# Your ChatGPT Traffic Is 5-10x Higher Than You Think

## Метаданные

- **Канал:** The Grow and Convert Marketing Show
- **YouTube:** https://www.youtube.com/watch?v=uLuL4axbMcQ
- **Дата:** 17.03.2026
- **Длительность:** 27:33
- **Просмотры:** 87

## Описание

Your ChatGPT traffic is probably 5-10x higher than what your analytics shows.

We used to think AI traffic was a small fraction of Google. We showed you the data a few episodes back — ChatGPT was about 3-4% of organic Google traffic across our clients.

We've revised our thinking.

In this episode, we break down why analytics tools are dramatically underreporting LLM-driven traffic and leads, why Google AI Overviews are creating the same attribution blind spot, and what you should actually be doing about it.

The core issue: when someone discovers your brand in ChatGPT or a Google AI Overview, they don't click a trackable link. They open a new tab, Google your name, and land on your homepage. Your analytics says "direct" or "organic." The real source? An LLM.

The only way to know? Ask them.

We walk through real examples from our own data and client accounts, show exactly how this plays out in ChatGPT and Google AI Overviews, and share what we're changing in how we measure and report on content performance.

Check out our latest article on Invisible Prompts: https://www.growandconvert.com/ai/invisible-prompts/

Timestamps:
00:00 – Why your ChatGPT traffic numbers are misleading
01:00 – The attribution gap: what analytics tools are missing
04:30 – Our previous data: ChatGPT traffic as % of Google organic
06:30 – Why that data dramatically understates the real number
09:30 – What our own lead form data is telling us
10:45 – The 5-10x multiplier rule of thumb
12:00 – Google AI Overviews are creating the same problem
14:00 – Impressions vs. clicks: what Search Console is showing
17:00 – Why showing up in AI Overviews still matters
20:00 – How AI Overview attribution actually works (it's tricky)
24:00 – What you should do: self-attribution, broader measurement, and GEO
26:00 – How to actually get mentioned in LLMs (hint: bottom of funnel SEO)

## Содержание

### [0:00](https://www.youtube.com/watch?v=uLuL4axbMcQ) Why your ChatGPT traffic numbers are misleading

Whatever chatbt traffic you see in your web analytics, I think you should multiply by 5 to 10. — The way people are searching is changing and analytics tools and the way that we attribute leads hasn't really caught up to how quickly this trend is changing. — You can't ignore this anymore. And we have, you know, this semi-quantitative, semi anecdotal data suggesting this is a big deal. Like there's a sizable number of people coming into ChatgBT and your analytics tool cannot measure it. So that means you need to be doing GEO and we're focused on that with every single account. Now — what's really changed in the last year is how people are discovering and making decisions about software or services. They used to mainly use Google and now they're using multiple different platforms to make decisions. So they may start their search on an LLM, then go into Google and do more research, then go to your site, then go back to the LLM and ask follow-up questions, and then come back and convert.

### [1:00](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=60s) The attribution gap: what analytics tools are missing

Okay, welcome to the Grow and Convert Marketing Show. I'm Davish. That's Benji. Today we're going to talk about this discrepancy between AI traffic that you see that seems like AI traffic is low. It seems like chat GPT traffic is low, but how what we're seeing anecdotally and also quantitatively from us and our clients is that leads are actually coming from AI, Chat GPT, Google AI overviews and why you could be misled into thinking it's not a big deal because the traffic looks low, but you're actually just not able to measure the vast majority of traffic on LLMs, primarily not only chat GPT but also Google AI overviews. — Yeah. And even on the conversion side, what's interesting is a lot of the conversions that do are selfattributed to chat GPT when we look at analytics tools, they're actually coming in as direct or organic search. And so the only way that we're able to know that it comes from an LLM is just by asking people. And it's if you don't have that on your forms, you could be misled into thinking that you're getting a lot more direct traffic or conversions branded uh organic conversions that are coming to your site and you just might think maybe my brand is growing or people are going directly to the homepage and converting that way. But when we ask them, "How did you hear about us in the form? " they're actually saying, "Shhat GBT or Grock or Claude or one of the LLMs. " And so it used to be where we could use Google Analytics or HubSpot or a lot of the analytics tracking tools to measure this. And we have a whole slew of them for both ourselves and our clients that we're using. Um, but yeah, we're kind of being misled in terms of what's actually happening. And the only way to figure out where people are actually come from coming from is to ask them because I think what's really changed in the last year is how people are discovering uh and making decisions about software or services. Uh they used to mainly use Google and now they're using multiple different platforms to make decisions. So, they may start their search on an LLM, then go into Google and do more research, then go to your site, then go back to the LLM and ask follow-up questions, and then come back and convert. And so, that user journey has somewhat changed. And I think what's really important now is to just make sure you're showing up in all the platforms. So, when someone's asking a question in an LLM, you need to be showing up there. when someone's showing up or asking a question or searching a keyword in Google, you also need to show up there because if your brand is not showing up when people are running these searches, you're kind of losing out on those opportunities. So, I want to add some data to this. This is a spreadsheet that we shared uh a few episodes back. I believe the title was like AI versus SEO traffic, what our client data really tells us. And we looked at the data from seven anonymized clients comparing

