Melissa (00:00):
Go. Okay, I’m going to kick it off.
Arsen (00:02):
Yay.
Andrew (00:05):
Melissa, you’re already doing better than Ashley.
Melissa (00:06):
Thank you. No, but she is my role model, okay?
Andrew (00:12):
This new Ashley is awesome.
Melissa (00:13):
Well everybody, welcome to the 29th episode of SEO for Bloggers. Today we’ll be talking about how to come up with new topics that’ll drive traffic. We’re with our experts today, Casey Marquee, Arsen Rabinovich, Ashley Segura, she’s there but you can’t see her. And Andrew Wilder. I’m Melissa Rice. I am your very new, nervous but excited host. And so thank you all for joining us. Just a heads up that we will be having a Q&A at the end, so please feel free to drop any and all questions in the Q&A section below. And if you haven’t already, we’d love to know where you’re tuning in from, so please message us in the chat box. And yeah, let’s get chatting about topic ideation.
Arsen (00:55):
Woohoo.
Melissa (00:57):
Arsen, I’m going to kick this off to you first. A question we get a lot is people want to come up with brand new topic ideas. So how do you know if a topic is going to give you traffic and is there research someone can do before they get started writing to ensure that a topic is going to be worth their time?
Arsen (01:17):
Right, so awesome question. So obviously if you’ve taken the courses, most of you should have taken the course, The Cooking with Keywords course by now, and that will teach you on how to, not so much ideate, but more figure out if your topic actually has demand. If the keywords for your topic actually has demand and any keyword research tool will do. You’re not looking at actual numbers, those numbers are not very accurate. Most of the time, you’re looking at it as a point of reference. Is there actual demand? You can also use tools on the Google Trends and Keysearch, Ahrefs, whatever tools. But once you’ve identified a topic in a handful of those keywords, you want to go to the best free tool that any blogger has access to and that’s Google, put in that query, put in that keyword and actually see, what’s ranking?
(02:11):
You want to make sure that A, the intent aligns with the type of content that you produce. So if you put in as an example, tacos, Google’s going to show you something of a kitchen sink result. You’re going to see the history of tacos from Wikipedia, a bunch of taco videos. Do you want to order these tacos? Do you want to have them delivered? Do you want to make them? So you’ll only have opportunity for maybe a handful of blogs, three or four appearing on page one or within the top 10 because there’s no more page one, within the top 10. So always go to Google to check your keywords, to check your topics, see how Google is handling it. I give advice on my coaching calls with my clients to focus on keywords that have the word recipe in them. So add the word recipe and check those volumes because we can safely assume that every time you add the word recipe at the end that that will be a recipe result because that’s what you’re asking Google to show you. So that’s what I would do.
Melissa (03:16):
Right on. I feel like I’ve heard a lot of this on our [inaudible 00:03:22]. Ashley, next question to you. And this is another question we get a lot on all the consultations I’ve been on with Arsen especially. How often should you focus on creating new content and coming up with new topics versus updating your existing content?
Ashley (03:39):
Yeah, definitely. This has changed a lot. It used to be just pump out as much content as you possibly humanly can on a regular basis and at least have something like twice a day. For big bloggers, they were doing so much content. Now things have really changed and instead of focusing on constantly creating new content, we’re always trying to recommend to our clients that they should focus on updating what they already have, focus on what’s working, what is potentially almost working, what has really good ranking potential, but just isn’t there yet, and really optimizing the content that you’ve already put resources and effort into. I know we all like to get really specific here. So from a number percentage, safe rule of thumb is 60/40. 60% could be new content, 40% is working on the content that you have. This obviously can change as far as percentage based on who you are as a brand, and it could be a little bit higher, a little bit lower, but thinking of it like the old days of 90% new content and 10% just updating your existing has definitely changed, and we much rather you focus on updating and optimizing the content that you’ve already put resources into that actually has a chance to rank.
Casey (04:56):
I would just add very quickly here, one of the themes that we see when we do all these consultations, and it’s not just me but most everyone on the call, is that you’re going to find that 70% of your traffic is from a very small percentage of your content. When I do an audit and I immediately visit with a food blogger and they have 1,000 posts and they’re only averaging maybe less than 50,000 sessions a month, we know that there’s a disconnect between that content and the intent or the content and the quality, because there’s no way they should have 1,000 posts and be generating a relatively small amount of traffic like that. And so we always want to go back and understand that the days of your site being a book in Google are over. Ever since 2011 with the content farmer panda updates, and now with the helpful content updates, your site is not a book that we need to continue to add pages to. Your site is something that we need to really start to audit competitively on a regular basis.
(05:47):
Less is more. We can actually generate 10 times the traffic from 10 times less the content. So it’s really focusing on making sure that the bottom line recipe and the intent that you’re publishing matches and that we’re not just publishing on a formula. “Oh my god, I’ve been told I need to publish two or three times a week and blah blah blah. I need to continue to add new content.” I hear that a ton and it’s also a theme that we’re seeing in a lot of relatively low quality SEO training out there in the recipe and lifestyle space. So I’d urge you all to ignore that. Focus on the bottom line quality of your recipes, and always focus on going back and updating content on a seasonal basis as needed.
Melissa (06:27):
Okay. Andrew, if you’re going to update your existing content, how do you figure out which pieces should be updated and which ones you should always leave alone?
Andrew (06:37):
I think if you’ve been watching us for a while now, you know if it’s in the top three position, don’t touch it please. If you’re ranking really, really well, there’s only one way it can go if you change things and that’s down. So I think that’s the very first thing. We hear from a lot of people who are like, “I’m ranking number two.” And we’re like, “Great, that’s it. Stop there.” From an SEO perspective, if you’re ranking top three, you are winning. It can change the amount of traffic you get, but really, that’s winning. So that’s the first rule is if it’s in top three, don’t touch it. Beyond that, I’d say if you think your content sucks and needs to be updated, then it’s time to update it. You all have those cringey photos from when you started and you want to work through that.
