WEBVTT
1
00:00:00.000 --> 00:00:32.000
Hello, I'm Jeremy Rivera, your unscripted SEO podcast host. I'm here with Anne Smarty of Smarty Marketing. If you haven't been around SEO for any amount of time, then you haven't seen her. But if you've been around, you know that she has a big footprint in our industry. Super excited for this interview. But for those unfortunate souls who aren't familiar with you and what you've been working on, give yourself an introduction, tell us about your industry niches, some of your experience, and share with the people who you are.
2
00:00:32.000 --> 00:01:20.000
Interesting question. I've been an SEO for more than 20 years and I've done pretty much everything in those 20 years from SEO journalistic reporting to SEO community management to SEO analysts to everything. So I've seen pretty much all sides of it. Currently, what we are doing SEO audits plus geo. I don't do those separately anymore because they overlap so much. And Reddit, we are very much focusing on Reddit. Not because we really chose that, but we've been on Reddit for more than 15 years and suddenly Reddit became a thing. So we started selling it as a separate service because we do have a global experience in how it works and what to do there.
3
00:01:20.000 --> 00:01:47.000
The Narwhal does Bacon at Midnight. Tell us a little bit more about that Reddit activity. Is it just find relevant threads and dive bombing in with links, or is it building your own subreddits and trying to gain audience? Like what's the, what are a couple of the plays that you're offering people when it comes to Reddit?
4
00:01:47.000 --> 00:03:14.000
We definitely focus on long term transparent branded existence on Reddit, which unfortunately takes more time than the first method that you mentioned. But it's long term and it's not. It's definitely a long work, a lot of it, a lot of effort. Our average is 6 months. When the community starts really picking up and that subreddit starts ranking in, Google starts showing up in LLM citations, but on average six months at least. But then once you have your subreddit, you definitely can branch out to different subreddits as well and build your presence there in a meaningful way. With new accounts. It's the best way to get suspended right away. And not even with new ones, with established ones. Once you try, once you start trying to do sneaky stuff, you can lose accounts. And that's exactly why we don't do that for clients because we, we operate their official presence. We have to be very careful with it. It's a lot of responsibility and anything Spammy you do on Reddit can backfire on the brand brand reputation and that's too much of responsibility to take. So we focus on a safe, transparent and meaningful way of building Reddit reputation and Reddit presence. Basically.
5
00:03:14.000 --> 00:03:26.000
Well, you stole a question from me because I was going to ask how do you crash out fast on Reddit? But I guess if you'd sum it up, what's the fastest way you've seen a project crash and burn?
6
00:03:26.000 --> 00:05:22.000
If you think about that, it's not very much different from SEO in general. There are quick ways to do stuff in SEO and achieve those quick results and there are long term and that all relies on how much you value your actual entity or brand. The there is a gateway to Reddit if you are ready to gain quick results and then flip to something else and lose everything. That's an approach too. Can happen as fast as one day to six months can. I saw very well established accounts, almost 10 years all being suspended on Reddit when they start trying to do something differently and quickly there. So regardless of how established your account on Reddit is switching to trying to be very promotional, linking to your website all the time is still risky. So it can happen quickly, it can happen in the long run, but it will happen once you start trying to game it. And it's not so much different from SEO if you remember our old days. And it's it's also a very popular model still to create a website, achieve traffic within 3 months by publishing AI generated high volumes of AI generated content, enjoy your 3 months of traffic and then get suspended and switch to another website doing pretty much the same. People make good money using that method and who am I to say if it's right or wrong because we work on behalf of clients, we just cannot afford that freedom and flexibility to game stuff, you know, and experiment with things. So we have to be more careful than anyone else
7
00:05:22.000 --> 00:05:54.000
does strike me that, you know, a lot of people think of their homepage as their home for their entity and business. But with geo, with AI overviews and LLM layers, it seems like a more distributed model of understanding your whole business footprint. Online seems to be a much more holistic and useful model in this moment of SEO. Would you agree?
8
00:05:54.000 --> 00:08:30.000
I definitely agree. I would still not neglect my own website for sure. I've seen a lot of changes in LLM answers once I went back and updated my very old personal website because they kept pulling information from it and it was like 22 years old and much touched for at least five years. So I would definitely do the homework. That means making sure your website is up to date. You have a very detailed about page, contact details, all the trust factors, everything that you can do on your website, definitely do it. Don't forget about that step. But it's definitely not enough. You have to have. I refer to that as LLM consensus. This is why LLMs are able to find a lot of different sources about you and they are very consistent in their messaging. Don't be like me. My footprint is a mess. Some of my bylines still talk about my projects that are no longer there, have not been there for 10 years. Many roles, many inconsistency and I sometimes I try to go and fix stuff but it's never ending. I have a lot of footprint accumulated for 20 years, but I'm also not as much concerned about LLM visibility because I have another ways to build my business. If you are very concerned about being visible in AI answers, fixing this whole digital footprint to be up to date and consistently saying the same thing about you, actually your value proposition and how you are different from your competitors is the very first step. It's LLM consensus. Audit can be done in house, takes lots of time, especially for older businesses as well. But you will see the difference right there and then identifying channels that make a difference in LLM training data or citations and work for your competitors where you're not present and gradually develop your presence there as well. Usually Those are Reddit, YouTube, Wikipedia and a few very high profile publications which are not as easy to handle, but it's doable. But it's definitely not just your website these days or even not just your rankings.
9
00:08:30.000 --> 00:10:39.000
Yeah, that came up with my interview with Grant Simmons. We were talking about, you know, Kalicube and thinking about it in terms of, you know, building an entity home on your about page. It's crazy how many people like I worked with four or five brands and they had no about page at all. Or their about page was like one or two paragraphs and it was just the most generic, fluffy stuff possible. I'm like, come on, there's. I know more. I haven't even worked with you yet, but I know more about your company than is spelled out on your about page. So there's definitely that. And then you're right, you gotta, you know, educate your customer support your exterior customer support representatives, Claude and chatgpt, they need good intel, you know, that's what my friend Matt Brooks at SEOT says, train them, you know, give them the materials like, and you're right there's for older entities and businesses. You do have that spread of old profiles, social profiles, that this snippet of information and it can be a headache trying to find credentials for, you know, a five year old, you know, blogger account and trying to get things updated. So I guess that's where, you know, we as consultants and can earn our keep of hunting those things down, impressing the value of that consistency. But I'm curious if there's anything, if there's a playbook or like a walkthrough. What are, what are the top five things aside from, you know, hiring you to go set up Reddit for them, that if a new business, you know, because I'm working with a new project, you know, they're doing AI energy power for data centers. That's like clean or whatever. But what would you suggest for a new project getting off the ground? How do you think about gaining that visibility? Where do you go? Where do you. Is it buying press releases? Is it popping up profiles? Is it classic directory submissions? What's your top five?
