WEBVTT
00:00:01.280 --> 00:00:12.880
Hi everybody, I want to talk to you today about a really important piece that Paris Childress of GEO Forge and I talked about on a live on LinkedIn a couple of days ago.
00:00:13.279 --> 00:00:26.480
We surmised, or we had a bit of an argument around is social media engagement leading to citation in any of the LLMs?
00:00:26.800 --> 00:00:38.159
Now I've said for a long time that if uh social media posts are engaged with heavily, that should then lead to rankings in Google.
00:00:38.399 --> 00:01:08.879
Um, Google would never admit to this, um, and a lot of SEOs would push heavily back on the fact that um that doesn't seem to necessarily be the case, but we are now entering a world as we know where prompts uh rather than keywords and key phrases and questions are important, hence me working on question dot marketing and AEO, geo and working across the different LLMs.
00:01:08.959 --> 00:01:19.200
Now, the findings in uh Paris' report here, which I've got on the screen, are incredible as far as I'm concerned.
00:01:19.359 --> 00:01:20.480
I'm a nerd, right?
00:01:20.560 --> 00:01:21.040
Why not?
00:01:21.120 --> 00:01:22.480
I like this sort of stuff.
00:01:22.719 --> 00:01:43.040
But he pulled all of the LinkedIn and uh YouTube citation data from GEO Forge, so they have over 10 million citations in their database across uh I think about 850 prompts that they're currently tracking for clients.
00:01:43.439 --> 00:01:46.239
Now, what is evident?
00:01:46.400 --> 00:01:50.640
And if we just you know, I would encourage you to read this article of his at length.
00:01:50.799 --> 00:01:54.239
Um, I'm not going to go through every single piece of it, I'll give you my take.
00:01:54.560 --> 00:02:05.040
But um, Paris runs the GEO show, it's published as a podcast, as um, I also talk to people about how you get your businesses listed in AI.
00:02:05.439 --> 00:02:23.520
One of the things that is very apparent is, uh and I've been talking about this for a while now, Google is losing its grip and trying to ring fence its own data, and one search engine does not mean the data set for all of the LLMs.
00:02:23.680 --> 00:02:49.840
That is evidenced by the fact that um 47% of the YouTube videos that are cited have uh less than a thousand views, so that's not a necessarily engagement, but what it is showing you that Google's AI references uh YouTube videos far far more uh than for example chat GPT.
00:02:50.000 --> 00:02:58.479
Chat GPT pulls a lot of its citation um from uh LinkedIn, for example.
00:02:58.719 --> 00:03:11.759
Now, if you look at 25,000, 25,000 chat GPT citations, um only one points to YouTube.
00:03:12.560 --> 00:03:17.759
So clearly, you know, we people are going crazy over does Reddit point to this?
00:03:17.919 --> 00:03:27.680
Which you know, which of which social property or which uh sites or which authority domains do you need to publish on to be cited?
00:03:27.919 --> 00:03:35.599
Clearly, making YouTube videos does not get you into chat GPT, whatever you do.
00:03:35.840 --> 00:03:46.719
Um we're we're also looking at uh the fact that if you're publishing on LinkedIn, that is working and getting you cited in Chat GPT.
00:03:46.800 --> 00:03:49.039
So articles, post articles.
00:03:49.280 --> 00:03:51.840
Now, what you know what does that say to us?
00:03:52.080 --> 00:04:19.920
Um I I thought that um you know if you take a post and it gets a lot of engagement, I would have made or I did make the assumption uh, or you know, I would have thought that if a post is getting a ton of engagement, we would imagine that it would go further in search, be cited more, and that doesn't appear to be the case.
00:04:20.079 --> 00:04:36.800
There does seem seem to be um a threshold, so you do need to break through certain levels, you can't be necessarily an absolutely brand new channel and get no engagement and expect to get a part a post cited, right?
00:04:37.040 --> 00:04:54.000
But um when you show that we got up to or once you got to past, for example, a thousand views on a video on YouTube, the citation rate didn't increase um you know dynamically, which you might have expected.
00:04:54.399 --> 00:05:17.680
Now GeoForged measured over 10 point million 10.2 million citation events across 336 unique cited your 336,000 uh unique cited your URLs, and that constituted 458,000 prompt runs over 848 prompts.
