Can AI Replace Your SEO Team? Think Again with Tom Winter
Show Notes
Can AI really replace your SEO team? Think again! 🚀 In this eye-opening episode of the Agency Insider Show, we explore how AI is revolutionizing SEO while uncovering why a human touch is still irreplaceable. Join host Navneet Kaushal and SEO expert Tom Winter, founder of SEO Wind, as they share actionable insights to help agencies and professionals rank your website number 1, drive more organic traffic, and stay ahead in the ever-evolving SEO landscape.
Discover step-by-step strategies for leveraging AI to humanize content, streamline workflows, and optimize search engine rankings. Whether you're curious about GPTs, AI workflows, or the role of AI agents in SEO, this podcast breaks it all down. Plus, find out how agencies can use AI-powered tools to achieve a #1 ranking without losing the personal edge that clients value.
Perfect for both beginners and seasoned pros, this episode is packed with latest SEO strategies, practical tips to improve visibility, and smart ways to future-proof your approach. Don’t miss this chance to learn how to harness the power of AI while keeping your SEO game strong!
🔥 Want to stay competitive and rank #1? Watch now and start turning insights into results!
Chapters:
00:00 - Intro
00:59 - Tom’s Journey: What sparked Tom's passion for SEO?
02:52 - Can AI Content Rank: Is AI-generated content effective for SEO?
06:44 - How Can Agencies Leverage AI for Growth: What strategies can agencies use with AI?
09:31 - Humanizing AI Content: How can we make AI content resonate with audiences?
13:39 - Will AI Replace SEOs: Are SEOs at risk of being replaced by AI?
15:34 - How to Keep Up with AI in SEO: What are the best practices for staying current?
18:44 - When and How to Use AI, GPTs, AI Workflows, and AI Agents: What tools should you implement for efficiency?
29:44 - How to Use AI for Local SEO: What are the benefits of AI in local search optimization?
31:50 - How AI Always Answers: How does AI handle user queries effectively?
34:30 - Biggest Mistakes Agencies Make When Integrating AI: What pitfalls should agencies avoid?
38:08 - Importance of White Label SEO Solutions: Why should agencies consider white label services?
40:14 - How to Make AI Content More Engaging: What techniques enhance AI-generated content?
43:34 - Importance of Long Form vs Short Form Content in Today's SEO Landscape: Which format performs better in SEO?
44:55 - What It Takes to Rank #1 on Google: What are the key factors for achieving top rankings?
48:49 - Advice for SEO Consultants and Agency Owners: What insights can help you succeed in SEO?
52:12 - Rapid Fire Round: What quick tips can you share for SEO success?
54:01 - Where to Find Tom Winter Online: How can you connect with Tom Winter?
Transcript
Intro
0:00 hello everyone and welcome back to G agency insider show i am your host Navit Koshel and today we have a very special
0:07 guest joining us Tom Winter tom is a serial entrepreneur the founder of SEO win and an expert in leveraging AI for
0:14 growth he has helped countless SAS companies scale globally today we are diving deep into how agencies and SEO
0:21 professionals can leverage AI for growth optimize SEO strategies [Applause]
0:33 [Music] Tom welcome to the show welcome Nit like
0:41 it's amazing to be here uh and I hope to share with you as much of information
0:46 things that you can implement as possible so welcome everybody yeah and
0:51 Tom you have built SEO wine from the ground up work with major companies at DevScaler and you have deeply been
0:58 involved in the AI space for our listeners who may not be familiar can you tell us a bit about your journey and
Tom’s Journey: What sparked Tom's passion for SEO?
1:04 what you are currently focused on okay so I'm a serial founder uh half a
1:10 developer half a marketer uh I started with a marketing agency over 15 years
1:16 ago then I started to work with a couple of friends on devs clear as you mentioned which is a tool to test
1:22 developer skills and map out the skills inside the company uh we work there with
1:28 Fortune 500 companies and with companies in over 100 countries but right now what
1:33 I'm focused on is seowind.io which is a tool to write articles based on research
1:40 uh using AI but like for us the research part is 80% of the process so like we want to feed AI with as much knowledge
1:48 as possible for that we're using multiple ways like we're sourcing data
1:53 from different places we're using retrieval augmented generation uh a lot of vectors to combine things together
2:00 and be able to feed it into AI and this is what I'm focused on right now i'm in
2:05 love with AI uh but to be honest most of AI that I'm using is like through API so
2:11 this is a lot more fun uh because you have more control over the whole thing
2:17 and one more thing that I really love to use with AI is embeddings something that
2:22 most of the people's don't hear about like on normally normally but it's just basically turning everything into a
2:29 vector so we can tell turn any kind of text any keyword any heading any whole
2:34 article or an image or a video into a vector and then start comparing it together and trying to pull things that
2:42 are relevant uh and that's magical that actually sounds like magic
2:51 yeah so when you talk about the content so because a lot of time people say AI content doesn't rank of course and
Can AI Content Rank: Is AI-generated content effective for SEO?