### [4:30](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=270s) Our previous data: ChatGPT traffic as % of Google organic

Google just organic regular Google organic traffic to ChatPT traffic. And for those of you that aren't watching the video, don't worry, I'll read it out. Basically, the conclusion was the average percent of Chat GPT traffic as a percentage of Google like the ratio was chatbt was like 3 to 4%. Some of them as low as 02% high one of the seven as high as 7% but the average was 3. 8%. Meaning like you saw ratios like this 2,000 Google organic visitors for 51 chat GPT 35,000 Google organic for 1,000 chat GPT. So we made the argument in that video of hey cha GPT is growing and we showed graphs and said like AI traffic is growing but it's still a small fraction of Google and I forget exactly how many months back that was. — Yeah, maybe two or three months ago and there's a lot of people who've showed almost the same data. I know this is a small sample size, but I think Hrefs or Gayana or someone else shared similar findings across the clients that they're looking at that is somewhere around 5% is what they're seeing attributed in analytics tools to chat GBT and LLM versus Google. But what we're saying today from what we're seeing in the conversion data and just getting more experience and data in the last few months since that episode is the opposite that this is actually a massive under representation of how many of your customers are actually using chatbt for product discovery where you need to be present. Why? Simple answer that is trackable chatpt traffic by Google Analytics. How does Google Analytics know that someone came into your site via chatbt? It needs that referral source, the UTMs. It needs the UTM parameters in the URL. Those come if

### [6:30](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=390s) Why that data dramatically understates the real number

someone clicks the source in ChatGpt where you're mentioned. So it where the where you're cited as a source by chat GPT. So the question to you whoever is listening is how many times in your own LLM usage whether it's chbt claude or whatever do you click the sources for me it's like one in 30 sometimes less — yeah it's super rare — right and so what if it's 1 in 10 so like there it was 3% right if it's one in 30 you multiply by 30 you get too close to 100% that. So, I don't know what it actually is, right? But the point is there is a sizable amount of people that are searching things related to your product space that are trying to discover you via chatt. How else do we know this anecdotally? Um, we can't share this on the screen, but you and I right before we recorded looked at the analytics data of a client that you know anecdotally on the calls is reporting more and more leads are coming in saying I heard about you on chatpt. We see in their tracker data they are getting mentions in chatpt like brand mentions and we can talk about that later briefly at the end how you need to be tracking brand mentions versus citations as well. And when you look in Google Analytics, they're attributable. What Google says, what G4 says is the tragedy traffic to our pieces is like 30 to 40 pe uh visitors a month, almost nothing, but they're saying people are hearing about us. And the big reason is you just can't track it because if you ask chatbt, in fact, we can go into tracker and look at an example. So we'll use ourselves as an example in tracker. Content marketing agency chatbt mentions us as a brand. I can click into that as a brand mention and see in the screenshot where it mentions grow and convert here. Examples of content marketing agencies well-known agencies buprow grow and convert. The citation is from some article by 10speed eight best content marketing agencies. So if you wanted to reach out to grow and convert if we get a lead from this what are they going to do? They're going to copy this name and they're going to open a new tab and Google us and click in through our homepage. So that may show up as Google organic. And so in that previous spreadsheet, we were showing an example. Let me go back to that. In that previous spreadsheet, we're saying, "Oh, look, ChadByd traffic is so low, Google organic is higher. " But in that example I just showed you, which is probably happening over and over again for like every business. it's going to show up as ch as Google organic even though it actually came from chatbt because they're just going to open a new tab and Google your brand name and then click onto your homepage and so we're seeing that sort of anecdotally but also

