is to get this tool, Screaming Frog SEO Spider, and you can run an analysis and get you to basically run your whole site through this and get a report on the number of internal links per page.
Here's a quick example. So this is branch. And we see here that some pages get linked to a lot, which is pretty common, and then some get linked to a little bit.
And then there are a bunch that have one link or no links. So all these URLs that have one link, no links, are not going to get crawled as much and the traffic is going to be as high. So now we can go back and fix the algorithm to get these more links crawled more and more traffic. The second thing is page rendering. So this is a little bit more obscure, but essentially Google needs to be able to read the entire page.
So here's a quick way to look at that. So basically, Google sometimes has trouble crawling JavaScript, and sometimes it doesn't. So if your page is server-side rendered and everything appears, then you're good. Google can render everything. It's when you're using JavaScript that you want to double-check.
And the specific case you want to double-check for is the case where you'll see here for Masterclass, they're doing... part server side load. So I turned JavaScript off and you'll see that not the whole page shows up. There's a bunch of stuff underneath here. And so we don't know whether or not this got picked up.
A lot of times you'll see this with a product page will have reviews at the bottom and the reviews are loaded in asynchronously with JavaScript and they're not seen or related links are not seen, but the body is seen. And so you basically want to go through the whole page and check to see. whether or not each thing is indexed. The best way to do that is to just grab the URL and to do site colon and then grab a snippet like this.
And then I can see whether or not Google found it and they did. So I'm good to go. But this is surprisingly common where half the page can't be indexed or isn't being indexed because of this hybrid server-side JavaScript rendering. So it's important to check that, especially the stuff lower on the page. Those are, okay, yeah.
So this is an example of a project that we worked on where something wasn't getting indexed properly. We fixed it and then traffic went up quite a bit. So this is a really easy, quick fix.
If Google can't see part of the page and you fix that, now they can. That's pretty impactful. So technical SEO, pretty simple in terms of the stuff that actually drives impact. recommendations are make sure you have enough internal links for each page um and then make sure that google can see everything on the page if you're using javascript and that's pretty much it rejecting less you uh editorial seo so editorial seo is probably the most impactful general strategy programmatic is the second but editorial more and more is is most impactful So when we say editorial, we basically mean articles, resources, things that a human sat down and wrote versus a product page where it's grabbing information from a database. The programmatic SEO is what we're referring to as things like product pages, category pages, search pages, stuff like that, that's grabbing information from the database.
And then editorial is where a human sat down and wrote content. And then technical is the infrastructure piece underneath that. And this is the list. of impactful stuff.
So the first thing that's important to note is that SEO is not targeting keywords, they're targeting topics. So by that we mean it's a group of keywords, each page that we have is actually targeting many different keywords, not just one. So this masterclass article, for example, is targeting about 700 different keywords, I think 300 different keywords, all these different variations.
And so as we're building our strategy, we want to know all this stuff underneath our topic, not just butter lettuce. Here's another example with BetterUp. Their five-year plan article is targeting personal five-year plan template and example.
And this is important because these are different subtopic themes that we might not know about. So if we wrote a five-year plan article and it was all about what is a five-year plan, but we did not have a template and no examples, then we're not fulfilling the intent of what the user wanted. And we're not going to rank as well.
And we're not going to get as good conversion engagement. So it's really important to understand what these subtopic themes are for the keywords that are part of your topic cluster. So the next thing is topical authority.
So Google started with page rank and backlinks and stuff like that. That's still true. but authority is more complex and it's topical. So you might be really authoritative for electronics, but you're not going to rank for food queries or vice versa. And so we want to look at not just domain authority and keyword difficulty.
We want to look at that at the topical level. So for example, here, our domain authority is really high, but Masterclass is not going to rank for best digital cameras because that's not what they're topically known for. So how we can look at that.
Okay, so we also did an analysis to validate this. And Google recently came out a few months ago stating that their authority score is topical. So we know that this is true. And we have data validating that this is true.
So what we saw here is that a high topical authority score for a given topic that I go after versus low has is much, much slower and ranks less well. And the earlier a company is in the The less domain authority you have, the more this matters. So if you don't have a lot of domain authority, but you're known for something really specific, you can actually rank for that ahead of someone with a lot more authority because you're topically known for that. And Google wants niche domain sites ranking for stuff above these very large authority sites so that you don't have the same 50 sites ranking for everything.