10
00:10:39.000 --> 00:15:11.000
Don't dismiss directory submissions that easily. These days they are being very well used by LLMs, which is like they made a very good comeback from, I don't know, 20 years ago. So my answer is very it depends basically on the business, what their budgets are. I would usually suggest do the fundamental homework which is establish your profiles, name consistency, get registered where it makes sense for many businesses that could be Yelp, Trustpilot, all of those places that are being popular in your niche. Basically, if it's B2B that be could we go to what, whatever, whatever is in your industry. Claim your profiles everywhere, create a very well organized system of how you contribute everywhere. It's very doable with repackaging, recycling your actual website content if one asset can be a good update for 10 profiles. So just create that system that works for you at the bare minimum, very consistent. Contributing is in many cases enough to start being visible, especially at the beginning. And then you can also branch out to newsletter sponsorships, paid advertisement ads on Reddit and elsewhere, wherever your audience is. But this organic work should be the first step because once you start growing, the people will start searching for you. And that's when all the problems start. Because brand being a brand is great, but no one talks about how it actually you lose control over it because stuff starts ranking for your name and that is the most important search query. That's your brand name variations, brand name reviews and all of that. And unless you do that homework before you start being very actively Outbound, you will see trouble very quickly. So the steps would be establish your organic footprint first, then find where your audience is and there are lots of ways to reach out to that audience. Being a podcast guest or starting your own podcast, live streams, whatever you can afford the time and team to do and let if you do that every week, even a little bit counts. But start doing something before investing in ads would be my I've seen a lot of, a lot of startups that for a month or two they were enjoying themselves and then people start using them and start discussing them on Reddit and now all the fun stops right there because no one wants to deal with them. Everyone is searching for brand names, including LLMs. Before they recommend a brand in an AI answer, they will do the branded search. They will find stuff about you that your customers may not want to know. So start doing something real and establish your branded queries very well. Control as many results there as you can. And I think in your newsletter somewhere I forgot which one I actually shared that trick. Do the bare minimum you can afford, just do it consistently. For example, I cannot afford a very well polished videos doing guest podcasts because that's a lot of coordination and my team is slammed with client work and I obviously cannot do all of that myself. But the bare minimum for me is my live stream is on my calendar. I make sure every Wednesday I join it and I talk about what happened in the industry, on how, what to do about that and that alone drives a lot of business just because I do that every week consistently. So everyone can decide depending on what they can afford. But consistency is the key.
11
00:15:11.000 --> 00:16:51.000
It's fascinating because I've been doing this for about 20 years too, and I am embarrassed to say that I recall very clearly a meeting where I said the words I don't care about your branded search in your Google search console. I'm filtering it out so that I can focus on unbranded keywords and phrases. And the paradigm has shifted so hard, you know, partly because of HCU, but also you're right, there's LLM factors and LLMs are a lot less forgiving for bad bad mouthing that it goes against your brand anywhere online. They're, they're programmed to surface. Oh, is there a trouble in the product? And you know, when I interviewed John Henshaw, he's like some of the the best thing you can do for SEO sometimes is turn inward and fix your broken ass product. You know, if you have something that's straight up janky or broken and it's gotten bad reviews. Whether it's your workflow and delivering results, if you're a SaaS, if there's a feature that's not doing what it's supposed to, people are going to talk about that. So, you know, you need to, you know, turn the eye inward. Are you finding as an SEO that you're now with these changes, you're able to get deeper into these organizations and you're not just held at bay and like, oh, can you SEO our website?
12
00:16:51.000 --> 00:19:45.000
Interesting. We used to be an island. Remember those times when we were absolutely not aware of what else is going on with our clients? As long as we have our keywords and can report some growth, that's everyone, what everyone cared about. Now I'm much more involved in reputation management. Organic, especially organic search. But Even otherwise, because LLMs search spontaneously, they can search Google or they can search Reddit itself for everything that you were trying to hide from Google search results. And you're right, I have to deal with so many departments, sales, customer support. You would not believe how much of reputation problem is caused by this communication between what the website says, what sales people promise, and that causes so much problem on Reddit and thus everywhere else from there that I have to have almost separate meetings to identify why those problems occur. Like, where is that disconnect? In many cases that, okay, we cannot say that publicly, but we can say that privately. And I mean it's no longer. You cannot hide anything. If there is an upfront price and you are not talking about that upfront, people will find that on Reddit. And in the quick AI answer right in the AI overview will tell you how much you, you charge upfront and you cannot hide that information anymore. So alignment inside companies becomes the key to LLM visibility and it's a fight. I spent too much time figuring out what we cannot say publicly why and what does it take to change that policy. Because it's already there public, and there is no hiding that. And in 50% of cases, I cannot win that battle because those executives are still like, no, that's how we make money. We cannot say that. And I mean like, okay, then this is, this is what we end up with. And then let's change the strategy instead. Instead of trying to fix it, we should be doing something else there. And usually that's building brand awareness in a different way. We have a lawyer client which charges more than everyone else and we're not allowed to talk about why, even though there is a clear reason. They take more difficult cases overall. But for some Reason they don't allow us to tell that. So we focus on hosting MaaS for them and just build their trust and authority that way without acknowledging the big elephant in the room as well. But we do what we can.