2:56 second is if you want to do rank AI content you can't do it at scale so what
3:01 with SEO is this the same thing or does it can you scale it can you scale it
3:09 definitely you can but like when you're talking about AI writing uh most of the
3:15 tools like apart from AI uh itself like so chat GPT claude Gemini Grog and so on
3:23 uh when you're looking at most AI writing tools what they're doing they're using AI as search engine so like it's
3:31 rather simple way of using that like they're asking a question write me an outline for this and that and then based
3:37 on this outline write me an article for this article but like for this specific outline but this way you're
3:43 just using the knowledge that AI is trained on h and to be honest like when
3:50 looking at how AI are trained like first of all there's a cut off in data like
3:55 that is at the moment of training AI so this is the first problem so like there
4:00 will be no up-to-date things the other thing is that it doesn't mean that AI when it was trained on specific data it
4:07 has all the data in one finger it will hallucinate it will try to logically
4:15 look at the things that it would train on and provide you with information we try not to do it this way we use AI as a
4:22 logical engine so it's similar to how you use your brain if you're a human writer if you have like a specific topic
4:30 even if you're like very familiar with a topic if you would be writing the article you would go to the internet you
4:35 would find additional data maybe some statistics maybe some uh information about top like from topical authorities
4:43 uh maybe some thought leadership insights the case studies and you would pull it all in and you would try to
4:50 write based on this information this is exactly what we're trying to do we're going into multiple data sources from
4:57 SEO sources to specific places on the internet to source additional data based
5:02 on the outline that we're creating and to pull the data from there to be able
5:08 to provide it to AI in a contextual digestible way so it actually knows what
5:14 to do it's not about asking just like write me this or that but like look AI I want to write this but these are
5:20 statistics that you can actually include here this is what our special matter expert said about this part of the
5:26 article these are additional case studies that you can use please write uh the article based on
5:34 the data that I'm providing you with and this way you're not creating fluff but actually an article that is full of
5:40 value so exactly what EAT talks about uh so you're providing the information
5:46 inside the article that people are looking for so Google I'm always saying Google hasn't changed much for the last
5:52 25 years even that everybody says that SEO is dead multiple times already
6:00 and all the algorithms change and like make the world upside down google wants
6:06 a simple thing the best value to the end user this is the definition of SEO like
6:11 if you're thinking about it today best value to the end user and it was like that 20 years ago right google became a
6:18 far better referee in the last 25 years but the goal that they have quite the
6:23 same nothing different so this is the main difference between AI writing tools
6:28 like so if you're using AI as a search engine or actually you're using it as a
6:34 logical engine that takes into account many data points provided to AI and use
6:40 it to make valuable content okay now let's talk a little bit further about AI
How Can Agencies Leverage AI for Growth: What strategies can agencies use with AI?
6:46 i know of course it's an hottest topic and you have been sharing of AI related insights now what are the biggest
6:53 opportunities you see for agencies when it comes to leveraging AI for growth
6:59 so first of all I can see a big trend because I talked to many agencies i I can see a big trend that many marketing
7:06 agencies are establishing something that they call uh AI councils inside the the
7:13 agencies this means that they gather people from every niche of their company
7:19 to talk on regular basis about where we can input information like input AI into
7:26 the task that we do so this is like really amazing what they do because
7:32 they're actually looking at the company from the perspective okay we might figure out like here and here we can
7:38 tweak it and make it better uh normally what they're after uh from what I see is
7:44 any tasks that are repetitive uh that everybody hate uh that AI can
7:51 take over and make it better so one of the key things of course like from my
7:58 perspective is uh building content and how to use AI for research part and then
8:03 writing the content based on research but it can be for example something like connected with reporting so like every
8:10 agency needs to do reporting and this is a repetitive task that you do every
8:15 single week or month and this is a perfect thing to do so like one of the
8:21 things that they're doing they're like talking to everybody inside the company they're asking people what are the
8:26 repetitive tasks that you would like to be taken over reporting nobody likes it
8:32 uh so uh they try to automate these things so there's multiple things that
8:37 you can do uh using AI okay so I I like the concept of AI
8:43 council i mean I'm I'm sure bigger agencies are but smaller agencies with less number of people will still be
8:49 stuck in uh doing it i think it's still worth to talk inside the company even if
8:54 your company is like four to five people it's worth to talk like just to find a little bit of time during a month just
9:01 to be able to talk and exchange thoughts between people because when you're brainstorming like you will get better
9:08 ideas better results and don't be afraid of testing things out testing means that
9:13 you will break things that means that you will make mistakes and that's really
9:19 okay but do test don't be afraid of testing and trying out new things that you can do uh with AI or AI uh leveraged
9:28 tools okay now uh you recently mentioned in your LinkedIn post that humanizing AI
Humanizing AI Content: How can we make AI content resonate with audiences?