(07:20):
So that’s the low hanging fruit, the stuff you know just off the top of your head is really needing to be updated. And then from there, you can start to really dig in and see what’s in the results of 11 to 20, where if you get a lift, you’re going to get higher up. That’s not necessarily as impactful now that we have the infinite scroll, but that’s where you want to start looking. Moving from position 11 to position 5 is going to give you a big lift compared to position 50 to position 40. So you want to start looking strategically at where that content is that needs a little extra lift. I also want to drop in a couple links. We did an episode on updating seasonal content a little while back, so I’m going to paste that into the chat. Oops, that is adding a whole tracking thing. Give me a second. Thanks, Google. There’s that. And then we’ve actually compiled on the NerdPress site some other info about updating and republishing content, so I’m just going to drop that in there for you to peruse at your leisure.
Melissa (08:19):
Thank you. Next question for Casey. I feel like there’s a lot of conflicting information from people online, but what search volume and competition level should someone be looking for in keywords as a new blogger who isn’t very competitive yet?
Casey (08:38):
And that’s a good question, and I believe that Arsen’s going to provide some answers, some knowledge to this as well. But basically the simplest answer to this is to aim for a minimal volume of about 100 to 1,000 searches per month for most new bloggers. This varies wildly though by industry in 10 or seasonality. Information keywords, for example, tend to be easier to rank for than transactional keywords. So in the lifestyle and the recipe niche, we tend to focus more on informational keywords, which is funny because food bloggers would absolutely vehemently disagree that it’s easier to rank in that niche. But the thing to understand about these search volume and competition metrics is that seasonality plays a large amount in which of these keywords to go after. And a good example would be the keyword of mint juleps.
(09:29):
It’s a keyword that averages less than 8,000 searches a month. Basically every time of the year except May when the Kentucky Derby happens and it jumps to 200,000 searches a month. The reason I use that example is that there’s plenty of sites who rank throughout the year for mint juleps when it’s only at 8,000, but when the Kentucky Derby comes up, Google tends to filter out some of this and they change the top 10 considerably. It’s very volatile and other sites tend to jump up there based upon clickthrough rate, based upon the history of the keyword phrase, based upon bottom line quality on the recipe. When we talk about competition-wise with things like search volume, and you’ve heard, for example, we said what does mean with keywords? Well, when we talk about the metrics for SEMRush or Keywords Everywhere and Ahrefs and others, they use this scale.
(10:23):
Many of you have seen it, it goes from zero to one. Well, what does that mean? Well, that’s the level of competition based upon Google Ads. So if you have a keyword that’s a 0.9, that’s an incredibly competitive keyword with regards to the bidding that would occur if you were to buy that keyword on Google Ads. If instead you were using something like SEMRush or Keywords Everywhere and you installed the plugin and you saw, okay, this keyword has a 0.20 competition metric, that would mean that the keyword is 0.2 competitive between 0 and 1. So it’s not as competitive as are the keywords of it is still pretty competitive. And it’s a really interesting metric to use because it’s supposed to be an examination of how many competitors are competing for that keyword, but it’s hardly ever true in the recipe space considerably because no one buys keywords for recipes.
(11:11):
That’s just not something that’s done, although some people have played around with it. But when you talk about the level of competition, especially for a newer blogger, we see a lot of advice to go after, for example, zero volume keywords. And I know that’s a question I believe that someone has just asked. The thing about zero volume keywords or keywords that are either so new or so small relatively or maybe so seasonally influenced that they look like they have no competition, but they really do have a lot of competition and that’s very common regardless of which tool you use. So never say, “Okay, I don’t want to go after this keyword because it doesn’t have any volume.” It’s very possible it does have volume, it’s just that it’s a relatively new keyword, or the keyword itself is so seasonally focused that the keyword data is not absolute.
(11:58):
In other words, you might have to wait a couple of months for a graph to build and for the tool to come back with the full information data. So answering a couple questions here, if you see a zero value keyword, it doesn’t necessarily mean that it’s bad, it just means that that is a relatively new keyword or a keyword with a lot of volatility and that shouldn’t necessarily exclude you from using that keyword. So again, a summary here. If you’re looking for keywords as a blogger, as Arsen mentioned earlier, we do recommend Cooking with Keywords. Alexa now has a Cooking with Keywords for food and lifestyle bloggers. She now has a more general keyword course for just non-food and lifestyle bloggers. She covers this in extremely excruciating detail where she talks about how you can use the relative strength of your site, the age of your site, and these various tools to understand what keywords you could possibly rank for and which keywords you can’t rank for and adjust accordingly.
Arsen (12:54):
So I’m just going to quickly my own 2 cents on this. So with what I’ve been seeing, and a lot of you look at keyword difficulty as a metric to guide you whether you should be writing content around that keyword or optimizing towards that keyword, it’s an okay metric to look at, but don’t let it discourage you. Again, I see a lot of keyword difficulties that are 60 and below with high search volume keywords that are fairly easy to rank. At the same time, I do also see high volume keywords with keyword difficulty that’s 40 or 30 or even 20. And then when you take that keyword to Google, you see who’s ranking there and you’re like, “There’s no way I’m going to make it there,” because you have some of the biggest bloggers, you have some of the biggest sites cooking.nytimes, all recipes ranking there.
(13:52):
So keyword difficulty as a metric is not as accurate, but it should be used as a point of reference. I typically recommend if you are going to be doing topic research and keyword research, I typically recommend looking at anywhere, I would say about 6,500 plus monthly search volume and you can drop it down to around three and a half to 4,000 if you’re just starting out. Again, these are not very accurate numbers from any tool, but if you are going to put in the work and create the content, shoot the content, cook it, might as well already do it for something that has a higher potential of earning you money and driving more traffic to your site, even if it’s not going to be instant gratification for you right away. But yeah, keyword difficulty should be just a point of reference, should not be the deciding factor for you.
Melissa (14:45):
Okay. All right. With all that keyword research in mind, Ashley, what is your strategy when you’re coming up with new topics?