00:05:17.839 --> 00:05:23.199
This is a significant piece of data that I think people need to need to pay attention to.
00:05:23.279 --> 00:05:42.639
So all the top um people that are talking about AEO, geo and SEO, um, and SEO being part of these new disciplines need to pay attention to the fact that this data shows clearly how uh citations are driven.
00:05:42.720 --> 00:05:50.319
Um, and social platforms account for 8.8% of that corpus.
00:05:50.639 --> 00:05:58.639
So of that data that is out there in terms of citations, we're looking at nearly 10% of it, but coming just from social media.
00:05:58.879 --> 00:06:15.600
Um that is you know, that is something it it may well be a minority at the moment, um, but it is meaningful, and it's not something that um you know B2B brands in particular should be ignoring.
00:06:15.759 --> 00:06:32.879
So you do clearly need to be publishing across all social media, more so you need to be publishing unique knowledge, and Paris does talk um a lot about that needing to have knowledge that's new to the LLMs right now.
00:06:33.040 --> 00:06:42.319
This is the bit um that showed this part of the report which AI engines actually cite social media.
00:06:42.480 --> 00:06:48.879
Now, if you look at ChatGPT, Google and Google, they are not reading the same internet.
00:06:49.519 --> 00:06:55.360
Let me repeat that for you, and this this will be the title of this episode from me.
00:06:55.680 --> 00:07:00.959
Chat GPT and Google are not reading the same internet.
00:07:01.920 --> 00:07:03.759
Remember that, that is key.
00:07:03.920 --> 00:07:41.199
I've been talking about Google losing its grip, Google, you know, somebody called me, Am I on mushrooms or am I on crack by saying that maybe in years to come we don't have well, we will have Google in some respect, but it is definitely slipping because if you look at Chat GPT with over a million citations in this piece of data from GEO Forge, look at the percentages 0.004 for YouTube, 0.2% for LinkedIn, 0.09 for Reddit.
00:07:41.439 --> 00:07:45.680
Now I see everybody chatting about Reddit, Reddit must be this, Reddit must be that.
00:07:45.759 --> 00:07:48.720
Well, not for chat GPT clearly.
00:07:48.879 --> 00:07:52.480
Um Instagram 0, right?
00:07:52.959 --> 00:08:03.839
Um Google AI mode conversely, YouTube 6%, LinkedIn 1.5%, Reddit 1.8%, Instagram 0.72%.
00:08:04.160 --> 00:08:18.879
So a clear, huge differential between the way things are getting picked up, and then obviously go Google AI overviews, YouTube six percent, LinkedIn one and a half, Reddit 2.1, and Instagram 0.56.
00:08:19.120 --> 00:08:24.560
Now I would imagine, and again, maybe this is a massive assumption.
00:08:24.800 --> 00:08:39.200
Um YouTube videos do not necessarily um build as much content because people typically don't make enough video, right?
00:08:39.440 --> 00:09:04.559
I would imagine that should we be able to um and we'll come on to this later in this article, but should we be able to put a video in front of um Google that answers a very very specific topic, i.e., a prompt that we want to be listed for clearly in Google, which is not going to work in uh Chat GPT, but I'm gonna come on to another experiment in a bit.
00:09:04.720 --> 00:09:17.039
Um, if that video matches a prompt almost exactly and fills a knowledge gap, does it then become cited more easily?
00:09:17.279 --> 00:09:22.399
Does it does it get picked up almost like an exact match keyword phrase, right?
00:09:22.559 --> 00:09:24.000
Long tail keyword phrase.
00:09:24.080 --> 00:09:26.240
If you're an SEO, think of it that way.
00:09:26.399 --> 00:09:36.639
Can we get a prompt uh straight into Google AI mode or Google AI overviews using a video which matches that prompt and fills a knowledge gap?
00:09:36.799 --> 00:09:44.480
That's one of the questions that we're gonna ask and try and answer in this series of um performance that we're working on.
00:09:45.279 --> 00:09:57.840
So if we see that Chat GPT only cited 42 videos out of one over a million citations, it's that's not gonna work, right?
00:09:57.919 --> 00:10:09.759
We can see that that is basically Chat GPT reading its own index referencing something completely different to Google I Google AI mode or um Gemini, etc.
00:10:10.000 --> 00:10:10.320
Right.