13
00:19:45.000 --> 00:21:43.000
I'm curious about how AI, you know, we, I, it's, it's sad that we only have the one word for it, but I, I think there is a true differentiation between the workflow and augmentation side, the practical utilization of Claude code or desktop between your team, multiple different models of AI to unlock capabilities. You know, I was talking with Brett Sklar just yesterday about how if you just take, take away one headache per week and just ask your team, use these AI tools and cure a headache you have this week. You know, be inventive and see what you come up with. That small progression instead of trying to replace workers, just eliminating tasks is just like SEO on steroids. There's so many workflows, so many creative things that it's way better than spending your SEO budget. You know, I know that I charged somebody $2,000 to edit Meta titles and descriptions because they demanded them they be bespoke and human written, every single one and they wanted to pay for the hours for it and they thought that was worth it. But I would absolutely not Recommend that now 10 years later it's not worth it because you can do so much more with $2,000 worth of budget and you can still get the titles and descriptions written really well, possibly even better, you know, if you really have honed your agents or your skill set to do so. So I'm curious what you're finding within your company, within your agency, when it comes to leveraging AI as a workforce element.
14
00:21:43.000 --> 00:24:41.000
Well, it's an interesting question. My team is mostly at this point they have mostly involved in two big tasks, which is one, community management and we use AI for tracking, analyzing, sentiment, all of that. But there is a lot of manual work there for sure, because it's Reddit. We write manually there, we don't automate it, the actual posting. So aside of analytics and monitoring, it's all manual work. The other part that they are doing is pr, digital PR assets. Likewise, we can use AI for ideation, competitive research and stuff like that, but it's not a good way to invent something actually original and unique. So we use for direction. But most ideation comes from brainstorming actual ideas by people. Building AI is handy for internal linking to that asset, to coming up with suggestions on how to reach out to journalistic research, things like that. But the core of that product is still manual work. I personally am not as much involved in those tasks, but I do entry audits and I mean, I use AI here and there. It's very connecting. Semrush or Ahrefs to quote, is a great way to kind of get insights that you didn't wouldn't have thought about. Competitive research. Great with AI as well, and analysis. But I also do a lot of manually there as well because I feel like maybe I'm old school, but unless I dig deep and see actual search results, actual queries, sort them, filter them, I don't feel like I get a good understanding of the health, SEM, positioning health of the website and where I should go. This is my brainstorming method before I create the roadmap of the strategy. I actually spend week, three days, depending on the size of the website, just looking at the data myself. Again, I can use help with Claude or Gemini for usually those two. I use things like analysis, brainstorming, things like that. But definitely a huge core of work in our company is still manual. Maybe because we picked up the services that are not easy to automate.
15
00:24:41.000 --> 00:27:58.000
It makes sense that, you know, we, that humans find the place where, you know, you're dipping into multiple different pools of data and have to have in mind, you know, a certain filter and parameter on it. Like if you, you know, I wrote about this on Open SEO. Like if you treat Google Search Console data as totally credible, then you're overlooking, you know, the dark queries, the fact that between 30 to 80% of the query table details, they're gone, you just can't see. And so you're basing your entire strategy of, you know, oh, we got a thousand clicks, but out of those thousand clicks, you got 200 of those that were told, you know, where they were, which pages for what keywords. That's less than a flip of a coin that, that's sticking your finger in the wind and saying what, what direction is the wind blowing? Kind of accuracy. That's it. But so. So it's about choosing and descri between these sources. For me, a big unlock has been honestly connecting MCP for Google Search Console, Google Analytics, Microsoft Clarity and keyword research data platforms like Open SEO or Semrush or Data for SEO and just asking for it to pull from the API and put the data next to each other for me to evaluate. You know, my favorite thing with Claude code is open this in a browser, show me all of the different types of data, slice it this way, and not ask it to analyze it, but just show me the data that's so hard to navigate to in GA4. If I have to open GA4 again, I'm going to hit my head on the desk till I pass out. But I can ask Claude, hey, show me GA4 data in a GA3 format. You can get it to do that and pull the same information and display it in friendly stacked sections where you know, oh, these are the pages. Sort by sessions. You can. You have that on your finger, fingertips. So it enables human analysis at a much quicker pace. If you can pull from. Pull via API and mcp and then you can apply your human knowledge of. Okay, GSE is cropped this way. Microsoft Clarity and GA are going to overlap, but they're going to report the same metrics differently. Bing Webmaster Tools, actually surprisingly good amount of data there to be able to pull. Microsoft Clarity added bot information. So GA4 reported there were 80 visits and 40 of them went into the direct bucket. That's what Microsoft Clarity told me. Those 40 were bots. And that's not necessarily bad anymore. It used to be like, keep the robots off my site. But now 50% of the traffic is bots. And I gotta say, some of those bots are running around with wallets. They can spend money.
16
00:27:58.000 --> 00:27:59.000
Yes.
17
00:27:59.000 --> 00:28:03.000
So don't. Don't treat them like strangers anymore.
18
00:28:03.000 --> 00:29:18.000
Yep. It's interesting the, the time, right. The switch that we had. Well, totally with you, honestly, because I don't deal with keywords as much as I ever did. I'm not into those actual search queries that much and not by choice. Just happened gradually for me because somehow I ended up more on the branded query reputation management side of things. And with Reddit and jio, whatever you call it, analysis, that just happened very, very easily. Most people, mostly that I get, don't ask about Google rankings that much anymore. They come to me with, we want to be on Reddit and we want to be an LLM. And search queries especially digging deeper into those isn't that much of my routine anymore. I actually miss those times now that I heard you speaking about that. I'm like, oh my God, I haven't done that for four months.
19
00:29:18.000 --> 00:31:52.000
It's a changing game. It certainly is a different ecosystem, largely because, you know, there was that. It's because Google had a true monopolistic moment and squandered it. You know, they had the golden moment in the sun where being not a real competitor, no one could come close. Yahoo was a joke and they just got so high on their own supply. And we got high too, you know, because we could, like you said, we could stand on our little island separate. And you know, I had one client, honestly, I never spoke to, I was part of an agency. I did SEO for them. Never met anybody on their team, never was in a single meeting with them. There was a salesperson who talked to the customer support person who talked to me to do, quote, unquote, do their SEO. And that level of separation that, that doesn't exist anymore because Google's not the solo primary channel. And even they've. They did it to themselves. They, they very quickly rushed out AI overviews when they saw the threat of LLMs on the horizon and kind of spiked their own wealth. And now that I guess they're enjoying the, the dark box, the, the bringing back of the dark ages. But when there isn't clarity for these businesses of how much value that Google is driving to them, and there's all these LLMs that have messaging that's ascendant, that are reaching C suite through LinkedIn, talking about, I don't doom and gloom about anymore. There was a moment where I genuinely was really concerned about the industry. But I think there's a level of acceptance now of, hey, this is what it looks like when a monopoly loses its monopoly and we get true competition back onto the table. And Google's going to have to make its choices and we're going to have to make our choices as to what surfaces or places we try to optimize. But it's as I'm writing a book right now of Search Everywhere Optimization. The acronym doesn't need to change. It's not geo, it's SEO, but it's Search Everywhere Optimization, not Search Engine optimization.