9:34 content is crucial in 2025 so can you break that down for us
9:40 uh humanizing is important uh I just to
9:46 say out loud in the beginning I don't believe that AI detectors uh tell you
9:52 anything like or that Google is against AI so this is what I want to definitely
9:58 state for me humanizing means proofreading text and make it more
10:03 digestible for the end user so like anything that you're doing with content should have a single aim not to cheat
10:10 Google but for like to make the text better for the users so humanizing for
10:16 me means uh simplifying the text uh removing a lot of garbage from the text
10:22 repetitive things from the text uh when you do that you will also increase the
10:28 if you have like a AI written article uh you will increase like decrease the
10:35 score in AI detectors uh and you can humanize using AI that's the funny story
10:42 uh one of my friends that is not in the niche like was when I told him like look
10:47 we're humanizing using AI he was laughing a lot how can you use AI to
10:53 humanize AI content it's like no like basically it's rewriting the whole thing
10:58 with a specific objective and that actually helps to make the
11:04 content better uh make it more digestible give additional insights give
11:09 additional storytelling to make the content more digestible for humans
11:15 so is there a specific way of doing it like or is there a set of prompts how
11:21 how exactly does humanizing AI content which is not I mean fast I would say
11:27 magic uh but yeah basically uh there like what we for for example do like we
11:35 have a specific prompt that we worked and we did like iterated a lot on that probably couple of hundred iterations to
11:42 make it work uh but there are also like other tools like what I would look out
11:47 because many tools that tell that humanize it and I'm like really against it uh what they do they they include
11:56 additional mistakes human errors into the text to humanize it for me that's a
12:03 nogo like this is something that you shouldn't do because that shows you that you're unprofessional and the only aim
12:10 of adding human errors into that is cheating AI detectors uh I think there's
12:15 like I can see that there's like way around it like you don't have to make mistakes human mistakes to proofread and
12:23 to uh cheat AI detectors you can actually make it better and remember
12:28 when you're humanizing spam all you will get is humanized spam nothing else right
12:37 so so when you talk about this prompt and I'm sure uh the listeners would love to get hold of that prompt is this a
12:43 prompt you open offerly offer openly or how does it we don't offer it openly but
12:49 definitely there is a free tool on our website that you can use to uh humanize
12:56 your content so like uh there are some limits because we're offering it for free but if you would like to humanize
13:02 your content you can definitely go to seowwind.io and in free tools you will find that there is a tool for humanizing
13:09 content and you can do it on our website so it's it's better than software like originality for that matter
13:16 uh it's definitely different like I didn't test all of them to be honest so I don't know how originality works and
13:21 to be honest probably originality provides you with that for money we
13:27 don't like it's for us it's like a side project that we just gave it to to people because we use it also in our
13:33 system all right all right now uh many agencies
Will AI Replace SEOs: Are SEOs at risk of being replaced by AI?
13:39 owners are hesitant about AI fearing it will replace them and uh or their team
13:45 thinks so what what's your take on that so often when I talk to the agencies uh
13:51 I hear uh a sentence that goes along these lines that like we're not
13:56 implementing this or that from AI because uh we will have to fire half of
14:02 our staff uh so this is something that I hear a lot and in in in this situation
14:08 what I'm trying to tell them like look by making your company more efficient
14:15 with AI you're actually securing jobs in your company because if you will not
14:21 implement uh any kind of tool it doesn't have to be AI but like in right now like AI is the topic if you will not
14:28 implement things that will increase your ROI efficiency make your work better in
14:33 a year or two you will have to fire 100% of people not 50% because your
14:40 competition will actually implement that and if they do they will increase their
14:45 efficiency and you can't compete with them i'm a half a developer half a marketer in many cases I'm not able to
14:53 compete with AI so if I'm not able to compete with it I prefer to team up with
14:58 it and this way like if I will increase the efficiency of the work that I'm doing in my company instead of losing
15:05 the jobs in our company I can make help our company actually grow to actually
15:10 increase the people that we can hire because our efficiency goes up and we
15:16 are able to add more people to the company if you think it from this
15:21 perspective all of a sudden it makes sense that you actually need to do it
15:27 otherwise there will be a far bigger problem than you see right now right now
How to Keep Up with AI in SEO: What are the best practices for staying current?
15:34 for agencies struggling to keep up with this rapid change in AI and SEO what options are available to them to still
15:40 deliver uh cutting edge results without needing to build in-house AI teams
15:46 i would avoid building in-house AI teams i know a lot of agencies big ones that
15:53 are over 100 people that decided at some point like look we need to build this
15:59 tool in house uh we will have like four or five people dedicated to a project
16:04 that will resolve our internal problem and we will build the tool so what happened in most of them probably nine
16:11 out of 10 uh is that they had to kill the project after two years of working
16:17 on it because it didn't work out it was not scalable they weren't able to keep
16:22 up with the technology because they were still focused on the agency part but they were not focused on the tool so if
16:29 you're thinking about like building a tool that actually is wider don't do it
16:35 internally you can go to SEO or any other it's not about like going to me
16:40 it's about actually choosing a tool that best fits your needs and your strategy if you're thinking about some small
16:46 automation so like some small tasks of course build it internally that's not a
16:52 problem use a GPT make.com like to build a workflow uh maybe in the future like
16:58 use some uh predefined agents right now I still haven't seen ones that are
17:04 good but definitely for some small task rewriting copy for LinkedIn posts uh uh
17:11 proofreading something use a GPT that is designed for specifically for this okay and are there
17:20 partnership opportunities perhaps with companies specializing AIdriven SEO with the agencies can take or still too early
17:27 for those uh what do you mean by that can you elaborate a I mean you said that if if people are not looking to bring uh
17:34 the u in-house AI team are there partnership they can do with other
17:40 AIdriven SEOs for sure you can like bringing consultants like helps you to uh
17:48 streamline the whole process but I believe that many of these things you can resolve on your own uh even by
17:55 talking like having a partnership with chat GPT so just like ask sometimes chat
18:00 GP PTN chat storm with it how to resolve a problem that you have like the biggest
18:05 problem that there is is to define the problem how to solve it is far easier
18:12 because you can actually talk like even with chat GPT as I said like you can use some drag and drop tools to resolve
18:18 certain problems as an example uh make.com there is hunch.tools tools i
18:25 think another one like but definitely start working with AI like so
18:32 go to chat GPT go to cloud start talking to it brainstorm on solutions that you
18:39 need but you need to define the problems that you want to solve
When and How to Use AI, GPTs, AI Workflows, and AI Agents: What tools should you implement for efficiency?