00:10:10.559 --> 00:10:22.399
So we looked also at how big is the audience on these cited videos, and it showed that you know cited videos didn't have to have that many views.
00:10:22.559 --> 00:10:30.720
So if you look at um distribution over videos where there are under a hundred views, a hundred to a thousand, a thousand to ten thousand.
00:10:31.200 --> 00:10:52.639
Now the assumption was obviously if you've got a video that's getting hundreds of thousands of views, you may well uh presume that that would be cited, it's not anything um any greater, although um do we know whether you know we've referenced that many videos with over a hundred thousand views on them?
00:10:52.799 --> 00:10:53.840
Maybe, maybe not.
00:10:53.919 --> 00:11:01.120
There's probably not that many view videos that answer prompts that are over a hundred thousand views, so that would be another thing to test.
00:11:01.600 --> 00:11:08.159
Um LinkedIn, and if you're watching this on LinkedIn, different story, right?
00:11:08.559 --> 00:11:18.559
We've got um if you had in the same scenario, less engagement doesn't matter here, or it does to a little bit, right?
00:11:18.639 --> 00:11:36.799
And the story it's going to tell is 60% of the cited LinkedIn articles that are picked up on Chat GPT had less than 50 comments or uh likes or engagements on them, and 19 or 20% of them had less than 10.
00:11:36.879 --> 00:11:47.120
So maybe we could be thinking, right, as long as we can get 10-15 comments or engagements on a post, it shows the relevance, and then it maybe get picked up a little bit easier.
00:11:47.200 --> 00:11:49.039
So that's one thing to learn.
00:11:49.440 --> 00:11:56.559
Um so the answer is does engagement make a post more likely to be cited?
00:11:56.799 --> 00:12:01.519
And in reality, the answer is no, not really, not necessarily.
00:12:01.759 --> 00:12:10.720
Um does it matter if it matches the question or the prompt more closely and gives some um unique knowledge?
00:12:10.879 --> 00:12:12.320
Yes, I think it does.
00:12:12.639 --> 00:12:16.720
So there's there's tons more in this article.
00:12:16.960 --> 00:12:23.679
Um, particularly here, if we look at LinkedIn posts, they are they die after a couple of days.
00:12:23.759 --> 00:12:24.159
We know that.
00:12:24.240 --> 00:12:28.480
After 48 hours, typically a LinkedIn post is pretty much gone.
00:12:28.559 --> 00:12:39.759
Um, but we're finding that LinkedIn posts and articles that get picked up they're typically getting picked up 35 days or more after they were originally published.
00:12:39.840 --> 00:12:54.000
Now, whether that's a circle on the AI uh wheel that comes back round and spiders that content and checks that it's still there and has had its engagement and is still living, maybe that's a thing.
00:12:54.240 --> 00:13:04.639
Um the the findings that Paris put together, pick the engine that you um you want to list on.
00:13:04.799 --> 00:13:14.480
Now, my advice would be I want you to be able to be listed across all of them, of course I do, and that's you know, producing content and only publishing it in one way.
00:13:14.720 --> 00:13:24.799
So, for example, if you produced a YouTube video and you posted that on your website and you wrote a blog around it, etc., yeah, great.
00:13:24.960 --> 00:13:37.840
You're doing traditional SEO and it's working for for you on Google AI mode, maybe if you can get it uh picked up and spider and you've structured it correctly and it's got the right prompts in it, etc.
00:13:38.000 --> 00:13:38.399
etc.
00:13:39.120 --> 00:13:48.480
If you then want to then get that distributed and referenced and shared, um, it needs to go onto other platforms and you need content across all of them, right?
00:13:48.559 --> 00:13:53.440
And you need to work with, I would hope, an expert who knows how to do that, right?
00:13:53.840 --> 00:14:01.440
You don't have to necessarily get it published uh with tons of engagement, which is a great sign, right?
00:14:01.840 --> 00:14:11.440
Um it's also evident that you are better off publishing more and more content um rather than just trying to get winners.
00:14:11.519 --> 00:14:20.320
So the thought here is you're not trying to get an amazing post that gets loads of engagement because that will get you get you the reach and get you the citation.
00:14:20.480 --> 00:14:33.840
No, you're better off posting better content, more of it, filling in the gaps, um, and doing what, for example, Gary Vinerchuk says, and that is um getting a lot of content out there and looking for the winners.