20
00:31:52.000 --> 00:33:26.000
Yep, and I totally agree with you. It's very telling. I haven't had a lead in months that would come to me saying I want to be on Google businesses, just not just probably. There is a lot of, of course noise telling them that you need to be in LLMs and that's part of it. But also many businesses are just upset. They don't want to deal with Google anymore. After all of that, what happened? And unpredictability and lack of reporting from Google, you cannot put any price on rankings anymore because you can be viewed, never clicked and it just doesn't make sense to businesses. So they just, okay, no more SEO. We want, we want everywhere. Oh, definitely, definitely. It's very Italian. I'm not sure if you are seeing that or maybe I'm just attracting those kinds of Leads, but most of them come to with I want to be in chat, GPD and stuff, not even mentioning Google. And I always say I will look in organic search anyways whether you want me or not because that is definitely more data there than anywhere else. And it's the first confident step into LLM. Visibility is definitely SEO, but the initial request is very seldom Google at this point.
21
00:33:26.000 --> 00:36:31.000
I would phrase it this way. I also run a small business interview and it's usually posed as a question of, you know, how much of an overlap is there on what I need to do for SEO and LLMs. And I've changed my position on this a couple of times and I used to think they were adjacent and there were things you did for SEO that weren't overlapping. But no, my mindset is now this is SEO and you know, AI overviews completely overlaps. Everything that you can do for SEO, do it because it all does impact. But there are are so many genuine different zingy dingies, different aspects, different departments to interact with and levers to pull, buttons to push, people to hire, things to publish to unpublish that have no impact directly on SEO and might even be negative in some places. But they're required as part of the bigger picture now. So it is a question that comes up, but my answer is, you know, SEO is still at the core if you know, I think Bernard, Jason Bernard covered it pretty well of, you know, you've got the training data set within LLMs is still partially, you know, is is one piece of it and that data entry is similar to SEO because those are fundamental entry points, you know, knowledge graph which impact SEO. But then you have the rankings itself. There's many searches, many results where AI, whether it's cloud, GPT, LLM or overviews, they're just rehashing search results. So you gotta have the search results to impact LLM outputs. But there are LLM outputs that you can't impact through SEO. So it's not a two way street everywhere. There's certainly a core element to it. But anybody that is hanging their shingle on only doing SEO, you know, you either need to be doing that for entity or for elite agencies and enterprises to be able to call that ticket home. But you got to start hybridizing, you got to start looking at additional skill sets and adjusting to the times. I mean we adjusted from, you know, I remember when guest posting was du jour ascendant and it was what we did and then Google came stomping along and the industry, you know, that industry standard got stomped on and you know, that didn't work anymore. Well, we gotta. We have to maintain that flexibility and if we pine for the good old days, we're just not going to get paid.
22
00:36:31.000 --> 00:40:11.000
Yep. And. Well, my point here is that SEO is my first step in auditing LLM visibility for several reasons and the main thing is that it has the most reliable data still. We really cannot look at LLM and figure out what people are even prompting. We don't have that data. But if we look at search keywords we may assume all the prompts behind them in a much more educational way. So I look at SEO data a lot still and for that exact reason we don't have firsthand data. We don't have as good predictability with with intense questions people may be asking unless we look at keywords and base. And yes, also there is a huge correlation. If the website is well represented in organic search, it will be in AI answers as well. There is it's a direct path and again the most predictable one. We don't have a better predictable path into LLM answers except for organic rankings at this point we have good theories, good playbooks, not very good reliable ways to measure the success of all those efforts that we are doing. Direct access success in MLMs. But SEO is still the most predictable path into LLMs, not the only one. Definitely not 100% overlap there, but 30% overlap is better than no overlap at all. Like some of those, some for some of our tactics even like Reddit and I'm in the Reddit pool a lot. How do you report on the success? We report on Reddit official subreddit LLM citation. But if you look further into those Just because your thread is being cited in MLMs doesn't mean you will be necessarily part of the answer. There is a huge level behind that where they again like just like we mentioned LLM consensus, they have to see you in many more sources. Just one mention in Reddit thread will not make you visible in AI answers. So Reddit visibility doesn't necessarily translate into LLM AI. AI answer presence can be citation, but who cares about citations. They don't. They are not even clicked. The goal is for the brand to be in the answer and unless you are the only predictable path into that is again SEO. You can do all other stuff like YouTube, credit, all of that. But the predictability is much of those efforts is much lower than SEO success still maybe changing, but at the most at this point effective way to have at least some visibility predictability in AI efforts will fall. So for that reason, even though most of my leads do not ask me anymore for that, I always say I will start with you and SEO and that's how I will understand most of what where we are going with LLM visibility as well.
23
00:40:11.000 --> 00:40:38.000
Give a final shout out for your brand if you have any resources. A newsletter where people can find you, discourse with you, learn. You said mentioned a live stream. I'll make sure that all of these things get linked in the show notes for anybody that's listening. A copy of this episode will go up on unscripted SEO.com so if you want more details, pop on over there. But in the meantime Anne we close out let them know where they can find you.
24
00:40:38.000 --> 00:41:46.000
Absolutely. Thank you so much first of all for having me here. Great discussion, lovely catch up with old friends of SEO. So my company website is Marty Marketing and there you can find a free option to schedule a 15 call with me. I'm a terrible salesperson so chances are I will not even try to to sell you anything. But I will be helpful. That that I can promise. Unless you ask me to sell you something. I never do that. So schedule a 15 call with me. We'll exchange ideas, brainstorm together. I will probably help you at least give you direction of where to look for and what to do with your small or big businesses. I really don't care. And smarty.com is where my newsletter lives. I announce my live streams there, give the link to join and also give some context for the topic. And other than that, those two I guess would be great. I'm also on LinkedIn as an smarty on X SEO smarty as well. Thank you so much.
25
00:41:46.000 --> 00:41:49.000
Thanks for stopping by.