18:44 okay now uh all right now uh let's talk about when and how to use AI in GPTs and
18:52 AI workflows and of course the AI agents which are hot today so
18:57 uh now let's get into something technical but of course super valuable understanding when and how to use
19:03 different AI tools now in one of your post you talked about GPTs AI workflows and AI agents can you break down the
19:10 difference for our audience okay so basically what we see like with
19:15 the development of AI uh going back two years two years back uh we mostly got
19:23 most of the lowhanging fruit when it comes to the quality of the model AI
19:28 model that we can reach of course it will get better but to be honest it will not be such a significant increase in
19:35 the quality of the model that we saw throughout the last year since like there's no simple options to increase
19:43 the quality of the model we have other ways we found like uh other ways of how
19:49 to use the model to make it more efficient and let's start with a simple
19:54 chat so this is the first thing that we saw two years back over two years back when chat GPT introduced the chat online
20:02 we were able just to talk to it like back and forth like we asked a question like it retrieved an answer this is the
20:08 simplest way of using that uh AI and then there were GPTs so GPTs if you
20:14 don't know them like you can go to that GPT probably you can also like define them somewhere else it's a specialized
20:21 AI chat that is designed to resolve a specific problem so what you can do in a
20:26 GPD you can define the instructions you can upload specific data that AI needs
20:32 to look at uh chat GPD will turn them into something that is called rag
20:37 database so retrieval or augmented generation will pull data from the documents that you provide there in the
20:43 GPD and it will resolve certain problems gpts are really amazing for solving
20:48 specific tasks so short task that you need to resolve so for example writing a
20:53 LinkedIn post like with your tone of voice based on the data that you provide it's really really good because you can
21:00 create it to def like to resolve a specific task you can also use there's like a whole store of GPTs that you can
21:07 use proofreading um a lot of them really go to the store check them out but still
21:14 simple task like so one-stop solution you go there you ask a question you get
21:19 an answer based on the data that you provide okay then there is AI workflows
21:24 AI workflows is anything that you have as a human the defined process let's go
21:30 with content writing so If I as a human not using any external AI tools write an
21:38 article I go to Google uh check what's like what are the search results in
21:43 Google then I go to hrefs or semrush to pull SEO data then I create an outline
21:50 then I write a initial draft then I add internal links then I add external links maybe I source additional data but this
21:57 is a defined process one by one so like it's really defined uh straightaway process and it's a
22:05 workflow right just so you know when you go to LinkedIn and you see like look at
22:11 my new AI agent 99% of these are not agents they are workflows because it's
22:20 done with make.com or hunch.tools or n8 or there's like different ones but
22:28 still their process if you see that they branch out so like they go both ways that's just an if statement so if you
22:35 see this then do that if you don't see this then do that it's still a process
22:41 linear process that is has a defined pathway and now what we're getting into
22:46 which is amazing part is AI agents ai agents are
22:53 uh agents that answer nonlinear question not obvious without a beaten track
23:01 without a process that is predefined so it's kind of like a tool that answers an
23:07 open-ended question and defines the process along the way which is really
23:13 amazing because uh for example what we have a we have like two AI agents at the
23:20 moment in the in our system one is AI research engine so basically what we do
23:25 with AI research agent we provide it with all the data about the brief and we
23:31 ask it like look AI agent here's the brief that I want to write about like
23:36 this is the article the definition of my article can you help me with the research please tell me what kind of
23:43 data and where should I look for it to add the data into my outline so I
23:50 can provide it later on into as context to AI so then the agent does like okay I'm looking at the article I need case
23:57 studies here I need uh some statistics here I need to go to our page uh to
24:03 source some case studies from our page I need to uh take some thought leadership
24:09 insights from here from there and it makes decisions and then it executes
24:14 each one of them to provide the information but it's not that I'm telling the AI agent like look for every
24:21 article I need find case studies about this heading right something about this
24:27 it's not a process like it defines the path and things that it has to do on its
24:32 own and then it executes them so so it's like the uh it's like it's
24:38 like there is no workflow for example every time it has to find case studies find thought leader it just based on the
24:44 topic it decides itself so it has a list of tools that it can use so kind of like
24:51 what kind of things it can use but it's up to him what he will do in this
24:57 specific uh task like so we always define like all the tools that like uh
25:03 an agent can use and what are the options but it's like a list of options that it can go with but it defines them
25:10 on their own will those be just an example uh with this one it's basically just
25:17 going like so what are the types of data that it can go after uh what are the