00:14:33.919 --> 00:14:42.320
So this backs up what he's saying about publishing a lot of content and get get the engagement across as much of it as you can.
00:14:42.480 --> 00:14:53.679
Not you don't need to worry that if you're not getting tons of engagement, it's about publishing unique knowledge as much as you can across as many platforms as you can and distributing that.
00:14:54.159 --> 00:15:00.159
Um so we we're gonna run a test, uh Paris and I.
00:15:00.399 --> 00:15:13.919
We are gonna publish some very specific content absolutely tailored to some some prompts that we know uh don't have answers behind them at the moment or exact answers.
00:15:14.000 --> 00:15:26.720
So one of the things that fascinates me, I was I was looking at the knowledge gaps in a corpus and saying, right, okay, if certain models don't don't know the answers, um, how does it answer?
00:15:27.039 --> 00:15:32.080
And the answer is it will give you its best guess or its nearest approximation.
00:15:32.320 --> 00:15:37.840
Now, if the then exact answer appears, does it then get cited?
00:15:37.919 --> 00:15:39.279
And that's what we want to know next.
00:15:39.360 --> 00:15:41.279
So we're gonna publish some content.
00:15:41.519 --> 00:15:50.480
Um, we're going to then have that match a knowledge gap, we're going to publish it as widely as we can for clients.
00:15:50.639 --> 00:15:53.360
Um, one client in particular, Stitch AI.
00:15:53.519 --> 00:16:04.320
We're going to publish that as part of this study, and we will then report back to you how that does and where it appears, if it appears, and if it is cited.
00:16:04.399 --> 00:16:05.919
So that's the the next experiment.
00:16:06.000 --> 00:16:08.080
I'd love you to keep following me.
00:16:08.240 --> 00:16:11.200
Um, please do share and like and comment.
00:16:11.279 --> 00:16:12.639
I would love to hear from you.
00:16:12.799 --> 00:16:17.200
Anything that you're finding out as well, um, this is an open book, right?
00:16:17.360 --> 00:16:18.720
And we're all here to learn.
00:16:18.879 --> 00:16:22.879
So love to hear from any of the audience and speak to you again soon.
00:00:01.280 --> 00:00:12.880
Hi everybody, I want to talk to you today about a really important piece that Paris Childress of GEO Forge and I talked about on a live on LinkedIn a couple of days ago.
00:00:13.279 --> 00:00:26.480
We surmised, or we had a bit of an argument around is social media engagement leading to citation in any of the LLMs?
00:00:26.800 --> 00:00:38.159
Now I've said for a long time that if uh social media posts are engaged with heavily, that should then lead to rankings in Google.
00:00:38.399 --> 00:01:08.879
Um, Google would never admit to this, um, and a lot of SEOs would push heavily back on the fact that um that doesn't seem to necessarily be the case, but we are now entering a world as we know where prompts uh rather than keywords and key phrases and questions are important, hence me working on question dot marketing and AEO, geo and working across the different LLMs.
00:01:08.959 --> 00:01:19.200
Now, the findings in uh Paris' report here, which I've got on the screen, are incredible as far as I'm concerned.
00:01:19.359 --> 00:01:20.480
I'm a nerd, right?
00:01:20.560 --> 00:01:21.040
Why not?
00:01:21.120 --> 00:01:22.480
I like this sort of stuff.
00:01:22.719 --> 00:01:43.040
But he pulled all of the LinkedIn and uh YouTube citation data from GEO Forge, so they have over 10 million citations in their database across uh I think about 850 prompts that they're currently tracking for clients.
00:01:43.439 --> 00:01:46.239
Now, what is evident?
00:01:46.400 --> 00:01:50.640
And if we just you know, I would encourage you to read this article of his at length.
00:01:50.799 --> 00:01:54.239
Um, I'm not going to go through every single piece of it, I'll give you my take.
00:01:54.560 --> 00:02:05.040
But um, Paris runs the GEO show, it's published as a podcast, as um, I also talk to people about how you get your businesses listed in AI.
00:02:05.439 --> 00:02:23.520
One of the things that is very apparent is, uh and I've been talking about this for a while now, Google is losing its grip and trying to ring fence its own data, and one search engine does not mean the data set for all of the LLMs.