1
00:00:00.000 --> 00:00:32.000
Hello, I'm Jeremy Rivera, your unscripted SEO podcast host. I'm here with Anne Smarty of Smarty Marketing. If you haven't been around SEO for any amount of time, then you haven't seen her. But if you've been around, you know that she has a big footprint in our industry. Super excited for this interview. But for those unfortunate souls who aren't familiar with you and what you've been working on, give yourself an introduction, tell us about your industry niches, some of your experience, and share with the people who you are.
2
00:00:32.000 --> 00:01:20.000
Interesting question. I've been an SEO for more than 20 years and I've done pretty much everything in those 20 years from SEO journalistic reporting to SEO community management to SEO analysts to everything. So I've seen pretty much all sides of it. Currently, what we are doing SEO audits plus geo. I don't do those separately anymore because they overlap so much. And Reddit, we are very much focusing on Reddit. Not because we really chose that, but we've been on Reddit for more than 15 years and suddenly Reddit became a thing. So we started selling it as a separate service because we do have a global experience in how it works and what to do there.
3
00:01:20.000 --> 00:01:47.000
The Narwhal does Bacon at Midnight. Tell us a little bit more about that Reddit activity. Is it just find relevant threads and dive bombing in with links, or is it building your own subreddits and trying to gain audience? Like what's the, what are a couple of the plays that you're offering people when it comes to Reddit?
4
00:01:47.000 --> 00:03:14.000
We definitely focus on long term transparent branded existence on Reddit, which unfortunately takes more time than the first method that you mentioned. But it's long term and it's not. It's definitely a long work, a lot of it, a lot of effort. Our average is 6 months. When the community starts really picking up and that subreddit starts ranking in, Google starts showing up in LLM citations, but on average six months at least. But then once you have your subreddit, you definitely can branch out to different subreddits as well and build your presence there in a meaningful way. With new accounts. It's the best way to get suspended right away. And not even with new ones, with established ones. Once you try, once you start trying to do sneaky stuff, you can lose accounts. And that's exactly why we don't do that for clients because we, we operate their official presence. We have to be very careful with it. It's a lot of responsibility and anything Spammy you do on Reddit can backfire on the brand brand reputation and that's too much of responsibility to take. So we focus on a safe, transparent and meaningful way of building Reddit reputation and Reddit presence. Basically.
5
00:03:14.000 --> 00:03:26.000
Well, you stole a question from me because I was going to ask how do you crash out fast on Reddit? But I guess if you'd sum it up, what's the fastest way you've seen a project crash and burn?
6
00:03:26.000 --> 00:05:22.000
If you think about that, it's not very much different from SEO in general. There are quick ways to do stuff in SEO and achieve those quick results and there are long term and that all relies on how much you value your actual entity or brand. The there is a gateway to Reddit if you are ready to gain quick results and then flip to something else and lose everything. That's an approach too. Can happen as fast as one day to six months can. I saw very well established accounts, almost 10 years all being suspended on Reddit when they start trying to do something differently and quickly there. So regardless of how established your account on Reddit is switching to trying to be very promotional, linking to your website all the time is still risky. So it can happen quickly, it can happen in the long run, but it will happen once you start trying to game it. And it's not so much different from SEO if you remember our old days. And it's it's also a very popular model still to create a website, achieve traffic within 3 months by publishing AI generated high volumes of AI generated content, enjoy your 3 months of traffic and then get suspended and switch to another website doing pretty much the same. People make good money using that method and who am I to say if it's right or wrong because we work on behalf of clients, we just cannot afford that freedom and flexibility to game stuff, you know, and experiment with things. So we have to be more careful than anyone else
7
00:05:22.000 --> 00:05:54.000
does strike me that, you know, a lot of people think of their homepage as their home for their entity and business. But with geo, with AI overviews and LLM layers, it seems like a more distributed model of understanding your whole business footprint. Online seems to be a much more holistic and useful model in this moment of SEO. Would you agree?
8
00:05:54.000 --> 00:08:30.000
I definitely agree. I would still not neglect my own website for sure. I've seen a lot of changes in LLM answers once I went back and updated my very old personal website because they kept pulling information from it and it was like 22 years old and much touched for at least five years. So I would definitely do the homework. That means making sure your website is up to date. You have a very detailed about page, contact details, all the trust factors, everything that you can do on your website, definitely do it. Don't forget about that step. But it's definitely not enough. You have to have. I refer to that as LLM consensus. This is why LLMs are able to find a lot of different sources about you and they are very consistent in their messaging. Don't be like me. My footprint is a mess. Some of my bylines still talk about my projects that are no longer there, have not been there for 10 years. Many roles, many inconsistency and I sometimes I try to go and fix stuff but it's never ending. I have a lot of footprint accumulated for 20 years, but I'm also not as much concerned about LLM visibility because I have another ways to build my business. If you are very concerned about being visible in AI answers, fixing this whole digital footprint to be up to date and consistently saying the same thing about you, actually your value proposition and how you are different from your competitors is the very first step. It's LLM consensus. Audit can be done in house, takes lots of time, especially for older businesses as well. But you will see the difference right there and then identifying channels that make a difference in LLM training data or citations and work for your competitors where you're not present and gradually develop your presence there as well. Usually Those are Reddit, YouTube, Wikipedia and a few very high profile publications which are not as easy to handle, but it's doable. But it's definitely not just your website these days or even not just your rankings.
9
00:08:30.000 --> 00:10:39.000
Yeah, that came up with my interview with Grant Simmons. We were talking about, you know, Kalicube and thinking about it in terms of, you know, building an entity home on your about page. It's crazy how many people like I worked with four or five brands and they had no about page at all. Or their about page was like one or two paragraphs and it was just the most generic, fluffy stuff possible. I'm like, come on, there's. I know more. I haven't even worked with you yet, but I know more about your company than is spelled out on your about page. So there's definitely that. And then you're right, you gotta, you know, educate your customer support your exterior customer support representatives, Claude and chatgpt, they need good intel, you know, that's what my friend Matt Brooks at SEOT says, train them, you know, give them the materials like, and you're right there's for older entities and businesses. You do have that spread of old profiles, social profiles, that this snippet of information and it can be a headache trying to find credentials for, you know, a five year old, you know, blogger account and trying to get things updated. So I guess that's where, you know, we as consultants and can earn our keep of hunting those things down, impressing the value of that consistency. But I'm curious if there's anything, if there's a playbook or like a walkthrough. What are, what are the top five things aside from, you know, hiring you to go set up Reddit for them, that if a new business, you know, because I'm working with a new project, you know, they're doing AI energy power for data centers. That's like clean or whatever. But what would you suggest for a new project getting off the ground? How do you think about gaining that visibility? Where do you go? Where do you. Is it buying press releases? Is it popping up profiles? Is it classic directory submissions? What's your top five?