25:22 places that it can go out like where it can use like what are the APIs that it can use and so on and so forth so
25:29 basically based on that it decides which tools are needed to proceed with the
25:34 task that is given but it's an open-ended question there is no predefined way of doing that can you do
25:41 it with an workflow probably uh and this is the
25:46 biggest change because like if I would do it with a workflow I would have to state a lot of ifs so if you see this
25:53 then do that if you see this then do that and the magical thing that is happening right now with uh AI is that
26:02 I'm as a developer I have to get rid of the idea the if state of my mind uh when
26:10 working with uh AI so like instead of thinking what are the edge cases of the
26:16 task that can we can do this is the way that I was programming for the last 10 years uh I have to think about the end
26:23 goal and forget about like all the ifs because I don't need to ai will figure out how to get to my goal what kind of
26:32 things it has to do like to resolve the problem but still AI agents are extremely early the most uh known AI
26:40 agent would be 01 uh model because they're having the reasoning uh part
26:46 like so they're defining what's to what to do first and then they're executing the whole thing so this is one of the
26:53 things perplexity has a bought for buying for you so like it in United
26:58 States it can actually purchase items for you based on your uh definition so
27:05 AI agents are more often like you can see them more and more but still most of
27:11 the cases that people on LinkedIn say that they built an AI agent not an AI it's a workflow
27:20 so do do you have AI agents with SEO win or another place we do but like we have
27:26 we use them in a combination of workflow and AI agent because it's not that like just just to tell you mine AI agents are
27:33 not a solution for everything because like when you're working with an AI agent you lose a little bit of
27:39 predictability if you have like a really good process for something use the process it makes no sense to have uh to
27:47 lose predictability because if you go through the process you know that AI will go and not only AI but also
27:54 basically you will execute specific tasks based on the process that you used as a human the big problem that I'm
28:00 seeing with humans using AI is that they use the process that worked for them
28:06 like for two three years then they go to AI and they forget about the process like don't do that like makes no sense
28:15 so just for example you mentioned how people do when they are writing content they go to SER they go to uh HRFS they
28:22 go to SCMS this is exactly the process which they should be replicating with a workflow right I would start with that
28:28 because that makes sense so if you're sourcing for keywords like uh don't all
28:33 of a sudden go to AI and tell it like find me keywords for this and that it doesn't know that it it doesn't have the
28:40 knowledge but the biggest problem with AI there is like one huge problem with AI it always answers okay yes that is
28:49 this is a big problem uh because if it knows or if it doesn't it will answer
28:55 yeah it's it's not like Google where it says result.found it's always supposed to answer no even thank you it will
29:02 still continue talking i know exactly because like this is how it's built and if it would not answer we wouldn't like
29:08 it so uh it's really brilliant but it doesn't have access to everything and it
29:15 will often not tell you that it didn't do something uh along the way so for me
29:21 when I'm using AI most important part is predictability because if I'm writing articles like for multiple customers
29:28 like our system does it uh I need to know what it did like it's not a black
29:34 box for me i need to know what kind of data it had to make decisions where it
29:40 took it from and how it was done okay okay now let's talk about a
How to Use AI for Local SEO: What are the benefits of AI in local search optimization?
29:46 specific example use case uh let's say I'm running an agency focused on local SEO now how could I leverage a GPT
29:53 specifically for the that purpose and what about an a AI workflow can you give
29:59 some tangible examples if possible i'm not an expert in local uh SEO but uh if
30:06 I would do that I would probably look into all the tasks that I'm doing normally and I would try to figure out
30:12 what uh should we do so like if it's connected to updating meta titles or
30:20 meta description I would definitely create a GPT that will help me out uh I
30:25 would probably scrape some kind of data to be able to figure out and cluster all
30:32 my content on my website uh because AI is amazing with understanding the
30:38 language because of that it's amazing with clustering things uh that are
30:43 really difficult for humans to cluster so if I source any data to be able to
30:48 perform my task I would uh use scraping for that so not AI and then I would
30:55 provide AI with the data that it needs to draw conclusions from specific things
31:01 ai is really bad with scraping so whenever you're putting into chat GPT a
31:06 link don't expect it to scrape it uh sometimes it will sometimes it will not
31:13 and you don't have you don't know what it did uh scraping is something that can
31:20 be done programmatically using Python uh using uh framework called beautiful
31:25 zoop it's not AI and it's amazing with scraping data uh AI is really bad with
31:33 that so like I wouldn't use for example AI pure AI for optimizing page right
31:39 because AI will not go to the code with will not be able to extract the code
31:44 that you're looking after like uh so be careful what you ask for
How AI Always Answers: How does AI handle user queries effectively?