So the earlier you are, the more important it is to look at your topical authority. to prioritize based on that and not based on keyword difficulty or long tail. So how do we look at topical authority? An easy way to do that is to go through your...
Google Search Console and to download your various keywords that you're ranking for and to basically look for patterns of keywords that you're ranking for. I included a tool. You'll see this bitly link here at the bottom is a tool to do this. So that's this here.
And you basically download your various keywords and then you look for these patterns. And so what this is saying to us is that Google thinks that we are, this is for Graphite's website, Google thinks that we're topically authoritative for SEO, agency, growth, marketing. And so as we're looking for new topics and new keywords that we want to target, we should look for stuff that's as similar as possible to these things. If we instead looked at something like paid media, and we have no topical authority for paid media, we're not going to rank for that.
Even if it's a long tail term, we're not going to rank for that. So we basically look at our search console data. summarize it in a tool like this.
And again, this is linked to this, you can use this, and then try to come up with ideas that are closely related to this. An example for masterclasses that early on, before they had done SEO, they had authority for the names of the celebrities, like Gordon Ramsay. And we ranked for Beef Wellington, which is similar conceptually to Gordon Ramsay.
But we did not rank for Butter Lettuce, which is not. adjacent or close to Gordon Ramsay conceptually. So again, we just took the authority that Masterclass had, which are the names of the instructors, and then found stuff that those instructors are uniquely known for. So when you're starting out with your strategy and the earlier you are, this is especially important so that you can rank with less authority and then you can build a fly wheel. So Masterclass ranks for Beef Wellington and then kind of saturates food and then eventually they can rank.
for butter lettuce because they've uh they've built out early on from the top topical authority and that topical authority is expanded the the next thing is content enhancements so if you already have content i would start by enhancing the content that you already have before doing net new content this is an example of a bunch of different content enhancements we did and essentially what we're doing is we're going through the content that we already have and finding for uh for this example with better up we're looking to see are we ranking for template and examples and we don't have a template and we don't have examples and if we don't then we should add examples or maybe we could add to the title tag uh the word template because people are looking for that so basically go through the content that you already have live you and then try to find keywords that your content's ranking for but you're not you're not uh mentioning and add those sections to the body and add those sections to your title tags one example we did was we had a bunch of resource content about how to do stuff and we saw that we were ranking for templates generally but we did not have templates in the body or in the title so we basically went back and added templates to our top performing articles and add a template to the title tag. And then we started ranking the templates. And that was like a 25% plus, 20% traffic increase.
So if you have content, go through everything that you have right now. Typically, I would do your top articles versus everything and find stuff that you can add to the body and to the title. So yeah, this is an example where we doubled traffic by doing content enhancements. So very high impact. And the impact we'll see is within a few weeks, not months.
So to summarize, editorial SEO is probably typically the number one long-term growth strategy in SEO, even more than programmatic. And top recommendations are go through your top performing content and make it better, enhance it, find stuff that you haven't covered. add to the body, add it to the title tag. The second is do topic research, not keyword research. Look at what you're topically authoritative for on a site-wide level, and then find topics that are similar to that and prioritize based on that.
If you're not sure whether or not a topic has intent or not, we've seen that CPC strongly correlates with intent. So you could just look at the CPC and you'll know whether or not. that topic is likely to drive conversions or not.
This is for new content. And then I didn't cover this fully because this is sort of a separate subject, but For editorial SEO, it's common for your conversion rate to be really low. And the reason why the conversion rate would be low is because you didn't do conversion optimization. So as you get traffic in editorial, assuming your conversion rate's low, I would spend time on conversion rate optimization. And I did a class on that, which you can review here.