Ashley (14:54):
Yeah, I mean, keyword research is definitely a great place to start and that’s where most bloggers start. There’s a couple other strategies though. One of the things we like to do at TopHatContent is do content gap analysis, which is a fancy way of seeing what your competitors are ranking for that you aren’t, and then picking through all those tumbleweeds and seeing what’s actually relevant to you. So say you’re a vegan blogger and one of your competitors ranking for a cake recipe that uses eggs. That’s not necessarily relevant to you, and so that wouldn’t be worth creating new content. But say they’re ranking for a vegan birthday cake recipe and you aren’t, you have any content on it, that’s a great indicator that okay, clearly if this is your direct competitor, you guys share a very similar demographic, you share the type of person who’s heading to your blog and reading your content, so it makes sense for you to create a recipe.
(15:48):
And so you can get a lot of new topic ideas just from seeing what your competitors are doing well. You can run competitor gap analysis yourselves, but if you have SEMRush, there’s another secret trick which if you go inside the Topic Research tool, we’ve talked about this on this webinar before. It’s my most favorite and most frequently used tool out of all the SEO tools other than Google Analytics of course. Again, it’s called Topic Research tool. I’ll drop a link into, excuse me, the chat box, but what it does is you can put in a keyword, so say your competitor is ranking for that vegan birthday cake and you’re thinking, “I think that might be something worth talking about and creating new content for,” you can drop the vegan birthday cake into this tool and it’ll populate the top 10 articles that are currently ranking for that keyword, and it’ll show you all the questions that people are asking, and it’ll more importantly show you the similar topics to this.
(16:48):
So you can essentially create a bunch of different topics from this just one single vegan birthday cake idea. There’s a lot of free tools that do this as well. I think, and correct me if I’m wrong, is it Answer the Public or? Yep, Answer the Public is another one. A lot of people like that one because they do a visual component. So there’s a lot of tools out there to come up with new topic ideas, but I would say after you’ve done your keyword research, go straight into looking at what your competitor’s doing and what’s working for them.
Melissa (17:16):
Right, and I think that SEMRush tool is the mind map, right, as part of their content resources and their guru plan.
Ashley (17:23):
Exactly.
Melissa (17:24):
Yeah. We’ve had questions with people because they’ve got an entry level signup plan, and then the guru plan is the one that actually has the content marketing tools. Next question, Casey, what are some of your favorite methods for creating new topics?
Casey (17:40):
Well, I think I have a couple here that I’m going to talk about. The first one, we’ve talked about several times in the past, and it’s making sure that you use the Keywords Everywhere browser extension. I’m going to go ahead and paste that over here specifically right here, Keywords Everywhere. And that browser extension is exceptional because it provides a lot of data detailed to individual keywords, including related keywords. PAA, People Also Ask queries, long tail keywords. All of that information is included as an overlay on the Google results page on desktop when you enable the browser extension. So you can go in and type in, for example, something like banana cream pie and it’ll tell you all the People Also Ask queries associated with that. It’ll also tell you all the various long tail keyword phrases for a banana cream pie that are broken down by average search volume.
(18:33):
And these are, again, all presented on the right side of the screen for ease of use. So I’m a big believer in using that for new topic ideation and to look at possibly increasing the bottom line topical relevancy of existing recipes by making sure that we’re covering all known entities that might be related to that keyword phrase. Now the second thing that we’re going to talk about, and it’s very similar to the entities situation I just mentioned, is a concept called topic graphing. Topic graphing. I want you to take a look at a tool called topically.io. Now you guys are going to love it. It’s very easy to use. And Topically is a new tool that allows you just to put in any keyword phrase and it’s going to spit out all the related topical entities for that keyword based upon what Google is serving.
(19:26):
So it’s pretty relevant in that it allows you to dissect knowledge graphs and related entity configurations and the like. So for example, using my good old-fashioned banana cream pie, it’s telling me that, and some examples if I was to type that in, one of the related examples that it gives me is old-fashioned banana cream pie, which is one of the more easier examples to reign for based upon competition. And it tells me that if I wanted to try to rank for old-fashioned banana cream pie, my recipe specifically should be thinking about very specific concepts, recipe obviously, but things like graham cracker, banana pudding, meringue, crust, slice, convinced milk, pudding, pie crust, those are individual entities actually on the knowledge graph given values that I would want to make sure that my content, my recipe post and my recipe card specifically make mention of.
(20:19):
And the point is that if I was looking to compete for this keyword, I would want to make sure that my source was complete and touching all these individual topical entities were possible. And that is one of the things that a lot of bloggers do not understand and do not take advantage of. So this is a very simple tool that will allow you to put your existing recipes in and see here’s what Google is saying topically wise, entities would be a good match for this. And then you could go and look at your existing recipe and ask yourself, “Am I missing any of these entities? Is there anything we can expand upon?” And then focus on the bottom line quality in that regard.
Melissa (21:00):
Everyone’s really excited about those tools you mentioned.
Casey (21:03):
Good. Good to know. Good to know.
Melissa (21:04):
Very cool. Andrew, this question’s for you. What if your regular readers want content from you that isn’t highly searched? So how do you balance creating content for social media engagement versus creating content that’s going to help improve your traffic and rankings?
Andrew (21:20):
Well, I think that’s great. I think if you can start getting some traffic from something other than search directly, that’s awesome. So lean into that. Not every post has to rank great, but every post should be valuable for your readers. So you can use some posts to rank well and be competitive on keywords and to get new people in, but you should absolutely be putting effort into pleasing your existing readers and building that audience and building your email list and building your social media presence on whichever platform works best for you. Because it’s not really about getting search, it’s about getting traffic. So if you can provide content that your existing readers want, they’re going to view your site and you’re going to have more traffic. So I think the balance is going to be a little bit different depending on your site and your audience. I don’t think we can say spend 42% of your time on this, but you can feel that out and maybe over time, you start leaning into the building for your audience and writing more for your audience, which will actually make your traffic more stable long term too. Everybody relying 100% on Google is pretty precarious. So the more you can spread that footprint around, the better. Especially now that Pinterest doesn’t drive as much traffic as it used to.