31:51 because AI will answer again going back AI always answers but sometimes you
31:56 don't know based on what sometimes like AI will if you provide a URL like for example and ask a question optimize me
32:03 this page for this and that or tell me like how to optimize this page it will
32:09 look at the slag and based on the slag of the URL it will tell you how to
32:14 optimize the page never going there so even when you are using web pilot or
32:20 those type of GPTs it doesn't go there uh I'm not saying never uh the question
32:25 is what you're using so if you're using a tool I don't know how web pilot works if you're using a tool that combines web
32:33 scraping with AI so web scraping is a nonAI feature ai is just like drawing
32:40 conclusions then definitely it probably will work but if you will go just to chat GPD and even implement some
32:48 additional features for web scraping and you will give like five URLs I can bet that it will not go to
32:55 all the five pages just so you know scraping a page takes I would say 10 to
33:02 15 seconds one single page one request to scrape five pages it would take 50
33:08 seconds so if it answers within 10 seconds do you think it went to the page
33:14 or no no right right right okay we don't like
33:19 to wait yeah we are getting restless these days with the advancement of technology we
33:25 all are losing patience that's true now what's your favorite AI workflow for
33:32 SEO and content creation i would say the workflow that we have uh
33:38 inside SEO win so like uh because we worked on that like a lot and it's the
33:44 it looks at multiple data sources but to be honest from top of my head I'm not
33:50 able to tell you because there's like first of all there's like a lot of ifs so if certain things are happening if we
33:56 see certain types of data we do certain things but to complete the whole article
34:03 we probably go over I don't know 20 30 points of data uh we and we use AI 30 40
34:12 times in the whole process so I'm not able to even tell you how it works from
34:18 top of my head even though that I designed I was like part of like every single step uh I worked directly with
34:26 somebody on that okay okay fair enough now
Biggest Mistakes Agencies Make When Integrating AI: What pitfalls should agencies avoid?
34:32 uh what do you think are some of the biggest mistakes agencies make when integrating AI into their processes i
34:38 know you spoke about few but are there any other bigger major mistakes you could highlight mhm sure thing uh great
34:47 question so basically when you're implementing uh any AI tools uh or any
34:54 AI methods like into the organization uh if you go to LinkedIn you will see like
35:00 my top best uh prompts for this or that uh to be honest uh if you will compile a
35:07 list of prompts to how to do certain things most of the time it will not work
35:13 because it's rather about understanding how prompting works and I would invest
35:18 in that like first inside the agency so everybody understands exactly what they're doing because uh if you just
35:25 share prompts with your team first of all most of them will not follow them uh we'll have some
35:32 additional thought like how to make them better and it will probably not work out uh but like help people to understand
35:40 that they can talk and try to do things with AI uh and let them break things let
35:48 them make mistakes because learning part of learning is basically making mistakes
35:54 so and I would give uh people a lot more responsibilities
36:02 connected with how to implement uh the AI like on their own and try things out
36:08 inside the company uh I think the biggest problem of agencies is not
36:13 giving confidence to the team that they can do it they are able to do it and
36:20 understanding that implementing AI will not mean that they will get fired uh
36:25 because a lot of employees are really really afraid and that's normal that's okay uh that they will get fired if they
36:32 will implement the AI and in my opinion as I mentioned already it's quite the opposite they will get fired if they
36:39 will not do it this is the new reality that we're in if you don't learn it like
36:45 how to use it you will have problems If not now you will have problems in two years and start thinking about small
36:52 steps don't implement everything at once it will not work out small iterations
36:58 small improvements step by step and if you can find tools that already resolve
37:04 certain problems for you so as an example again I would take give my
37:09 company as an example use a tool like SEO win for content creation which will
37:16 let you not think about how to go through the process and how to prompt
37:21 the articles like how to get the articles what kind of prompts do I need to do because it will do everything on
37:27 the back end for your perspect from your perspective it's just drag and drop so there's plenty of tools in different
37:33 niches that you can use that they use AI underneath but it's not about prompting
37:40 it's about thinking and resolving a problem so like compiling all the things together to resolve a specific problem
37:47 and this is in my opinion the future of AI most AI tools that we will use we
37:54 will not even know that we're using AI like underneath like we will just like use them and they will resolve the
38:00 problem the prompting part will not be that needed because they will do it for
38:06 us okay now too many agencies are looking to
Importance of White Label SEO Solutions: Why should agencies consider white label services?
38:12 offer uh more comprehensive AI powered services but may not have the resources
38:17 to build everything from scratch now in such a scenario how critical do you think it is to rely on a white label SEO
38:24 solutions to provide a holistic suite of offering to the client uh definitely in my opinion if I would
38:31 be an agency I would start with buying tools like buying in sense of like
38:37 buying purchasing subscriptions uh and trying things out instead of
38:43 creating a tool internally as I mentioned if it's a small thing that you need to do like reporting I would
38:49 probably use make.com like just to simplify reporting but when it comes to any bigger things don't I would strongly
38:58 recommend not to build a tool not because I'm like selfish and I want to
39:04 be the only tool in the niche like I really don't care uh but it's really
39:10 most cases waste of your time and money uh because it's really extremely hard to
39:16 keep it up to date to uh improve the tool and it will cost you a lot more
39:22 than a simple subscription my way of thinking like in SEO wind when I'm like
39:27 building a new feature the first thing that I'm looking at is do I have an API
39:33 that can resolve my problem so basically can I buy a specific functionality from
39:38 somebody else to resolve my problem and I do it before I start to work on
39:44 specific thing because if I would think about resolving every single thing on my side on my own it would take me far more
39:51 time i would uh in like work a lot slower and probably I would bring in
39:58 like a lot worse result okay just not worth it okay
40:04 now uh great insight Storm it seems like understanding the new essence of AI
40:09 tools can really help agencies stay ahead of the curve uh speaking of uh
How to Make AI Content More Engaging: What techniques enhance AI-generated content?