That's sort of a deeper dive, but I don't probably have a scope for this presentation. Okay, so jumping into programmatic, this is the list of stuff that typically drives the most impact category pages, product pages, and then all the things related to that. So the first thing is curating the pages, what pages are indexed. And a common issue is that I might have thousand like five four hundred thousand different pages about towels and there's too many towel pages uh you don't there's coastal towels and modern bath towels and traditional bath towels rustic bath towels some of these might be good but a lot of these no one's looking for and are not getting traffic and they're pretty similar to each other and there's hundreds of thousands of them and so especially with commerce sites or marketplace sites a lot of times there's just too many pages and we don't want to have too many pages And what this says to Google is that our domain quality is not that great because we have all this automated content that people aren't engaging with. So how you can look at that is you can go through your various page types and look to see how many pages do you have live, how many are getting crawled, how many are getting indexed, and then ultimately how many are getting traffic.
where the gap is, you can diagnose that. So if you have a million pages, and they're not getting crawled, then you're probably not linking to them properly. And so you want to improve your internal links, like I mentioned at the beginning, if they are getting crawled, and they're not getting indexed, maybe there's some indexation issue, like weird redirects, or canonical tags that are broken or things like that. Now, if they're all getting indexed, but they're not getting traffic, then that's a very different problem.
And that's typically that you're not linking properly, maybe the content's not that interesting. And then a lot of times, it's that the page just isn't popular. So you might have a case where you have a job site like Upwork, and you say, Python developers in really obscure city in Texas, where there are no Python developers, and no one's hiring in that city. And, and you wouldn't want a page for that.
Because again, this percentage score of I have a million pages, but only 5,000 of them get traffic, tells Google that my domain quality is not that good. And that million pages actually harms the 5,000 good pages. So in general, you want to curate which pages are indexed and not, especially based on are they getting traffic?
Are they empty? And do they have search volume? I did a longer article on this here, which I'll link to. for further reference but um but yeah again uh indexation logic to curate what pages are indexed on your site um the second thing is uh category pages so category pages still do quite well i'll show you what i mean by that so a page like this where you have a list of people you have a list of products you have a list of restaurants hotels in a particular city and this is a common structure the uh the thing that matters a lot here is adding page level content so adding guides essentially to category pages has high impact the reason why is that there's not that much content in these grids and everyone's grid is kind of the same and you can't cover everything uh by just listing more jobs. So adding category page level content is a really good way to get category pages enhanced.
People are currently trying to do AI generated content and templated content. I think it can maybe work, but I generally am not bullish on that, at least not yet, because I haven't seen people sustain traffic for long periods of time. Some people are creating AI content and then they go up and then down. I think eventually there might be automation opportunities here, but generally I would recommend not automating and doing manually written editorial content. and then internal links so i talked a bit about internal links but we want to again cross link to all my other category pages so that they get crawled and they get traffic so this is a viral post that went out about i made a bunch of ai articles and my traffic went up however the traffic went went back down and you'll see this a lot with sites.
This is another example where you'll see houses going up and down these huge swings. This is basically the algorithm scoring houses domain overall saying that they're low quality, and then they're not low quality, and then they are low quality. In this case with houses is more related to all those hundreds of thousands of bath towel pages that I showed you. But it's common for something to work for a few weeks. And then the algorithm reruns and says this domain is not good or these articles are not good.
So that's what happened with this example. I do think that AI-assisted content can do well, and the AI content's pretty early on, so I'm not discounting that it can never work. But right now, if you just make a bunch of AI content, I probably wouldn't have that be my top recommendation. Title tags as well. So we talked about title tags for editorial, similar with programmatic.
So for programmatic, again, I would go through my top category pages and try to find themes of stuff that's not in the body and not in the title. So one example that we had was people were looking for templates and we want to add templates. So we added templates or sorry, people were looking for examples. So we added examples to the title tag and traffic went up by 12%.
We literally just added one word to the title tag because, again, it was popular in the keywords that were ranking for our category pages. And we. within days started ranking higher. In order to come up with your title tag ideas here, you can use the same strategy here where you basically go through all your different category pages and run it through this tool. And then you'll find some themes of stuff that you're ranking for, but you haven't added in the title and then you add it to the title.
Great. So to summarize for programmatic SEO, make sure that your domain doesn't have a majority of it with pages that are not getting traffic and that are low quality and very similar to each other. So curate the programmatic pages that you do have, test and optimize title tags, add content to category pages and to any other programmatic page type. And then the third is to explore new programmatic page types. So the way to do that is to basically, if I'm competing against Rippling, basically go through a tool like Ahrefs or SEMrush and try to find themes for pages, page types and pages that they're ranking for and we're not ranking for.