Melissa (22:30):
Yeah. Ashley, two part question, but at top high content, you work on a lot of content audit for bloggers. Do you use the info from the audits to come up with new topic ideas? And if so, what data do you use to say, I don’t know, “Hey, this would clearly be a great topic and for my blog in particular to write about”?
Ashley (22:55):
Yeah, definitely. Casey and Arsen touched on this a little bit earlier, but when we run through and do those content audits, that’s exactly when we’re able to see what’s working and what hasn’t been working. But even more, we’re able to see what has the potential to work. So these are the pieces of content that are ranking on page three to 10 to where you’re almost there, but you’re just not there yet. And then when you look at the competition and see who is on page one, it’s not necessarily the Allrecipes or the huge brands, but you actually have the opportunity to get on page one. And so those are the times where we can diagnose and see, “Okay, these are the topics where you should definitely update, which Casey and Arsen touched on earlier. But then we’re also able to see what’s working really well, especially in the past year.
(23:42):
And when you’re looking at your data, especially we just got done with the holiday period, what worked really well for you in the 2021 holiday and the 2022 holiday? And then this helps dictate what you should do in the 2023 holidays, the topics that worked really well, you can figure out how to create subcategories on those and dive a little bit deeper. If you’re like, “Well, I already wrote 3000 words on this vegan birthday cake recipe, how the heck could I possibly dive even deeper on that?” Look at the variations. And that’s when you can go into Google and say, you can type in the vegan birthday cake. And Google will say, well, the people also asked, have you answered all of those questions that Google’s populating and the people also asked, or maybe that might spark another idea of a second version of this type of recipe that you can publish or this specific type of content.
(24:32):
So as far as the data that you’re looking at from content audits to come up with new topic ideas, you’re really looking at what worked really well. The things that didn’t work well, the topics that just not even on social media, they didn’t even get engagement on social media, why keep trying to go down that road and make those topics work? Stop trying to put resources on that. Instead pivot, update the content that has the opportunity to and create more content in terms of what’s already been ranking and what’s going to continue to rank for you.
Melissa (25:03):
Excellent. And that was a lot of useful information. That’s always a popular question with people we speak to on consultation, so thank you. Arsen, with all this data in mind, are Google Search Console or Google Trends good tools to use to come up with new topics, or is there any other tools that we haven’t already mentioned that you recommend using?
Arsen (25:26):
So Google Search Console, GSC is good. Casey most likely will chime in once I’m done. He has a cool process to figure out which post should be written or how to extract topics, but I like to just look at which queries are getting impressions but no traffic. And then just extrapolating from that. Google Trends for topic discovery is really good. We had a whole webinar on that I think last year. We’ll post a link. Ashley mentioned Answer the Public, that’s a good tool. Very similar to what Casey told us I think was topic something, TopicIO. SEMRush, Content Marketing Tool-
Casey (26:11):
Importantly, I would just interject because I know we’ve said Answer the Public a couple times, did you guys know that Answer the Public was bought by Neil Patel?
Arsen (26:18):
Right, right, right.
Casey (26:19):
So it will absolutely completely fall apart within the next six months and there will be a paywall, so it’s very important that all of you understand.
Arsen (26:27):
I think it’s paid now. I think it’s fully paid now. Yeah.
Casey (26:29):
So that means the quality will go down considerably as was the case with Ubersuggest, so I would not rely on that long term. I believe there is an option for that. I’ll let Arsen finish and I’ll see if I can track that other one.
Arsen (26:43):
If you have SEMRush and you’re spending the crazy money that they charge for to give you access, the content marketing tool set inside of SEMRush is pretty good. We use it all the time. We use it for my coaching clients, we use it internally. Gives you a lot of information. Also very similar to that topic tool that Casey put out. You can get a lot of that. Overall, and Ashley mentioned this also. I love running gap analysis between my clients and they’re very close, super close competitors. You can filter the data to see where there are opportunities for content that you have either already published.
(27:22):
So where both of you are ranking and your competitor is outperforming you, or opportunities where your competitors are in the top five or 10 positions or top traffic driving positions and you are not, then you can easily dissect what they are doing so that you can reverse engineer that and you don’t have to reinvent the wheel. Keyword gap analysis is your best friend if you don’t want to spend crazy amounts of hours trying to figure out is this a good topic, is it going to work, can I rank there? Look at what your competitors are doing. If they’re doing a good job, they’re ranking there. You look at the result page and you see, “Hey, I have a really good opportunity to be here,” invest the time.
Melissa (28:04):
And Arsen, you always suggest people compiling it all into very fancy Excel spreadsheets.
Arsen (28:09):
Right. Always create a spreadsheet for yourself with everything that you’re observing on page one of Google to make it easier for yourself to analyze all that data.
Melissa (28:18):
100%. So once we figured out the topics, Ashley, what do you figure out to put inside the post? What information to include?
Ashley (28:31):
Yeah, it’s honestly really similar to the concepts and the steps of coming up with new topic ideas. So you want to go and see what’s ranking on page one, and again, take out the huge brands that are ranking because they’re just domain cloud if you may. And look at the other two to three that are on page one that are fellow bloggers like yourself and see, okay, how long is their article? What information do they include? Is it a bunch of images and then one video, do they have three videos? I’ve seen content that literally has four different types of videos in it and also 2,000 words in it and has the recipe card. And so start to look to see what patterns you can see with what’s currently ranking and look at the information that they’re providing. Whatever content, whatever info that they’re providing is clearly matching the user intent and satisfying users.
(29:23):
That’s why they’re ranking. So as you’re coming up with your new topics and you’re trying to figure out, “Okay, I’m going to do this vegan birthday cake recipe, but how much info should I include?” Go to page one, see how much info they’re including, check out what their headings are, see what kind of questions they’re addressing, if they’re doing alternatives and what kind of alternatives they’re including. And this will really give you a guideline of what you need to, at the very least, make sure to put into your content. Of course, we’re not copying and pasting here, you’re making your unique content, but that just provides you with a guideline of, all right, at the bare minimum I know I need to include at least this much information.