10
00:10:39.000 --> 00:15:11.000
Don't dismiss directory submissions that easily. These days they are being very well used by LLMs, which is like they made a very good comeback from, I don't know, 20 years ago. So my answer is very it depends basically on the business, what their budgets are. I would usually suggest do the fundamental homework which is establish your profiles, name consistency, get registered where it makes sense for many businesses that could be Yelp, Trustpilot, all of those places that are being popular in your niche. Basically, if it's B2B that be could we go to what, whatever, whatever is in your industry. Claim your profiles everywhere, create a very well organized system of how you contribute everywhere. It's very doable with repackaging, recycling your actual website content if one asset can be a good update for 10 profiles. So just create that system that works for you at the bare minimum, very consistent. Contributing is in many cases enough to start being visible, especially at the beginning. And then you can also branch out to newsletter sponsorships, paid advertisement ads on Reddit and elsewhere, wherever your audience is. But this organic work should be the first step because once you start growing, the people will start searching for you. And that's when all the problems start. Because brand being a brand is great, but no one talks about how it actually you lose control over it because stuff starts ranking for your name and that is the most important search query. That's your brand name variations, brand name reviews and all of that. And unless you do that homework before you start being very actively Outbound, you will see trouble very quickly. So the steps would be establish your organic footprint first, then find where your audience is and there are lots of ways to reach out to that audience. Being a podcast guest or starting your own podcast, live streams, whatever you can afford the time and team to do and let if you do that every week, even a little bit counts. But start doing something before investing in ads would be my I've seen a lot of, a lot of startups that for a month or two they were enjoying themselves and then people start using them and start discussing them on Reddit and now all the fun stops right there because no one wants to deal with them. Everyone is searching for brand names, including LLMs. Before they recommend a brand in an AI answer, they will do the branded search. They will find stuff about you that your customers may not want to know. So start doing something real and establish your branded queries very well. Control as many results there as you can. And I think in your newsletter somewhere I forgot which one I actually shared that trick. Do the bare minimum you can afford, just do it consistently. For example, I cannot afford a very well polished videos doing guest podcasts because that's a lot of coordination and my team is slammed with client work and I obviously cannot do all of that myself. But the bare minimum for me is my live stream is on my calendar. I make sure every Wednesday I join it and I talk about what happened in the industry, on how, what to do about that and that alone drives a lot of business just because I do that every week consistently. So everyone can decide depending on what they can afford. But consistency is the key.
11
00:15:11.000 --> 00:16:51.000
It's fascinating because I've been doing this for about 20 years too, and I am embarrassed to say that I recall very clearly a meeting where I said the words I don't care about your branded search in your Google search console. I'm filtering it out so that I can focus on unbranded keywords and phrases. And the paradigm has shifted so hard, you know, partly because of HCU, but also you're right, there's LLM factors and LLMs are a lot less forgiving for bad bad mouthing that it goes against your brand anywhere online. They're, they're programmed to surface. Oh, is there a trouble in the product? And you know, when I interviewed John Henshaw, he's like some of the the best thing you can do for SEO sometimes is turn inward and fix your broken ass product. You know, if you have something that's straight up janky or broken and it's gotten bad reviews. Whether it's your workflow and delivering results, if you're a SaaS, if there's a feature that's not doing what it's supposed to, people are going to talk about that. So, you know, you need to, you know, turn the eye inward. Are you finding as an SEO that you're now with these changes, you're able to get deeper into these organizations and you're not just held at bay and like, oh, can you SEO our website?
12
00:16:51.000 --> 00:19:45.000
Interesting. We used to be an island. Remember those times when we were absolutely not aware of what else is going on with our clients? As long as we have our keywords and can report some growth, that's everyone, what everyone cared about. Now I'm much more involved in reputation management. Organic, especially organic search. But Even otherwise, because LLMs search spontaneously, they can search Google or they can search Reddit itself for everything that you were trying to hide from Google search results. And you're right, I have to deal with so many departments, sales, customer support. You would not believe how much of reputation problem is caused by this communication between what the website says, what sales people promise, and that causes so much problem on Reddit and thus everywhere else from there that I have to have almost separate meetings to identify why those problems occur. Like, where is that disconnect? In many cases that, okay, we cannot say that publicly, but we can say that privately. And I mean it's no longer. You cannot hide anything. If there is an upfront price and you are not talking about that upfront, people will find that on Reddit. And in the quick AI answer right in the AI overview will tell you how much you, you charge upfront and you cannot hide that information anymore. So alignment inside companies becomes the key to LLM visibility and it's a fight. I spent too much time figuring out what we cannot say publicly why and what does it take to change that policy. Because it's already there public, and there is no hiding that. And in 50% of cases, I cannot win that battle because those executives are still like, no, that's how we make money. We cannot say that. And I mean like, okay, then this is, this is what we end up with. And then let's change the strategy instead. Instead of trying to fix it, we should be doing something else there. And usually that's building brand awareness in a different way. We have a lawyer client which charges more than everyone else and we're not allowed to talk about why, even though there is a clear reason. They take more difficult cases overall. But for some Reason they don't allow us to tell that. So we focus on hosting MaaS for them and just build their trust and authority that way without acknowledging the big elephant in the room as well. But we do what we can.