40:15 staying ahead of the curve uh let's shift gears slightly and talk about another cornerstone of success SEO and
40:21 content now uh AI generated content is often criticized for being robotic how
40:28 can agencies make AI content more engaging and humanlike do they all the time need to use a tool or is there
40:35 another way out so from my perspective when I'm looking at the content uh definitely not using
40:43 AI as a search engine just purely write me this write me that uh because this
40:49 creates robotic fluff content that from Google's perspective it's removed from
40:55 the internet because that's a definition of spam it's empty it's fluff and it doesn't bring the value to the end user
41:01 so like if you're looking and hearing about any pages that were de-indexed
41:07 from the internet or like they got hit because they were writing with AI this
41:12 is how they were using AI like they were asking simple question using AI as a search engine not adding value not
41:19 adding opinions and so so forth i'm a strong believer in something that I call cyborg method a human working with AI
41:27 together not against each other uh but actually together because together we
41:32 can do much more we as humans we have superpowers and AI has its own superpowers but we have to realize what
41:39 are our superpowers so definitely when you're looking at AI AI can research far faster it can read a lot more like a lot
41:47 more simpler faster it can write extremely fast based on the context so
41:53 our superpowers is our experience we talked to the product team we talked to the marketing team we have our opinions
42:00 uh we talked to the customers for the last three years we saw so many cases
42:06 that AI never saw we can add all this influence into the article like or into
42:11 the content that we provide if you don't think that you have something to
42:17 add nobody will employ a button pusher like you are a special matter expert in
42:24 the field so add values on top like so this brings additional value to the content itself like so what we're doing
42:30 for example in SEO win we have like a toggle that enables you to give your
42:36 feedback to the article what we're doing next we're parsing this uh input that you gave as a special matter expert on a
42:43 specific topic we're dividing it into separate parts then we're looking at the outline and the context of the outline
42:49 to put this these spots into the context of AI so when we're writing the article
42:55 we say to AI look for this specific heading our special matter expert said that this and this is important so add
43:03 value on top like give context as much context as possible but don't overwhelm
43:09 AI at the same time it's not that like you will have a book of 300 pages give it to AI and say like do something with
43:16 that it will not work out you have to give context to the things that you're
43:21 doing and this is the bigger problem because context means for most cases
43:27 when you're using just chat form of AI it takes a lot of time to do it
Importance of Long Form vs Short Form Content in Today's SEO Landscape: Which format performs better in SEO?
43:34 okay now uh how important is long form versus short form content in today's SEO
43:40 landscape and does AI changes the dynamics here what I can see from the data point that
43:46 the articles got longer at least two times uh for the last five years like so
43:52 we can see that they were like a lot longer so short content that you can use
43:58 in social media use it in social media this is where you probably should uh add
44:03 it when it comes to long form like I can see that most top 10 search results are
44:09 rather longer than they were like five to eight years ago uh so a lot of people
44:15 say like okay but people don't read the whole long blog post yes but what Google
44:21 does like they go to a specific place in the blog post and show you where your
44:26 answer where you can find the answer they even highlight a place where they show like here's the answer to your
44:32 question but then if you like the answer to your question what you do you screen through the blog and you read the part
44:39 of the blog that uh is written the article that is written to read more on
44:44 the topic so this is how it works why this is why long form in sense of SEO
44:51 works better okay now uh when it come to
What It Takes to Rank #1 on Google: What are the key factors for achieving top rankings?
44:56 ranking number one on Google what do you what are three things agencies should prioritize in 2025
45:03 uh so I can definitely tell you two big things uh let's see like if I can say
45:10 it's free uh first of all like figure out like how to increase the quality of the content because Google is all about
45:16 the value uh of the content but if you only increase the quality of the content
45:23 so actually your uh articles bring in more value it doesn't mean that it will
45:28 solve your problems because on the other side there is authority authority means that you have to build trust like in
45:37 eyes of Google that your page will answer specific question and when it
45:42 comes to authority I will very simplify it like extremely probably most SEOs would kill me for that but authority for
45:49 me means that other people site you so like you get actually the back links that uh probably most of SEOs said that
45:56 they're dead and Google definitely told that like we don't look at backlinks anymore uh and you have to see like what
46:02 are the citations at any social media so you have to be actually well known the other thing of the authority I always
46:10 say like be the Wikipedia of your niche but actually niche that means because
46:16 Google what they will do they will let you in in small pools of content and
46:22 they will do it gradually they will let you bigger pools and then bigger pools
46:27 but you have to start from a really specific niche don't expand your content
46:33 all over the place so if you're in the marketing niche it doesn't mean that you can write about anything in marketing
46:39 just look at your Google search console look at the data where Google is letting you in a little bit more and slowly
46:45 expand if one day I hope you will you will become hotspot of this world then
46:51 you will be able to write about the whole marketing world but until then focus on specific where do you
46:57 specialize niche out and like really focus on that uh where you can add the
47:03 most value and Google will help you out to build your authority within this niche and then slowly gradually expand
47:09 add topics on top of that uh so this is my recommendations when it comes to SEO
47:17 hubspot might not be the right example right and right now you know how they lost it
47:23 but like look at this this from different perspectives like what have
47:28 like they lost on blog.habspot.com when you look at their main domain so hopspot.com domain they
47:35 didn't lose traffic so like the main the main source of their money is still
47:40 there so like ownish keyword they were ranking i don't even think they were generating any value for them they were
47:47 just generating so traffic this is like the definition what they did of time
47:53 like so they were all over the place so uh again looking at the main domain
47:59 hopspot.com it didn't lose that much like so you can look at all the big ones
48:04 like they're all over the place they're going up and down because of the changes and all of them they're trying to use an
48:11 SEO to extend if you are a big company like they are you will try to bend their
48:19 so and that's okay like uh this is what we do at SE at as SEOs i always try to
48:26 see like what is the extent that Google will let me go and not slap me on the
48:32 hands so this is how I am i like to test out i do a lot of tests uh not all of
48:38 them are good uh but I need to know what works and what doesn't
48:45 right okay now um let's talk a little bit about
Advice for SEO Consultants and Agency Owners: What insights can help you succeed in SEO?