And that's how I come up with new ideas. So maybe Rippling is doing calculators and widgets and I didn't do that. And then I look at another competitor like deal.com and then I essentially look at my competitors. What's the gap between my strategy and theirs? And then you can come up with new programmatic ideas by doing this competitive analysis.
So the last thing is the SEO journey. A lot of times your strategy is going to be low impact and you come into this SEO death spiral. But we want to avoid the death spiral by selecting the 5% of stuff that's impactful. But even if we do that, it takes time.
And it's important to keep on. repeating the SEO journey and narrative to the broader team so that people don't prematurely conclude that things are not working. So if I buy an ad today, I'm probably going to have results within the next day or within the next week. And I can easily say this ad caused this amount of revenue and therefore I should spend less money or more money.
In SEO, you usually don't get an answer the next day and it might take weeks or months. So typically it takes a few months to start getting traffic. And you had your first cohort of traffic, and then you're ranking number eight for a few terms, and you're getting 50 visits a day.
And then you start ranking number three, and now you're getting 1,000 visits a day. And then you have more and more content coming in, and those are sort of summing up. Now you get traffic, but then the traffic on the blog is low conversion rate. So then you go back and you do conversion optimization, and now you start getting conversions in your editorial section. And your programmatic SEO changes that you made are starting to kick in.
And now it's growing over time. But this stuff takes months typically. And a lot of times... people will forget people will forget that and so it's important to constantly reiterate the journey where are we in the journey and what's next in the journey so that we don't prematurely stop working on seo cool so uh let's jump into questions um cool um i can uh i can call on folks that that ask questions that have uh the most upvotes and then we can kind of go back in time uh chuck you just asked a question uh about editorial seo do you want to come uh unmuted and ask your question sure yeah hey chris hey ethan um yeah i was just asking about length if you have any opinion on the length of editorial content we recently wound up working on a piece that wound up being thousands of words long and i was just wondering We wound up splitting it up and doing a lot of interlinking between four different posts, but I wasn't sure, I don't know, what you thought about length and whether that really has an impact on how Google ranks content. It does, but it's not just a straight word count.
Also, thanks for your question. So it does, but it's not just a straight word count. You need 1,327 words.
So what Google wants and users want is you want to sufficiently fulfill the intent of the query. So did you cover everything that I had questions about? And if you did not, like you're missing templates or you're missing examples, you should add that. Typically, we'll see a range of 800 to 1500 words for pages that rank well, and that's sufficient. Sometimes you need more than that, but typically 800 to 1500 words should suffice.
Even if competitors are writing more words, we generally see that just adding another 500 words doesn't have impact. What has impact is sufficiently. covering all the questions and subtopics that users had. Now, if you had 10,000 words, it's not harmful, but it's probably not necessary to fulfill the intent of a single article. Thanks, Chuck.
Next question is from Donvir. Donvir, would you want to unmute and ask a question? Yeah, I think you already answered this question in the slides, but I was just pretty curious about your take on not using AI tools to generate content.
You had mentioned that if I use AI tools, it'll go up a little bit and then it'll come down again. I'm just trying to understand your reasoning behind not using that. Is it because Google knows that it's automatically generated by an AI or is it grammatical errors or things like that?
As long as it's a structure and the English grammar is correct, shouldn't that be okay or am I missing something here? Yeah, so let's assume that AI content is 100% accurate and grammatically correct and spelling correctly and everything's great. And, you know, we're either there now or we'll be there very soon. So will that suffice?
Today, I'm skeptical that it will suffice for two reasons. The first reason, but I think it might in the future. So the first reason is that AI content generally is very similar to content that's already been written. And Google generally doesn't like that. Like in 2011, Google launched a bunch of quality algorithms to say, are you like if you're a shopping comparison site, are you scraping other people's content and repurposing their specs and reviews?
And Google generally doesn't like sites that grab content from somewhere else and remix it. And they're basically looking at similarity. So is your content very similar to something that's already online?