00:02:23.680 --> 00:02:49.840
That is evidenced by the fact that um 47% of the YouTube videos that are cited have uh less than a thousand views, so that's not a necessarily engagement, but what it is showing you that Google's AI references uh YouTube videos far far more uh than for example chat GPT.
00:02:50.000 --> 00:02:58.479
Chat GPT pulls a lot of its citation um from uh LinkedIn, for example.
00:02:58.719 --> 00:03:11.759
Now, if you look at 25,000, 25,000 chat GPT citations, um only one points to YouTube.
00:03:12.560 --> 00:03:17.759
So clearly, you know, we people are going crazy over does Reddit point to this?
00:03:17.919 --> 00:03:27.680
Which you know, which of which social property or which uh sites or which authority domains do you need to publish on to be cited?
00:03:27.919 --> 00:03:35.599
Clearly, making YouTube videos does not get you into chat GPT, whatever you do.
00:03:35.840 --> 00:03:46.719
Um we're we're also looking at uh the fact that if you're publishing on LinkedIn, that is working and getting you cited in Chat GPT.
00:03:46.800 --> 00:03:49.039
So articles, post articles.
00:03:49.280 --> 00:03:51.840
Now, what you know what does that say to us?
00:03:52.080 --> 00:04:19.920
Um I I thought that um you know if you take a post and it gets a lot of engagement, I would have made or I did make the assumption uh, or you know, I would have thought that if a post is getting a ton of engagement, we would imagine that it would go further in search, be cited more, and that doesn't appear to be the case.
00:04:20.079 --> 00:04:36.800
There does seem seem to be um a threshold, so you do need to break through certain levels, you can't be necessarily an absolutely brand new channel and get no engagement and expect to get a part a post cited, right?
00:04:37.040 --> 00:04:54.000
But um when you show that we got up to or once you got to past, for example, a thousand views on a video on YouTube, the citation rate didn't increase um you know dynamically, which you might have expected.
00:04:54.399 --> 00:05:17.680
Now GeoForged measured over 10 point million 10.2 million citation events across 336 unique cited your 336,000 uh unique cited your URLs, and that constituted 458,000 prompt runs over 848 prompts.
00:05:17.839 --> 00:05:23.199
This is a significant piece of data that I think people need to need to pay attention to.
00:05:23.279 --> 00:05:42.639
So all the top um people that are talking about AEO, geo and SEO, um, and SEO being part of these new disciplines need to pay attention to the fact that this data shows clearly how uh citations are driven.
00:05:42.720 --> 00:05:50.319
Um, and social platforms account for 8.8% of that corpus.
00:05:50.639 --> 00:05:58.639
So of that data that is out there in terms of citations, we're looking at nearly 10% of it, but coming just from social media.
00:05:58.879 --> 00:06:15.600
Um that is you know, that is something it it may well be a minority at the moment, um, but it is meaningful, and it's not something that um you know B2B brands in particular should be ignoring.
00:06:15.759 --> 00:06:32.879
So you do clearly need to be publishing across all social media, more so you need to be publishing unique knowledge, and Paris does talk um a lot about that needing to have knowledge that's new to the LLMs right now.
00:06:33.040 --> 00:06:42.319
This is the bit um that showed this part of the report which AI engines actually cite social media.
00:06:42.480 --> 00:06:48.879
Now, if you look at ChatGPT, Google and Google, they are not reading the same internet.
00:06:49.519 --> 00:06:55.360
Let me repeat that for you, and this this will be the title of this episode from me.
00:06:55.680 --> 00:07:00.959
Chat GPT and Google are not reading the same internet.
00:07:01.920 --> 00:07:03.759
Remember that, that is key.
00:07:03.920 --> 00:07:41.199
I've been talking about Google losing its grip, Google, you know, somebody called me, Am I on mushrooms or am I on crack by saying that maybe in years to come we don't have well, we will have Google in some respect, but it is definitely slipping because if you look at Chat GPT with over a million citations in this piece of data from GEO Forge, look at the percentages 0.004 for YouTube, 0.2% for LinkedIn, 0.09 for Reddit.
00:07:41.439 --> 00:07:45.680
Now I see everybody chatting about Reddit, Reddit must be this, Reddit must be that.
00:07:45.759 --> 00:07:48.720
Well, not for chat GPT clearly.
00:07:48.879 --> 00:07:52.480
Um Instagram 0, right?