(30:02):
You can also go inside on SEMRush’s topic research tool. Inside those cards where it shows you the top 10 articles that are ranking, right next to it, it shows you the who, what, when, where, why, how questions that people are actually asking. They are searching inside of Google and they want to know this information. Go use those questions as headings, make sure you are addressing those questions, the ones that make the most sense and actually relate to what the information that you can provide. Don’t go and put on a science hat if you’re a baker, but the information that you can confidently provide, add that into your article and use those questions as direct headings.
Melissa (30:41):
Very cool. I think that there’s also a tool that we use, Arsen, SEO Minion for these outlines to take a look?
Arsen (30:48):
Right? SEO Minion, it’s free Chrome extension, it’ll give you document outlines all the headings and everything else in there.
Melissa (30:59):
And even a word count for some post I’m sure would be useful, right?
Arsen (31:03):
Yep.
Melissa (31:03):
Very cool. Okay, Andrew, roundup posts. Even these confuse me. Can you please explain what roundup posts are and how to organize your topics using roundup posts?
Andrew (31:18):
So a roundup post is just a post that links to lots of other posts. If it’s done right though, it’s actually using the item list schema to help Google understand that. I want to drop in the recipe schema docs talks about the item list also. And then I’m guessing most of you are using WP Recipe maker, so there’s a good article on that. Even if you don’t use it, the article helps you understand what roundups are. So partly, you may be able to get a whole carousel of all of the links that are in that roundup. The more competition that’s there, the harder that is to get, of course. But you can also use it then to organize your content. So let’s say your Thanksgiving’s coming up and you’ve got 10 different recipes that are really great Thanksgiving side dishes. You can do a roundup of Thanksgiving side dishes.
(32:11):
So that’s providing value to your readers where you’re like, “Hey, check these out. If you’re looking for recipes,” and that gets you lots of internal linking, you can also link to other sites. It doesn’t have to be all internal links. I think the most important thing to think about, and I know I’m a broken record on this, but add value to your readers, give value to your readers. So make the roundup useful. We were talking right before we hit record about people who do roundups who have 50 recipes on it. You don’t want to do that because scrolling through 50 recipes on one post is really hard to do, especially on a mobile phone. There isn’t enough real estate. If you do want to have a lot of different items in the roundup, you can break it up into subsections and use a table of contents plugin. So if you’re sticking with a Thanksgiving theme, you could do Thanksgiving appetizers, Thanksgiving side dishes, Thanksgiving desserts. You could put all those in one post and have three different sections on your page. But you still have to organize your content properly in terms of your whole site and use your categories properly. We’ve done a bunch of episodes on that, so make sure your information architecture is still dialed in. So I think about this as supplementing your content organization.
Casey (33:23):
Just a couple quick points on the roundups. We still get questions all the time about adding the recipe card to roundups and Google has only come out once and it’s been about a year and a half now where they said that they were looking at closely going in and penalizing it because it was clearly a violation of the recipe guidelines and nothing has happened. Now Arsen and I both are going to be a PubCon Austin here in a couple weeks. I’ve already called ahead, so we’re going to be talking to a couple of the Google people in person about this to see what they’re actually doing about it because everyone’s really annoyed about it. Exactly. But in this case, it’s one of those things where if Google’s not penalizing for it, what can we say? I know it pisses a lot of people off to say, “Oh my God, I’ve got this fantastic 2020 recipe roundup for Valentine’s Day and I’m not in the carousel because it’s not a recipe, it’s a roundup.”
(34:13):
And the people that are putting a recipe card in the carousel, they’re showing up for Valentine’s Day recipes, which is crazy. And I would say that at this point, if you can just run the risk of doing that, of adding the recipe card and see what you see, I just understand that at any time, Google could decide to crack down on it. And I don’t expect it’ll be anything dramatic, you’ll just find that those posts will just fall out of the carousel, and then you’ll probably wonder, “What the heck why’d I fall out of the carousel?” And then you’ll realize, “Well, it shouldn’t have been in the carousel in the first place and it go on.”
(34:41):
Also, the thing about recipe roundups specifically is people just get really annoyed when there’s more than 30 items. Whenever we’ve done usertesting.com studies, people are like, “I don’t understand why there’s 98 items in this. I can’t get through them all.” So please understand that if you haven’t done a blog survey or anything or surveyed your audience, more is not more, more is not better, more is just more. So maybe you come and look at a quality issue, just because you see that there is a 47 item Thanksgiving roundup or a 47 item Valentine’s Day dinner ranking does not mean that you need to publish an 87 item roundup because you’re going to get that spot. That’s not how it works. More is not better, more is just more. So just be aware of that.
Andrew (35:25):
Hey, there’s a question that came in from Amy Katz that might be good to answer right now, and I’m not sure the answer. Regarding the WPRM roundup schema, if Amy includes other bloggers recipes as well as hers, can it still show in a carousel with the recipe side by side?
Casey (35:41):
Yeah, but it’ll just be your recipes. In other words, you can do a roundup of all external recipes, but there is no item list schema that’s going to be generated on that. So if you’re going to be doing a roundup and you want carousel consideration, understand that it’s only going to work for year own recipes, not for external recipes. Good question.
Andrew (36:03):
Thanks.
Melissa (36:05):
Okay, so I think we touched on this a little bit earlier, Arsen, but how do you make sure that you’re not creating new content that outranks your current content?
Arsen (36:14):
Well, just don’t do it. Don’t write about the same thing. If you already have a potato soup recipe, don’t create another potato soup recipe. And also keep in mind, and we started seeing this year, and I’ve talked about it before, I had it as a part of my presentation for the virtual Tastemaker conference, and I’ll post a link. I actually put it up publicly for those of you that didn’t attend it, I put it up publicly on our website, a link to it. But potato soup recipe as a result right now is going to show you only potato soup recipes. It’s not going to show you bacon potato soup. It’s not going to show you baked potato soup. It’s not going to show you any of those. Same applies for a lot of other keywords. Google is very specific. Google understands if I search for potato soup recipe, that’s what I’m looking for.