13
00:19:45.000 --> 00:21:43.000
I'm curious about how AI, you know, we, I, it's, it's sad that we only have the one word for it, but I, I think there is a true differentiation between the workflow and augmentation side, the practical utilization of Claude code or desktop between your team, multiple different models of AI to unlock capabilities. You know, I was talking with Brett Sklar just yesterday about how if you just take, take away one headache per week and just ask your team, use these AI tools and cure a headache you have this week. You know, be inventive and see what you come up with. That small progression instead of trying to replace workers, just eliminating tasks is just like SEO on steroids. There's so many workflows, so many creative things that it's way better than spending your SEO budget. You know, I know that I charged somebody $2,000 to edit Meta titles and descriptions because they demanded them they be bespoke and human written, every single one and they wanted to pay for the hours for it and they thought that was worth it. But I would absolutely not Recommend that now 10 years later it's not worth it because you can do so much more with $2,000 worth of budget and you can still get the titles and descriptions written really well, possibly even better, you know, if you really have honed your agents or your skill set to do so. So I'm curious what you're finding within your company, within your agency, when it comes to leveraging AI as a workforce element.
14
00:21:43.000 --> 00:24:41.000
Well, it's an interesting question. My team is mostly at this point they have mostly involved in two big tasks, which is one, community management and we use AI for tracking, analyzing, sentiment, all of that. But there is a lot of manual work there for sure, because it's Reddit. We write manually there, we don't automate it, the actual posting. So aside of analytics and monitoring, it's all manual work. The other part that they are doing is pr, digital PR assets. Likewise, we can use AI for ideation, competitive research and stuff like that, but it's not a good way to invent something actually original and unique. So we use for direction. But most ideation comes from brainstorming actual ideas by people. Building AI is handy for internal linking to that asset, to coming up with suggestions on how to reach out to journalistic research, things like that. But the core of that product is still manual work. I personally am not as much involved in those tasks, but I do entry audits and I mean, I use AI here and there. It's very connecting. Semrush or Ahrefs to quote, is a great way to kind of get insights that you didn't wouldn't have thought about. Competitive research. Great with AI as well, and analysis. But I also do a lot of manually there as well because I feel like maybe I'm old school, but unless I dig deep and see actual search results, actual queries, sort them, filter them, I don't feel like I get a good understanding of the health, SEM, positioning health of the website and where I should go. This is my brainstorming method before I create the roadmap of the strategy. I actually spend week, three days, depending on the size of the website, just looking at the data myself. Again, I can use help with Claude or Gemini for usually those two. I use things like analysis, brainstorming, things like that. But definitely a huge core of work in our company is still manual. Maybe because we picked up the services that are not easy to automate.
15
00:24:41.000 --> 00:27:58.000
It makes sense that, you know, we, that humans find the place where, you know, you're dipping into multiple different pools of data and have to have in mind, you know, a certain filter and parameter on it. Like if you, you know, I wrote about this on Open SEO. Like if you treat Google Search Console data as totally credible, then you're overlooking, you know, the dark queries, the fact that between 30 to 80% of the query table details, they're gone, you just can't see. And so you're basing your entire strategy of, you know, oh, we got a thousand clicks, but out of those thousand clicks, you got 200 of those that were told, you know, where they were, which pages for what keywords. That's less than a flip of a coin that, that's sticking your finger in the wind and saying what, what direction is the wind blowing? Kind of accuracy. That's it. But so. So it's about choosing and descri between these sources. For me, a big unlock has been honestly connecting MCP for Google Search Console, Google Analytics, Microsoft Clarity and keyword research data platforms like Open SEO or Semrush or Data for SEO and just asking for it to pull from the API and put the data next to each other for me to evaluate. You know, my favorite thing with Claude code is open this in a browser, show me all of the different types of data, slice it this way, and not ask it to analyze it, but just show me the data that's so hard to navigate to in GA4. If I have to open GA4 again, I'm going to hit my head on the desk till I pass out. But I can ask Claude, hey, show me GA4 data in a GA3 format. You can get it to do that and pull the same information and display it in friendly stacked sections where you know, oh, these are the pages. Sort by sessions. You can. You have that on your finger, fingertips. So it enables human analysis at a much quicker pace. If you can pull from. Pull via API and mcp and then you can apply your human knowledge of. Okay, GSE is cropped this way. Microsoft Clarity and GA are going to overlap, but they're going to report the same metrics differently. Bing Webmaster Tools, actually surprisingly good amount of data there to be able to pull. Microsoft Clarity added bot information. So GA4 reported there were 80 visits and 40 of them went into the direct bucket. That's what Microsoft Clarity told me. Those 40 were bots. And that's not necessarily bad anymore. It used to be like, keep the robots off my site. But now 50% of the traffic is bots. And I gotta say, some of those bots are running around with wallets. They can spend money.
16
00:27:58.000 --> 00:27:59.000
Yes.
17
00:27:59.000 --> 00:28:03.000
So don't. Don't treat them like strangers anymore.
18
00:28:03.000 --> 00:29:18.000
Yep. It's interesting the, the time, right. The switch that we had. Well, totally with you, honestly, because I don't deal with keywords as much as I ever did. I'm not into those actual search queries that much and not by choice. Just happened gradually for me because somehow I ended up more on the branded query reputation management side of things. And with Reddit and jio, whatever you call it, analysis, that just happened very, very easily. Most people, mostly that I get, don't ask about Google rankings that much anymore. They come to me with, we want to be on Reddit and we want to be an LLM. And search queries especially digging deeper into those isn't that much of my routine anymore. I actually miss those times now that I heard you speaking about that. I'm like, oh my God, I haven't done that for four months.
19
00:29:18.000 --> 00:31:52.000
It's a changing game. It certainly is a different ecosystem, largely because, you know, there was that. It's because Google had a true monopolistic moment and squandered it. You know, they had the golden moment in the sun where being not a real competitor, no one could come close. Yahoo was a joke and they just got so high on their own supply. And we got high too, you know, because we could, like you said, we could stand on our little island separate. And you know, I had one client, honestly, I never spoke to, I was part of an agency. I did SEO for them. Never met anybody on their team, never was in a single meeting with them. There was a salesperson who talked to the customer support person who talked to me to do, quote, unquote, do their SEO. And that level of separation that, that doesn't exist anymore because Google's not the solo primary channel. And even they've. They did it to themselves. They, they very quickly rushed out AI overviews when they saw the threat of LLMs on the horizon and kind of spiked their own wealth. And now that I guess they're enjoying the, the dark box, the, the bringing back of the dark ages. But when there isn't clarity for these businesses of how much value that Google is driving to them, and there's all these LLMs that have messaging that's ascendant, that are reaching C suite through LinkedIn, talking about, I don't doom and gloom about anymore. There was a moment where I genuinely was really concerned about the industry. But I think there's a level of acceptance now of, hey, this is what it looks like when a monopoly loses its monopoly and we get true competition back onto the table. And Google's going to have to make its choices and we're going to have to make our choices as to what surfaces or places we try to optimize. But it's as I'm writing a book right now of Search Everywhere Optimization. The acronym doesn't need to change. It's not geo, it's SEO, but it's Search Everywhere Optimization, not Search Engine optimization.