48:52 business most of our listeners are either SEO consultants and agency owners who wants to grow their business now of
48:59 course uh white label SEO is a massive opportunity for them what's your advice for agencies looking to pitch their
49:04 services effectively so I've had a webinar with really a
49:10 brilliant person uh that was talking about his agency and he mentioned like
49:17 he has an agency of 60 people in Dubai and he mentioned one thing to me that
49:23 often white labeling for other agencies makes a lot of sense because you have
49:28 one customer you don't have to run around look for additional customers if you deliver good quality content for
49:35 good quality services it's often worth to work on a lower price but for one
49:42 customer or like two customers if you partner up like with agencies and you white label your services for other
49:48 agencies you don't have to talk to the customers you don't deal with the customers like you don't have to look
49:54 for the customers so there's so many costs that like go off the table uh the
50:00 whole sales process like uh becomes really easy but you have to know that
50:06 you're delivering uh the value so like there are two ways either you will work directly with the customers and you will
50:12 like look for the customers and then look to retain the customers because retention is the biggest problem always
50:19 uh but like it's your choice i think both ways are good if I would be
50:25 building an agency like I would think about working with other agencies to
50:31 white label uh my services like I think it's really worth to
50:36 diversify and use this option right right and how does AI fit into white
50:41 label SEO can agencies use it to increase margins while maintaining quality definitely it's all about like
50:47 talking to your customers whoever the customer is is it another agency that sells for somebody else uh it's worth it
50:55 okay how but talk to them right now how do you think uh agencies should
51:01 differentiate themselves in the competitive SEO market in today's world
51:06 definitely giving more value at the end over overd deliver every single time and
51:12 if you just want to use AI to create posts because uh you just have a kind of
51:19 a headcount of articles that you can have to prepare every single month and they will not work for the customer they
51:26 will never bring in value organic traffic this is a this is a short way
51:31 like a quick way to lose a customer so always think about how you can help your
51:39 customer to get the results that they want so the first thing that I would ask anybody that I work with what is the KPI
51:46 that you're looking at what do you want to achieve so if I know that I can help
51:52 them out with that and to be honest like just bringing in 10 articles per month this is not KPI
52:00 right uh all right now Tom this has been an absolute master class in AI and of
52:07 course SEO strategies now before we wrap up let's do some rapid fire questions
Rapid Fire Round: What quick tips can you share for SEO success?
52:13 okay all right now you just answer what comes in the top of your mind and uh one
52:20 AI tools you can't live without and it is not SEO cloth
52:26 cloth okay uh biggest SEO myth that needs to die
52:34 uh that's ah that's a hard one
52:40 uh that AI knows everything okay and one book or podcast you recommend for agency
52:47 owners uh agency owners
52:53 uh to be honest like I don't know like I don't have uh a straight answer uh for
53:01 that one I I really I I don't have anything to recommend
53:06 definitely uh listen to Nami like the the agency
53:12 insider show like this is the one that I would really really recommend uh what I would recommend here definitely go to
53:18 LinkedIn find some people like on LinkedIn that share their knowledge uh look at what they're doing nid is
53:25 definitely one of these people thank you very much for your uh uh
53:30 praise and uh um finally where can our listeners find you and learn more about
53:36 SU so basically you can find me on LinkedIn so look for Tom Winter on
53:41 LinkedIn you will definitely find me there like I live there uh it's my next
53:47 house uh basically so I love to talk to people and if you want to just uh ping
53:54 me and talk shoot like I don't mind and I really like to talk to people
Where to Find Tom Winter Online: How can you connect with Tom Winter?
54:01 okay all right uh thank you once again Tom for joining us on the Agency Insider
54:06 Show and to all our listeners don't forget to check out Tom's work on LinkedIn and subscribe to his updates
54:12 and if you are interested in exploring white label SEO services to elevate your agency offerings feel free to reach out
54:18 to us at page traffic see you next time and once again Tom thank you for giving me your valuable time thanks a lot Anit