(37:06):
Whereas before, we had page one show a pretty decent diversity of potato soup recipes. A good example is cooking.nytimes, they had a baked potato soup recipe, which was I think number three, number two for a very long time. When I created the presentation for the Tastemaker conference, they were on page two. Today, they’re on page three or four to the point where Casey, they’re paying for ads for that result. So if you are going to create content and you want to start creating more content to create that topical depth, filling that topical silo, creating that category, again, Google is your best friend. Go in there, put in the query, see what’s ranking on page one, see how Google’s treating that. So if you want to create bacon potato soup recipe, put that into Google and see what’s ranking. If it’s all bacon potato soup recipe, then most likely it will never compete. I’m not going to say never. Most likely it will not compete with your potato soup recipe, or your baked potato soup recipe was not going to compete with your potato soup recipe.
Casey (38:10):
And there’s different carousels for each one of those.
Arsen (38:13):
Right, right.
Casey (38:14):
So I guess you’d mentioned earlier, let’s clarify that a little bit. Maybe not have five potato soup recipes, but let’s be honest, you could have five potato soup recipes if they’re different potato soup recipes and they would all necessarily trigger new carousels. I think that’s where it-
Arsen (38:31):
But you have to clarify, you don’t want to have pretty much the same recipe written in different ways.
Casey (38:37):
Exactly, and that’s where I think a lot of people are confused. I’ve got this big potato recipe that I can make in a slow cooker, so I’m going to take the same recipe and I’m going to go ahead and use it in the Instant Pot. Okay. And you’re thinking just because it says Instant Pot, this is a completely new recipe and we’re going to do well. Google is much more, I guess, more discretionary in that regard than they were even two years ago. I mean, I see a lot of bloggers trying to get a recipe post where they’re putting three, four, or five methods in the post to try to cover everything and it makes for a very muddy recipe, in fact, in many aspects. So I would urge you, if you think that you can have competent recipes and have them individually and there’s enough keyword detail to justify having the two recipes, then consider doing that. But we don’t necessarily need five baked potato recipes just because three of them are, because you made them in different methods. We find that that just doesn’t perform well long term. And there’s a definitely internal cannibalizing effect with regards to that related overlapping content.
Melissa (39:42):
Right. Okay. Just a reminder to everybody to put questions in the Q&A and thumbs up any questions that you’d like to see bumped up to the top, prioritized, and we’ll get to those shortly after our next question. Casey, I’m going to direct this to you. This is a pretty popular question. We received quite a few requests for this answer, but it’s regarding AI and content writing. How does everyone feel about using ChatGPT for upcoming new topics and just writing content in general?
Casey (40:13):
So I believe that we’re going to try to put together a whole presentation on this because there’s so much that we can talk about. So I’m going to just nerd off for just a couple minutes because I do have a background in large language models and transformers that hopefully I can dumb this down enough, so that unlike my wife who says on the couch and just looks at me with a blank stare on her face, you get something out of this at this point. But when we talk about ChatGPT, basically GPT stands for generative pre-trained transformer. So it’s just a fancy way of saying that this is also known as a large language model. It’s a chat bot. It’s just a chat bot on steroids. And how it works is it’s a gigantic neural network that’s trained on known information around the web.
(40:58):
So right now, with the most recent version of number three, it’s trained on about 175 billion parameters. That’s the weights that it uses. And this was invented about five years ago, and it’s been trained on a large fraction of all the text on the open internet. So everything that you can possibly think of, it’s been there, but it hasn’t been updated in a while. So just to understand that a lot of this is outdated. Matter of fact, anything that’s happened in the last 18 months, you’re not going to really be able to have a competent generated answer from ChatGPT. Now, the training specifically consists, how do you make something like this? And basically how it does is it consists of playing the following game over and over again, trillions of times. Predict which word comes next in this text string.
(41:42):
That’s what they’ve done. They’ve trained this chat bot to do that over millions and millions of times. And this is great for writing essays, it’s great for predicting answers to questions and solving problems because the large language model is basically saying, “Oh, well I know what should come back. I should know what should possibly come after this prompt, the answer to this question or the essay itself.” Then it will proceed to generate the thing you want word by word by word. That’s different than artificial intelligence. And I think where a lot of bloggers are confused is that this isn’t artificial intelligence. Again, this is a string intelligence here, and basically what it does is if you can put in a prompt, and I apologize, I was going to do a live screen sharing here, but it’s funny because ChatGPT has been down at various times, and it happens to be down right now. Good times.
(42:34):
But you can put a prompt like, “Generate me 10 romantic Valentine’s Day recipes,” and it will generate it. And you could take that information and you could make a roundup like we were talking about, and then you could link to those similar recipes on your site. But this is also a good way to do, remember Arsen was talking about content gap, I believe maybe it was Ashley, content gap analysis. This is your way of, if you’re going and you’re using something like ChatGPT and you’re saying, “Okay, I want you to generate 10 romantic Valentine’s Day recipes for couples.” You take the 10 that it generates.” It’s taking that information from all over the web, which means, hey, those 10 ideas that it’s generating must be pretty popular. That would be a strong signal to you to ask yourself, “Okay, I don’t have all those 10 recipes on my site. Maybe that’s a content gap that I can fill either now since I have about three weeks before the holiday, or start thinking about next year.”
(43:26):
But the thing to really understand about the ChatGPT is that it’s just parroting the training data it has. It’s not able to do original thought. And that’s why we call it a highly trained chatbot because that’s all it is. And I’m going to go ahead and paste over a great article on exactly how this works because it’s going to be really interesting reading for all of you on the call. This is a great overview of Chat right here.
Melissa (44:05):
I took some time to play with it over the holiday break, and I have to say it’s exciting. I feel like a lot of people are scared of it, but I had a lot of fun and it was a great way to get familiar with it for those of us who it’s new for.