20
00:31:52.000 --> 00:33:26.000
Yep, and I totally agree with you. It's very telling. I haven't had a lead in months that would come to me saying I want to be on Google businesses, just not just probably. There is a lot of, of course noise telling them that you need to be in LLMs and that's part of it. But also many businesses are just upset. They don't want to deal with Google anymore. After all of that, what happened? And unpredictability and lack of reporting from Google, you cannot put any price on rankings anymore because you can be viewed, never clicked and it just doesn't make sense to businesses. So they just, okay, no more SEO. We want, we want everywhere. Oh, definitely, definitely. It's very Italian. I'm not sure if you are seeing that or maybe I'm just attracting those kinds of Leads, but most of them come to with I want to be in chat, GPD and stuff, not even mentioning Google. And I always say I will look in organic search anyways whether you want me or not because that is definitely more data there than anywhere else. And it's the first confident step into LLM. Visibility is definitely SEO, but the initial request is very seldom Google at this point.
21
00:33:26.000 --> 00:36:31.000
I would phrase it this way. I also run a small business interview and it's usually posed as a question of, you know, how much of an overlap is there on what I need to do for SEO and LLMs. And I've changed my position on this a couple of times and I used to think they were adjacent and there were things you did for SEO that weren't overlapping. But no, my mindset is now this is SEO and you know, AI overviews completely overlaps. Everything that you can do for SEO, do it because it all does impact. But there are are so many genuine different zingy dingies, different aspects, different departments to interact with and levers to pull, buttons to push, people to hire, things to publish to unpublish that have no impact directly on SEO and might even be negative in some places. But they're required as part of the bigger picture now. So it is a question that comes up, but my answer is, you know, SEO is still at the core if you know, I think Bernard, Jason Bernard covered it pretty well of, you know, you've got the training data set within LLMs is still partially, you know, is is one piece of it and that data entry is similar to SEO because those are fundamental entry points, you know, knowledge graph which impact SEO. But then you have the rankings itself. There's many searches, many results where AI, whether it's cloud, GPT, LLM or overviews, they're just rehashing search results. So you gotta have the search results to impact LLM outputs. But there are LLM outputs that you can't impact through SEO. So it's not a two way street everywhere. There's certainly a core element to it. But anybody that is hanging their shingle on only doing SEO, you know, you either need to be doing that for entity or for elite agencies and enterprises to be able to call that ticket home. But you got to start hybridizing, you got to start looking at additional skill sets and adjusting to the times. I mean we adjusted from, you know, I remember when guest posting was du jour ascendant and it was what we did and then Google came stomping along and the industry, you know, that industry standard got stomped on and you know, that didn't work anymore. Well, we gotta. We have to maintain that flexibility and if we pine for the good old days, we're just not going to get paid.
22
00:36:31.000 --> 00:40:11.000
Yep. And. Well, my point here is that SEO is my first step in auditing LLM visibility for several reasons and the main thing is that it has the most reliable data still. We really cannot look at LLM and figure out what people are even prompting. We don't have that data. But if we look at search keywords we may assume all the prompts behind them in a much more educational way. So I look at SEO data a lot still and for that exact reason we don't have firsthand data. We don't have as good predictability with with intense questions people may be asking unless we look at keywords and base. And yes, also there is a huge correlation. If the website is well represented in organic search, it will be in AI answers as well. There is it's a direct path and again the most predictable one. We don't have a better predictable path into LLM answers except for organic rankings at this point we have good theories, good playbooks, not very good reliable ways to measure the success of all those efforts that we are doing. Direct access success in MLMs. But SEO is still the most predictable path into LLMs, not the only one. Definitely not 100% overlap there, but 30% overlap is better than no overlap at all. Like some of those, some for some of our tactics even like Reddit and I'm in the Reddit pool a lot. How do you report on the success? We report on Reddit official subreddit LLM citation. But if you look further into those Just because your thread is being cited in MLMs doesn't mean you will be necessarily part of the answer. There is a huge level behind that where they again like just like we mentioned LLM consensus, they have to see you in many more sources. Just one mention in Reddit thread will not make you visible in AI answers. So Reddit visibility doesn't necessarily translate into LLM AI. AI answer presence can be citation, but who cares about citations. They don't. They are not even clicked. The goal is for the brand to be in the answer and unless you are the only predictable path into that is again SEO. You can do all other stuff like YouTube, credit, all of that. But the predictability is much of those efforts is much lower than SEO success still maybe changing, but at the most at this point effective way to have at least some visibility predictability in AI efforts will fall. So for that reason, even though most of my leads do not ask me anymore for that, I always say I will start with you and SEO and that's how I will understand most of what where we are going with LLM visibility as well.
23
00:40:11.000 --> 00:40:38.000
Give a final shout out for your brand if you have any resources. A newsletter where people can find you, discourse with you, learn. You said mentioned a live stream. I'll make sure that all of these things get linked in the show notes for anybody that's listening. A copy of this episode will go up on unscripted SEO.com so if you want more details, pop on over there. But in the meantime Anne we close out let them know where they can find you.
24
00:40:38.000 --> 00:41:46.000
Absolutely. Thank you so much first of all for having me here. Great discussion, lovely catch up with old friends of SEO. So my company website is Marty Marketing and there you can find a free option to schedule a 15 call with me. I'm a terrible salesperson so chances are I will not even try to to sell you anything. But I will be helpful. That that I can promise. Unless you ask me to sell you something. I never do that. So schedule a 15 call with me. We'll exchange ideas, brainstorm together. I will probably help you at least give you direction of where to look for and what to do with your small or big businesses. I really don't care. And smarty.com is where my newsletter lives. I announce my live streams there, give the link to join and also give some context for the topic. And other than that, those two I guess would be great. I'm also on LinkedIn as an smarty on X SEO smarty as well. Thank you so much.
25
00:41:46.000 --> 00:41:49.000
Thanks for stopping by.