### [9:30](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=570s) What our own lead form data is telling us

for those that measure we're seeing that now in our own data we do ask in our field how did you hear about us and to be perfectly candid most people have not said chatbt we have had a couple but in the time in the last few months where you just tabulated we had two chatt leads that means leads who manually in the form fill the human itself wrote chatgpt into the how it hurt how did you hear about us field we probably had for those two we probably had what 10 Google that's that wrote in Google I found you on Google — yeah it's still much higher at this point but I think had we not had that form field we would have just thought it was direct so it's important for us to know and I think we've had two in the last couple months because we've only started to invest in doing AI specific activities uh since December. And so we are also seeing a direct correlation between doing things like updating some of our pieces to focus more around keywords, improving our SEO rankings, and then also doing citation outreach. And since we've started doing those activities, we've started to see that we've had much more visibility in the LLMs and we're seeing a correlation between those activities and then also getting more leads from CHA GBT. So, I'm

### [10:45](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=645s) The 5-10x multiplier rule of thumb

going to sort of go out on a limb and just give kind of a general rule that I want to make sure listeners don't get mad at me about or hold me to. It's not particularly quantitative. I think whatever chatbt traffic you see in your web analytics, I think you should me you should multiply by 5 to 10 and assume it's that much. Again, I can't defend that because like who has the data on how many people are clicking into searches? Open AI. You don't have that data. Nobody has that data of when somebody asks chatbt for like product related recommendations, what percentage of those chats does the user actually click into a source? It's that ratio that we don't know. But you could just pay attention in your own usage. I swear it has to be 5 to 10. Five would be one in every 20 chats you click into a source. And 10 is one in every 10. — It's not just LLMs. It's also Google too because — Google keeps I feel like every few months keeps releasing more and more searches that have an AI overview. — And so where you used to rank number one with an actual piece where someone would click into that piece and get the answer, now you just search for an answer and it gives you a quick answer. And so if you get the quick answer, there's usually no need to dig in anymore. You're like, "Okay, top content

### [12:00](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=720s) Google AI Overviews are creating the same problem

marketing agencies. I'm gonna go reach out to the top three or just do research on those top three. You might then go Google those people separately. And so — all we're saying is the way people are searching is changing and analytics tools and the way that we attribute leads hasn't really caught up to how quickly this trend is changing. — It's really hard to attribute. But I I'm going to make an even bolder statement and just say from our previous take where we said chatbt traffic is way lower. I have revised my thinking and my takeaway if you're listening to this the takeaway from me that I think everyone should have in the marketing space is you can't ignore this anymore and we have you know this semi-quantitative semianecotal data suggesting this is a big deal like there's a sizable number of people coming into chat GBT and your analytics tool cannot measure it so you can't just measure the traffic and say look Google is way higher it's way higher. That's why I said multiply by five or 10. So that means you need to be doing GEO and we're focused on that with every single account now is in addition to SEO, how are we doing on those chats? And maybe at the end if we have time, we can do a little bit of a recap to remind people of our prioritized GEO strategy. But the next thing I want to cover is exactly what you said, Benji, which is Google AI overviews. So, the other thing that a bunch of us are seeing that a lot of people have reported is this difference between your rankings and the actual clicks and organic traffic you're getting from Google and why it looks like your organic traffic is decreasing, but your actual exposure to the searches that matter where someone is googling for products or services in your space, how those actually seem to be staying steady and in some cases you can see increased impressions or whatever and so people are still doing that and but you're just not able to attribute it

### [14:00](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=840s) Impressions vs. clicks: what Search Console is showing

anymore and see those as clicks. So here's the example. We'll use us an as an example and two clients. So first of all this purple line every time you're going to see it as impressions. So we have this weird anomaly in impressions where when that numbum 100 tracking thing from Google happened um back here in September uh of this last year 2025 we saw a big decrease but overall you see before that you know a slow increase in impressions and then after that maybe another slow increase back up to the previous level in impressions. So it looks like if anything impressions what is impressions this is I'm looking at search console data but it's powered by SEO gets impressions in this case is how many times you're appearing in the SER in the Google results when someone Googles something. So it is a reflection an important reflection of your SEO work. If you are not ranking for any keywords with any content your impression number is going to be zero right at the beginning of a website. And if you're doing really well in SEO, your impression number is going to be high. But it's not clicks. Let's switch this to clicks. You see a different overall trend. You see an overall decrease from 2024. You see things have been decreasing. And then a sharp another sharper or or slight decline in kind of the fall of this last year of 2025, which a lot of people online I've noticed have reported as well. — Yeah. And I think that's largely due to Google rolling out more and more AI overviews. I know even last year if you were to search for top content marketing agencies, you would just get a like sites that produce lists of different content marketing agencies or maybe Clutch would show up. But now you're just getting an AI overview for that. And so I think that's a large reason people are reporting a decline in clicks or they're looking at their traffic graph and saying, "Why isn't my traffic going up? It seems to be in a downtrend. What's going on? " But then if you look at your rankings, they're actually improving. And we just had a client call the other day where in Hrefs, when you look at a lot of the number one rankings, it also is starting to count number one being an AI overview because that's just another way that someone is discovering your company. Now, you may show up number one, which means you're in that AI overview when it appears, but instead of you getting the traffic benefit of someone clicking into your article and reading your point of view on whatever topic it is you wrote about, now your brand is just getting shown in that AI overview. And so, there's a lot of people that are questioning, well, is that not as important? And yeah, you could view it like that. you're not getting the same traffic benefit or the conversion benefits of someone reading your article and then converting directly from that. But if you're not showing up here and this is how people are starting to do all of their searches, they're just looking for a quick list and then this is their starting place now. This is how they're discovering the different options and then they might go do deeper research in an LLM or on your website or