Casey (44:20):
And that’s great. And that’s great, except when it’s not. And so what I’ve done is I’ve also pasted over an article where it tells you about all the incredibly bad examples of the information that has been used by ChatGPT. As a matter of fact, I don’t know if it would wordpress.org, it was a big forum, they banned ChatGPT in December because the automated responses for the product report were so bad that they couldn’t allow it on the forum. And so they had to ban the whole bot from it. And there’s some great ways for those of you on the call to use ChatGPT and I do recommend that you’d use it, but just understand that it’s never going to replace you writing a whole recipe post. It’s never going to replace that personal touch that you provide in understanding what users are looking for and surveying your audience. But it will help you expand some content.
(45:17):
I’m going to go ahead and give you access to a Chrome extension, and the Chrome extension will actually allow you to use ChatGPT on your browser pretty easily. I’m going to go ahead and paste that over as well. Thank you. But the thing about any new technology is it’s all nice and shiny for a while until it’s not. We’ve already been hearing all these things about, “Oh my God, college students are going to use this all over the United States to cheat.” And the thing is that there are threads, and we’ve been making a folder of them on Facebook, of all these professors saying that’s exactly what they’re doing is they’re using this ChatGPT to write whole essays. But the problem is that it will keep outputting the same information.
(46:00):
So you’ve got 20 people who are told to write a essay on the illustrious origins of revolutionary political thought, and you’re going to find that they all shame share the same paragraphs because the students aren’t smart enough to rewrite the paragraphs. And so the professor’s able to go in and say, I can tell pretty quickly which of these essays have been written by the chat bot and which have not. And even you taking some of the answers and rewriting them is not enough. And so I have been asked by Search Engine Land to write an article on the various ways that you could detect this chat related content, and we’ll see how that goes. I’ll hopefully have that out by April. But what’s really going to be interesting is when they launch, right now it’s 3.0, and they’re looking at launching 4.0. And 4.0 is going to go from, I think I mentioned here that we’re at 175 billion parameters.
(46:58):
The new version will be 100 trillion. So it’s going to be interesting to see if that incredible increase in corpus in the amount of documents that it’s going to be using in tangential connections is going to result in some huge uptick. A lot of people don’t think it will, for example, it’s not at the point where, “Hey, can you write me…” And even Andrew has played around with this where it will write simple apps. It is able to write code for specific apps, but the more complicated the resource, the less chance that it’s actually going to work because it only takes one small mistake in the syntax of the code for it to break the entire thing. So we’ll see what we see, but it certainly is illuminating. And for everyone on the call, this is something that you should be aware of and start to educate yourself on as we go into 2023.
Melissa (47:49):
Yep, definitely. We’ll get a whole episode on this, I hope, soon, so that we can cover what this new update’s going to look like. When did you say that they’ll have the update for this ChatGPT?
Casey (48:00):
Well, they’re saying 2023, so we’ll see. But I’d be surprised if it was that soon. That’s from the founder, so who knows.
Melissa (48:08):
All right, let’s get to some Q&As. We’ve got, it looks like, 10 minutes left, so I’m going to start with what’s been uploaded the most from an anonymous. If I am in the top three for a small keyword, is it not worth updating to try to go after bigger keywords?
Arsen (48:27):
It depends.
Melissa (48:27):
It depends, oh.
Arsen (48:29):
It depends. I mean, look, you can definitely take a look, and again, this is one of those things where using tools like SEMRush, like Keysearch, Keysearch does this well. I’ve done this in across both tools. You can grab who’s ranking, who else is ranking for that keyword, plug them into that tool, see what other keywords they’re ranking and continue to click around until you get to somebody who is ranking for the same keyword, but with a piece of content that’s optimized for other keywords that are also bringing in traffic on that same topic or very close topic. And you can then refocus your post to rank for those other keywords.
(49:22):
Again, it’s not always going to happen. It’s a bit of effort that you have to put in there, but it’s one of the easiest ways to do this is to look at what your competitors are doing and then see, “Okay, I’m in the top three for,” I don’t know, let’s do an easy one. I use this one a lot. Oven baked salmon with garlic and lime. But then there’s another post that’s ranking for that keyword, but it’s also ranking for oven baked salmon. So you can then see, can I refocus or can I reoptimize or take focus away from garlic and lime and just turn it into a post that can potentially rank for oven baked salmon.
Melissa (50:03):
Nice. Thank you. We’ve got another question from Kaylee. Can you elaborate on the process you use after you see queries you are getting impressions on but not clicks? How would you improve on that?
Arsen (50:16):
Casey, all the way.
Melissa (50:16):
Yes.
Casey (50:16):
I’m pacing over a-
Arsen (50:24):
I know Casey had this one locked and loaded.
Melissa (50:26):
Right, okay.
Casey (50:29):
Oh hey, so there you go. I pasted something from ChatGPT. No, no. What was the question again?
Melissa (50:35):
Can you elaborate on the process you use after you see queries you are getting impressions on, but not clicks? How would you improve on that?
Casey (50:43):
Well, it certainly depends on the number of impressions you’re getting and not clicks. Usually if you’re getting a lot of impressions for something, that means that you’re close to the quality level, but you’re not necessarily there. You’re probably far enough back that you’re maybe in the top five, top seven pages, but the intent is so low, or it’s so competitive that you’re just not making inroads. I think that best thing that you can do with content like that is to first ask yourself, are you going to be able to compete on that? If I look on the first two pages and all I’m seeing are sites that are five to 10 times my size, it doesn’t really matter if you have the best beef broccoli recipe in the world, it may be very hard for you to compete on an organic level.
(51:25):
Now the carousels completely different. You could absolutely pop in absolutely in the top 20 in the carousel if you have a fully enhanced recipe card, if you have your nutrition filled out, if you make sure that everything is filled out there specifically. It’s amazing how many people don’t fill out all the data. You don’t have to worry about the guided recipes nonsense, that hasn’t changed in the last three years. But if for a lot of people are like, “I am against nutritional information,” I’m going to tell you on the call, if you’re against nutritional information, you’re probably costing yourself about 15 to 20% potential traffic, because people tend to want that information and it’s a good rich snippet. So I always want to fill that out and make sure that I pop into the carousel with as much information as possible. I would start there.