And if yes, then you get a bad quality. content score. And that's been around for 2011. And most AI content is very similar to content that's already been written. Now, you could force it to be more dissimilar, but I think that's one component. And then I think the second component is Google's generally building algorithms to probably to ding AI content.
I'm not actually sure about that. And there's not public information. And I haven't tested and validated that unique, high quality, accurate AI content won't work. I don't know. I also don't think Google really knows yet.
I think that they'll figure it out over the next few years about what their perspective is. But right now, I wouldn't just straight use AI with no human intervention at all. What I would do is I would do AI content and then give it to human editors and have them massage it and curate it and make it unique.
That could go quite well because it's cheaper than something that's written from scratch, but it still has a human in the loop. Got it. So uniqueness is more important here than pretty much anything else, right? As long as you can figure out that balance between the uniqueness.
Assuming that it's high quality, accurate, written well, then uniqueness, I think, is very important. If it's unique, I think it can work. If you have a human in the loop editing it, I think that it will work.
Got it. Thank you. Thanks.
All right. So our next question is from Ben. Ben, did you want to come on mute if you're still here?
Sure. Hi, thank you. Yeah, I was just curious, you mentioned about the internal links.
And you mentioned, you found that a good number was eight internal links. From what I've seen, sometimes we link from, you know, multiple sections in the same article to other pages. Did you mean or did your findings show that you needed unique pages?
So eight unique pages to link to one unique page? Yeah, unique pages linking to you versus links versus five links from same URL. I haven't specially analyzed that, but generally speaking, I would assume that it's unique.
Sorry, I know that unique pages, unique inlinks is what it's called in ScreenFrog. Unique pages linking to my page, we know that that matters. I have not separately tested five links in one URL, but I think I have. but I do know that this works and I would expect that this matters more than five links from one page. Right.
Okay. Thank you. Awesome.
Next question is from Andrew about external linking and backlinking. Andrew, did you want to give any more context to your question? Sure. Hey, Ethan, long time no see.
Backlinking used to be a critical part of any SEO strategy. You didn't mention it at all, which is striking. I'm curious to get your take on that.
So authority matters a lot. Backlinks do matter a lot. It's usually, it's a question of ROI.
So how much money do we want to spend per link relative to all the other stuff that we could be working on? And usually what we'll see is that a site with a thousand plus referring domains can actually rank for a whole bunch of different stuff without doing any link building. I'll show you a quick example.
So here's a site with 800 referring domains when we started 300 referring domains so very low authority and this is their traffic curve and they can rank for coaching sessions number one now if we built links would we get more yes we would get more but you can get a lot out of just leveraging uh your current topical authority and building around that having said that you'll eventually cap out probably So there's many different ways to build links. Usually I'll try to recommend not spending money. So the typical option is you pay an agency, you give them $3,000 a month and they go get you links from other, you know, entrepreneur.com or whatever.
That is impactful. You could do that. Usually when we do link building, we'll try to suggest something like do viral content or try to get your community sharing your stuff or stuff that doesn't have a cost. and that's usually what we'll recommend so i do agree that it's on the list of high impact stuff but it's usually not worth the money to do relative to all the other stuff that i shared awesome um question just came up in the chat which i think is is relevant ethan uh carolyn would you want to come on mute and ask the question uh that you just chatted in Yeah, sorry, took me a minute to go on mute. I don't run SEO, but some other members on the team do.
So I'm curious, as a business leader, how often should I be reviewing their work and say this is working or this is not working? So that's the first part of the question. Second part is based on what key dimensions should me, who obviously didn't do the work, but ask the right questions to say.
hey, based on XYZ signals, I'm curious if we're actually seeing success. Yeah. So it depends on where you are in your journey.
So if you're just starting out, it's going to take months to get early data about is this working or not. But the data that you would look at early on is, am I ranking competitively for... five different pages for 20 different terms and i'm ranking number 12 and number eight and i'm not getting much traffic but i'm at least ranking for something and there's some indication that this can grow over time so that's what you would measure for phase phase two here or phase one now then i want to curve up and get more and more traffic so then i'm going to be looking at visits and for visits i'm going to want to get 100 to a thousand a day before uh i you know This is meaningful to me. Then when I have 1,000 a day, I want to look at what's performing and what's not performing and do more of the stuff that's performing. And I'm going to be tracking visits, and I'm going to have goals around visits.