### [17:00](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=1020s) Why showing up in AI Overviews still matters

in Google. Um, but if you're not showing up here at all, you're not even in their consideration set. So, it is very important that you show up in the AI overview in the top spot if you have content there and in the LLMs because again, this is how people are now discovering the products or services that they want to use is through the AI overviews, through the LMS. And again, you're just not able to track how many people are viewing this AI overview and then coming to your website because of that. And so it's just a largely an attribution issue right now that we're facing. — Yeah. So here's an example. So I Googled content marketing agency for B2B SAS companies. After the sponsored results, the ads, you see one organic uh listing and then you see the AI overview. In the AI overview, it mentions us in two places. So this is important. So in the opening paragraph, it just mentions us by text. It says include specialized partners like animals, siege media, and grow and convert. — And side note, an important side note, — next to each of those, it puts a little parenthesis that's like it's — kind of like differentiator or value prop. Animals parenthesis premium thought leadership. — Siege media parenthesis SEO focused design. First of all, I don't know what that means. And then grow and convert parenthesis conversion focused. I'll take our value prop all day over those other ones. By the way, um that this is a bit of an aside. And I'll get back to the main thing, but this is also something we have an article on that we'll put in the show notes and we actually just re kind of titled it, but it's um why ultra specific bottom ofunnel content is critical or key in AI search because it needs to have read the content growing convert has produced AI in this case AI overviews and then it synthesizes it in this case into a little parenthesis value prop. And if you look further down, there's a full bullet with like a longer sentence of it synthesizing. And Tragybt does the same thing. If I were to open some something where it mentions grow and convert, it'll do the same thing. In fact, that previous screenshot I showed, you could rewind, does the same thing. But our sentence here is focuses heavily on conversion and bottom of the funnel content, case studies/o rather than just traffic. Great. Would I write it exactly this way? Maybe not. But like I'll take this from AI describing our business, right? So like you how you position yourself matters. But I'll leave that okay leaving that aside away back to what we were talking about. So if this is how people are discovering you like Benji said two things can happen. First they see your brand name up in here in this opening paragraph. What are they going to do? There's no link. Nothing. I'm clicking. There's nothing. So I would highlight this copy. Open a new tab and just Google grow and convert.

### [20:00](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=1200s) How AI Overview attribution actually works (it's tricky)