00:07:52.959 --> 00:08:03.839
Um Google AI mode conversely, YouTube 6%, LinkedIn 1.5%, Reddit 1.8%, Instagram 0.72%.
00:08:04.160 --> 00:08:18.879
So a clear, huge differential between the way things are getting picked up, and then obviously go Google AI overviews, YouTube six percent, LinkedIn one and a half, Reddit 2.1, and Instagram 0.56.
00:08:19.120 --> 00:08:24.560
Now I would imagine, and again, maybe this is a massive assumption.
00:08:24.800 --> 00:08:39.200
Um YouTube videos do not necessarily um build as much content because people typically don't make enough video, right?
00:08:39.440 --> 00:09:04.559
I would imagine that should we be able to um and we'll come on to this later in this article, but should we be able to put a video in front of um Google that answers a very very specific topic, i.e., a prompt that we want to be listed for clearly in Google, which is not going to work in uh Chat GPT, but I'm gonna come on to another experiment in a bit.
00:09:04.720 --> 00:09:17.039
Um, if that video matches a prompt almost exactly and fills a knowledge gap, does it then become cited more easily?
00:09:17.279 --> 00:09:22.399
Does it does it get picked up almost like an exact match keyword phrase, right?
00:09:22.559 --> 00:09:24.000
Long tail keyword phrase.
00:09:24.080 --> 00:09:26.240
If you're an SEO, think of it that way.
00:09:26.399 --> 00:09:36.639
Can we get a prompt uh straight into Google AI mode or Google AI overviews using a video which matches that prompt and fills a knowledge gap?
00:09:36.799 --> 00:09:44.480
That's one of the questions that we're gonna ask and try and answer in this series of um performance that we're working on.
00:09:45.279 --> 00:09:57.840
So if we see that Chat GPT only cited 42 videos out of one over a million citations, it's that's not gonna work, right?
00:09:57.919 --> 00:10:09.759
We can see that that is basically Chat GPT reading its own index referencing something completely different to Google I Google AI mode or um Gemini, etc.
00:10:10.000 --> 00:10:10.320
Right.
00:10:10.559 --> 00:10:22.399
So we looked also at how big is the audience on these cited videos, and it showed that you know cited videos didn't have to have that many views.
00:10:22.559 --> 00:10:30.720
So if you look at um distribution over videos where there are under a hundred views, a hundred to a thousand, a thousand to ten thousand.
00:10:31.200 --> 00:10:52.639
Now the assumption was obviously if you've got a video that's getting hundreds of thousands of views, you may well uh presume that that would be cited, it's not anything um any greater, although um do we know whether you know we've referenced that many videos with over a hundred thousand views on them?
00:10:52.799 --> 00:10:53.840
Maybe, maybe not.
00:10:53.919 --> 00:11:01.120
There's probably not that many view videos that answer prompts that are over a hundred thousand views, so that would be another thing to test.
00:11:01.600 --> 00:11:08.159
Um LinkedIn, and if you're watching this on LinkedIn, different story, right?
00:11:08.559 --> 00:11:18.559
We've got um if you had in the same scenario, less engagement doesn't matter here, or it does to a little bit, right?
00:11:18.639 --> 00:11:36.799
And the story it's going to tell is 60% of the cited LinkedIn articles that are picked up on Chat GPT had less than 50 comments or uh likes or engagements on them, and 19 or 20% of them had less than 10.
00:11:36.879 --> 00:11:47.120
So maybe we could be thinking, right, as long as we can get 10-15 comments or engagements on a post, it shows the relevance, and then it maybe get picked up a little bit easier.
00:11:47.200 --> 00:11:49.039
So that's one thing to learn.
00:11:49.440 --> 00:11:56.559
Um so the answer is does engagement make a post more likely to be cited?
00:11:56.799 --> 00:12:01.519
And in reality, the answer is no, not really, not necessarily.
00:12:01.759 --> 00:12:10.720
Um does it matter if it matches the question or the prompt more closely and gives some um unique knowledge?
00:12:10.879 --> 00:12:12.320
Yes, I think it does.
00:12:12.639 --> 00:12:16.720
So there's there's tons more in this article.
00:12:16.960 --> 00:12:23.679
Um, particularly here, if we look at LinkedIn posts, they are they die after a couple of days.