So now early on, when you're just starting out, I would check in monthly, but I wouldn't expect results until maybe two to three months after launching stuff. As traffic is growing, I would check in probably monthly as well and make sure that visits are increasing. Then as visits are increasing, I want to do conversion optimization and I want to look at conversions. And I'll probably also be looking at conversions probably on a monthly basis. Does that answer your question?
Yeah, that's super helpful. And I'm curious also as a mature business, not mature, let's say, if your business is already viewed as number one, number two in the industry, market share of whatever, 30%, is it fair to say your website traffic should... aligned to your market position to some extent. Let's say if your market share is 30%, but your web share across your competitive set is 5%, that is sort of a top-line signal that something needs to be done, right?
Yes. Yeah, so if we're a big brand and we have a lot of authority and other commerce sites are getting half their traffic from SEO and we're getting 5%, that means that there's a huge opportunity. Thank you so much.
Cool. Awesome. Ethan, we have time for probably two more.
So next question is from Honia. Honia, sorry if I'm mispronouncing your name. Would you want to come on mute and ask your question? Going once, going twice for Honia. Well, I understand the question.
So how would you do internal links? And the question is, would you do it based on categories or would you manually do it? So this is an example of doing it based on categories or facet-based links.
So Thumbtack Plumbers in Santa Monica. So other cities, which are tags, or other things within same city, other services within same city. So this is a common way to do tag or facet-based cross-linking.
This is impactful. However, we've seen that you'll basically have silos. Same with Upwork. So other in-demand skills.
And I think we did stuff in Citi. So then if you add cross-linking that's basically spread better, you'll get like 20% plus increase on top of these tag-based links. The last thing you can do is add stuff in the body. You can do that, but it's just a very manual process. Like Masterclass would have hundreds of hours spent adding links in the body, which is great, but it's just a lot of effort.
So. I would generally build an algorithm that does this automatically and make sure that your links are spreading so that all your pages are getting a lot of links versus silos or versus just a few. Awesome.
Last question of the day. How can we improve topical authority from MR? There's a few ways.
So the first way is... The best way is a masterclass. writes an article about Beef Wellington, they start ranking for it, they get good engagement. So people are searching for Beef Wellington, clicking on the article, it performs well, that Beef Wellington article gets more and more traffic. And what's next to Beef Wellington is steak and other types of food.
So basically, starting with where you have topical authority, doing something very similar to that, and then sort of branching out from there. Or if Graphite wanted to rank for paid marketing, and we're not known for that, we start by creating stuff around best SEO agency. And then we rank for that and we get more traffic there. And then we eventually do rank for stuff that's further away.
So again, to summarize, create content that's adjacent to your topical authority, get good engagement and then expand from there. However, you can be more direct and more precise. So to Andrew's question, you could build links. So if we...
If we really wanted to rank for paid marketing tomorrow, we basically just create and get a bunch of backlinks that say paid marketing anchor text point back to us. Or if a commerce site wants to start ranking for towels, they have no topical authority, you could do PR and paid backlink stuff around towels specifically, and you'll start ranking for it that way. So it's these two different paths.
Awesome. That's all the time we have for today. Ethan is teaching a topical SEO course in February.
If you go to reforge.com slash courses, you can check it out. We'll also send out the recording by the end of the day Monday with links to the slides. And Ethan, any last messages for the people left in the room before we sign off here? I think the main message is, again, reiterating that the majority of SEO work is a waste of time and prioritization is super important. especially at the beginning of a project, make sure that we can prioritize those few things that drive most of the impact and only focus on those things.
Obviously, everyone's site has differences, but in general, these are the things that we've seen cause and correlate with impact the most. I just included the link to the slides as well. So hopefully you can go through these slides and there are links off to internal link analysis tool, topical authority tool. So hopefully you can use these tools and the slides to build your 2024 roadmap.
Awesome. Thanks so much for your time, everybody. And Ethan, we'll see you around.
Awesome. Thanks.