(52:07):
I’d also make sure that I’m looking at my recipe and I’m looking at that keyword, making sure that the intent matches there. Now again, if you’re in the recipe and lifestyle niche, most of the intent is informational. We want to make sure that there’s no transactional elements that might be diluting the reach on your page. It’s why we don’t recommend shop pages unless you run your own products, because they’re a waste of time for the recipe and lifestyle blog. But those are things that we would look at in trying to get a post or a query that if I’ve got a query that’s giving me 1500 impressions a month, but not one click, I know that there’s some a mismatch on the page. I’m either having a quality issue where the post is just not high enough quality to get over the threshold for it to move forward, or my keyword phrase is not necessarily at the right level to fulfill the intent of that search term itself.
Melissa (52:58):
Okay. Thank you. Another question from Tammy. Hi, Tammy. When you use keywords organically in a post, long tail or not, is there a number of times it needs to appear so that you can possibly rank for it? Just throwing a keyword in a sentence doesn’t seem to make a difference.
Arsen (53:22):
So keyword density is not a thing that you should be worried about it. It’s all about how well you’re covering the topic. So I wouldn’t necessarily focus on, “Oh my God, haven’t used this keyword enough times in this post.” Obviously, if you’re structuring your content properly, you’re prioritizing intent and you’re covering the topic in its entirety or close to it, and you’re answering all the relevant questions and you’re satisfying that primary secondary intent, you’re going to have enough words in there to make sense. Obviously, look, your meta title, your H1, your H2s, those places should have some presence of keywords, whether they’re exact match or partial match, don’t overoptimize, but Google is now smart enough to understand what the post is, why that post is there, and what it’s about without you having to repeat or emphasize specific keywords over and over and over again.
Casey (54:19):
Yeah, I think if you have a plugin like SEO Quake, which I do recommend, I think it’s a fantastic browser add-on, it provides an ability for you to very quickly dial in and see keyword density on the page. And the reason that keyword density is still used is just a very physical, very easy way for you to compare your post to what’s ranking in Google. So if I go and I’m looking through Google and I’m like, “Okay, I’m seeing that the top five recipes here, I’ll have a keyword density of about 1.7 or 1.8.” There’s a correlation causation. It’s just one piece of data that you can use. So I say, “Okay, well that’s interesting. So now I’m going to look at my content.” And I look at my content and I see that I have a 3.5% keyword density for broccoli, in broccoli and chicken as an example.
(55:08):
Keyword density isn’t something you should rely on, but that’s probably a good warning sign that you probably overdid it in the overoptimization of this content. So maybe I would use that as just a very quick way to dial that down a little bit. Okay. Yeah, there are plenty of recipes and sites that rank with incredibly keyword stuffed pages, and there are plenty of recipes and sites that rank with literally not even a keyword on it. So that’s why we can’t lose too much sleep over a specific keyword density recommendation. In my audits, I do tend to say, “Okay, hey, if this is something that will make you feel better, try to keep it under 2%,” and I show them how to check this and compare their content to everyone else, because I tend to find it’s better to underoptimize than overoptimize in most situations, and I’d rather them always err on the side of caution, if anything.
Melissa (55:59):
Okay. I know we covered this a little bit earlier, but it had the most upvotes, so I just want to throw it out there. How do you find your close competitors? If there’s an easier way to explain it or if there’s a best tool? What’s everyone’s consensus on this?
Arsen (56:16):
So I don’t know about Keysearch. I know that SEMRush does keywords in common. Well, they will tell you who your competitors are based on keywords and common assessment or analysis.
Casey (56:25):
Correct. Yeah.
Arsen (56:26):
I don’t know. Does Keysearch do this, Casey?
Casey (56:29):
It does, yeah. But again, I think the SEMRush, and I’m not a fan of a lot of what SEMRush offers, but their competitive comparison’s pretty good, because again, it takes a bottom line. Here’s two sites, and they’re trying to rank for 17,000 identical keywords. That’s a pretty close competitor, so that’s a good way to understand that. And I have bloggers who onboard for, not all the time, they don’t know who their competitors are. Well, if you’re a vegan site, then clearly other vegan recipe sites are your competitors. If you are a gluten-free site, then clearly other gluten-free bloggers are your competitor, and you can either do a search in Google for some of your top keywords and see who Google is returning in the carousel. Or you can use a tool like Ahrefs, or you can use a tool like SEMRush, or you can use a tool like Majestic and see what’s going on there. That’s-
Arsen (57:18):
Yeah. But if-
Casey (57:18):
… pretty common sense.
Arsen (57:19):
… like Casey you said, if you’re performing searches and if you’re looking at what keywords you’re ranking for and you’re looking at who else is ranking next to you, I’m sure over time, you already know you figured out who-
Casey (57:29):
Yeah, those are your competitors.
Arsen (57:30):
Right, who’s chasing you around the web kind of a thing.
Casey (57:32):
Exactly.
Melissa (57:35):
All right. Well, we’ve wound down, we’ve got about a minute left. How did I do everyone?
Casey (57:42):
Good.
Arsen (57:43):
Yay.
Casey (57:46):
Very nice.
Melissa (57:46):
Oh, goodness gracious.
Andrew (57:47):
Way better than Ashley.
Arsen (57:49):
Is she even there? What’s going on with her? She’s taking a nap.
Casey (57:52):
Ashley who?
Ashley (57:53):
I took a nap real quick.
Arsen (57:57):
What?
Melissa (58:00):
Thank you to everyone in the chat for the kind words. Really, I appreciate it. But to just round this out, as always, thank you for joining us. Just a reminder that we’ll be posting a recap on our site next week. Then I’ll have a replay of this episode and answers to questions that we didn’t have time to get to today. I hope you can all join us again in February. I will be hosting again. Yay. And everybody, just have a great week. Thank you.
Arsen (58:25):
Bye, everyone.
Casey (58:25):
Happy Winnie the Pooh Day everyone.
Melissa (58:27):
Happy Winnie the Pooh Day.
Andrew (58:27):
Take care. Bye.
Melissa (58:27):
Bye.