00:12:23.759 --> 00:12:24.159
We know that.
00:12:24.240 --> 00:12:28.480
After 48 hours, typically a LinkedIn post is pretty much gone.
00:12:28.559 --> 00:12:39.759
Um, but we're finding that LinkedIn posts and articles that get picked up they're typically getting picked up 35 days or more after they were originally published.
00:12:39.840 --> 00:12:54.000
Now, whether that's a circle on the AI uh wheel that comes back round and spiders that content and checks that it's still there and has had its engagement and is still living, maybe that's a thing.
00:12:54.240 --> 00:13:04.639
Um the the findings that Paris put together, pick the engine that you um you want to list on.
00:13:04.799 --> 00:13:14.480
Now, my advice would be I want you to be able to be listed across all of them, of course I do, and that's you know, producing content and only publishing it in one way.
00:13:14.720 --> 00:13:24.799
So, for example, if you produced a YouTube video and you posted that on your website and you wrote a blog around it, etc., yeah, great.
00:13:24.960 --> 00:13:37.840
You're doing traditional SEO and it's working for for you on Google AI mode, maybe if you can get it uh picked up and spider and you've structured it correctly and it's got the right prompts in it, etc.
00:13:38.000 --> 00:13:38.399
etc.
00:13:39.120 --> 00:13:48.480
If you then want to then get that distributed and referenced and shared, um, it needs to go onto other platforms and you need content across all of them, right?
00:13:48.559 --> 00:13:53.440
And you need to work with, I would hope, an expert who knows how to do that, right?
00:13:53.840 --> 00:14:01.440
You don't have to necessarily get it published uh with tons of engagement, which is a great sign, right?
00:14:01.840 --> 00:14:11.440
Um it's also evident that you are better off publishing more and more content um rather than just trying to get winners.
00:14:11.519 --> 00:14:20.320
So the thought here is you're not trying to get an amazing post that gets loads of engagement because that will get you get you the reach and get you the citation.
00:14:20.480 --> 00:14:33.840
No, you're better off posting better content, more of it, filling in the gaps, um, and doing what, for example, Gary Vinerchuk says, and that is um getting a lot of content out there and looking for the winners.
00:14:33.919 --> 00:14:42.320
So this backs up what he's saying about publishing a lot of content and get get the engagement across as much of it as you can.
00:14:42.480 --> 00:14:53.679
Not you don't need to worry that if you're not getting tons of engagement, it's about publishing unique knowledge as much as you can across as many platforms as you can and distributing that.
00:14:54.159 --> 00:15:00.159
Um so we we're gonna run a test, uh Paris and I.
00:15:00.399 --> 00:15:13.919
We are gonna publish some very specific content absolutely tailored to some some prompts that we know uh don't have answers behind them at the moment or exact answers.
00:15:14.000 --> 00:15:26.720
So one of the things that fascinates me, I was I was looking at the knowledge gaps in a corpus and saying, right, okay, if certain models don't don't know the answers, um, how does it answer?
00:15:27.039 --> 00:15:32.080
And the answer is it will give you its best guess or its nearest approximation.
00:15:32.320 --> 00:15:37.840
Now, if the then exact answer appears, does it then get cited?
00:15:37.919 --> 00:15:39.279
And that's what we want to know next.
00:15:39.360 --> 00:15:41.279
So we're gonna publish some content.
00:15:41.519 --> 00:15:50.480
Um, we're going to then have that match a knowledge gap, we're going to publish it as widely as we can for clients.
00:15:50.639 --> 00:15:53.360
Um, one client in particular, Stitch AI.
00:15:53.519 --> 00:16:04.320
We're going to publish that as part of this study, and we will then report back to you how that does and where it appears, if it appears, and if it is cited.
00:16:04.399 --> 00:16:05.919
So that's the the next experiment.
00:16:06.000 --> 00:16:08.080
I'd love you to keep following me.
00:16:08.240 --> 00:16:11.200
Um, please do share and like and comment.
00:16:11.279 --> 00:16:12.639
I would love to hear from you.
00:16:12.799 --> 00:16:17.200
Anything that you're finding out as well, um, this is an open book, right?
00:16:17.360 --> 00:16:18.720
And we're all here to learn.
00:16:18.879 --> 00:16:22.879
So love to hear from any of the audience and speak to you again soon.