you're going to see that as coming to the homepage because obviously if you just grew Google grow and convert the you're just going to get like our homepage or whatever. But be careful in this case we don't know what actual piece of content had us in was responsible for getting us into the AI overview and it may not be just one our own content. So it could be that we are mentioned in other people's list posts and you see in the actual organic listings one. So there's one above. So let's count this. One, two, three, four. Some more ads. Wow. An ad spot right in smack in the middle of the SER. I hadn't seen that yet. — Even this is changing a lot. — Five. Six. We have the sixth organic spot with what? With an article. That's best SAS content marketing agencies in 2026. In fact, we had someone on our team recently do an update and create creator updated this article. So, from us, from a marketing decision-making standpoint, we would we it would be if you're just looking at your analytics, you have no idea the chain of events that led you to get like if we got a good lead from this, how did you hear about us, Google? and you look at when the lead came or whatever something and you try to do some sleuththing in your G4 you're going to see it if they did what I just did which is just Google grow and convert in a new tab you're going to see it as like I don't know it came in from the homepage so we don't know what they googled and you think oh just the homepage brand like you said you think oh maybe our brand is growing maybe it's the LinkedIn I'm doing I'm just so clever on LinkedIn right but re but the reality is down here was a blog post that Google AI overview read and decided, okay, I'm going to include that. And maybe other things, maybe it's the outreach, maybe it's some getting our us mentioned in other people's blog posts. And the sum of those gets you into the AI overview, and you're not going to figure out that was the reason by looking inside your analytics. It's the same thing as I just mentioned on chat GPT, but the Google AI overview version of it. The second way that you can use it is in here in the bullets. If you're mentioned in the bullets, you see I can hover over grow and convert and there's this dotted line underline. Let's see what that does. I'm going to click it. You think it's going to go to the grow and convert site. It doesn't. It looks like it Googles grow and convert, but it doesn't quite. This is extremely tricky and misleading. So, it looks like it just Googles grow and convert and then you click into the homepage. So, it's the same thing but just kind of that as I showed you opening a new tab and googling it, but it's not actually googling grow and convert. It's extremely misleading. In fact, I think it's like borderline like not appropriate by Google cuz so for those that are just listening, the actual SER shows a Google field where it looks like all I did was Google grow and convert. But the reality is in the URL in the query parameter it shows that the words I googled it googled was grow and convert content marketing agency for B2B SAS companies. It's actually googling the combination of your brand name and your previous search together. And what I've noticed is when you click into these dotted line stuff, you don't always get that company's homepage. Sometimes you get a random blog post or whatever. But the point is it's comp AI overview has completely muddied attribution but it's not fair what you said it's not fair to say oh it's not that useful to be in the AI overview blah no it actually is like this is where people are making their decisions I mean I use it all the time if I'm looking for quick information I just Google something I look at the AI overview I get what I need and then I go do more research or I go on my way and look into that brand or look into that company so I just know I've noticed my own search habits have been changing and this stuff is just not as trackable as

### [24:00](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=1440s) What you should do: self-attribution, broader measurement, and GEO

it used to be. So, I know we only have a few minutes that we can keep talking here. So, what should people do? I think number one, they should if they can and I know this is difficult as some companies cuz there's some politics around lead forms. Obviously, people don't want to add more fields cuz they feel like that hurts conversion rate. But I do think it is important both in your lead form and on sales calls for people to ask the prospect how they heard about you because that self attribution becomes really important. And for us, it's kind of given us a new lens to look at conversions through which we didn't have if we were looking at just analytics tools. And then the other couple things that we're doing is we're starting to instead of just looking at purely directly attributable leads, we're now starting to look at just overall organic lead growth, overall direct leads that are coming from the homepage and making sure that those numbers are going up. Because again, if we're saying LLMs, people's search habits are changing and a portion of the those leads are coming from LLMs are just they're influenced by SEO, but they're not fully attributable to it. We just want to be able to look at directly attributable leads. We branded searches that are coming to the homepage. We want to look at direct traffic and direct conversions. And so we're just starting to broaden out and look at things more holistically because if we're just looking at direct attributions from content, just our analytics are starting to break down and it doesn't give the full picture of what's happening from SEO and content. — Yeah. Some other takeaways to add to that. How do you actually get mentioned on here? What should be your strategy? How do you know what people are typing into chatpt and all? You can read prioritize GEO for it. We have an entire video talking about this and we'll link to the article below. But the TLDDR of that is that you need to still rest on Google that the way to control your mentions on LLMs is when the LLMs

### [26:00](https://www.youtube.com/watch?v=uLuL4axbMcQ&t=1560s) How to actually get mentioned in LLMs (hint: bottom of funnel SEO)

go and do a web search to figure out what to say. So for none of these product and service related queries is chat GBT. First of all, Google AI overview is by definition a search summarizer. So obviously you need to be in the SER AI overview. That's what it does. ChachiBT is different. It's a general purpose LLM, but it's smart enough to know that when someone's asking for a product recommendations, I have not seen, and we've done a ton of testing, not even once have I seen it just do it off the training data without any um searching. It does searching and then it blends it with what it knows from the training data. So, good luck getting into the training data. That's hard. Your way to do it is by in the web searches. And those keywords to search show up for and target with SEO is bottom offfunnel product related searches. So, leave aside your 10 tips, intro guides, and all that top offunnel nonsense you were doing. If you're listening to Grow and Convert, hopefully you weren't doing that in the first place. But more importantly, these LLMs are googling best blah blah software, best blah blah agency business to figure out what to say to its user. And now, it's not going to use it verbatim, and organize and rank it by who's ranking highest on Google. Those old school SEO things don't apply. It'll use its own reasoning and logic to decide what to recommend. But you need to be there somewhere, otherwise you have no shot. If you like this video, don't forget to subscribe. You can also get the audio only versions of these shows wherever you get your podcasts. And you can follow us at growandconvert. com/newsletter for any articles and updates for when these videos come

---
*Источник: https://ekstraktznaniy.ru/video/46212*