ChatGPT Ads Are Here: What Happens Next?

Hey. [music] Hey. Hey. Hello and welcome to the dojo, a search marketing news podcast by Exposure Ninja. My name is Dale Davies. I'm the head of marketing at Exposure Ninja and I'm joined by our fabulous CEO and extraordinary speaker, Charlie Martian. How you doing, Charlie? Hey, Dale. Doing great. How are you? Pretty good. Uh my mind is buzzing at the moment. Uh not just because of all of the wonderful speaking announcements that I hear you've been making over the last couple of days, but also because we have uh some fantastic news finally from Chat GBT and possibly well I say possibly as you can see from the above if you're watching live. It says chat GBT ads are here. I think the industry's been waiting for it and finally it's actually happening. It's almost hard to believe the time has come. Yeah, I feel like we've been talking about this for at least the last two or three years. I recall we probably did a podcast in the first year or two of CHP being released and we started to see like just how impactful it would be. We're like, well, advertising has got to be part of the revenue system, revenue uh side of things uh in the future. So, what do we know about chatbt ads so far? So what we know chatbt is launching ads in the next couple of weeks. USbased users only for now which is normal same kind of pattern as Google. US gets to play with it first and then we can expect once they've made some tweaks there may be a wider roll out of it. It's going to only be ads served to users who are on the free tier and what's called the go tier, which is one of their uh lower tiers in terms of cost to users as well. What we know about the ads is probably the most interesting thing. So, OpenAI did release a couple of mock-ups of what the ads would actually look like. And the plan at the moment is that those ads are going to go at the bottom of the answer that you get at the bottom of chatbt's answer. Then you'll get like a a little square image, a rectangular image that has both an ad creative and a short amount of text. Based on GBT's mockups, it looks like actually not that many words of text are going to be in the ad. Looks like there's going to be like five to 10 words. It's it's quite a small ad space on the previews of them. Um, the interesting thing I think is that they are going to be placed at the bottom of the answers and not inside the text, which is what a lot of users were initially worried about with the ads is how disruptive they might be to the experience of reading a chatbt answer and having a conversation with an LLM versus are they actually going to look like paid ads? Is it going to be clear what's an ad and what's not an ad? So, this first iteration to me looks like ChachiBt trying to make it very clear what an ad actually is distinct from the answer being given. Do you think people are going to scroll far enough to be actually able to see the ad? That's my first thought. I wondered about this as well. And I think sometimes it will be a yes because most of us when we're conversing with chatbt, we're actually conversing. So, we do actually want to get to the the text box to then ask our follow-up question or or do the next task. So, in those cases, yes. But sometimes, I think not always. If it's able if if it's a short question, quick answer, potentially not. Um, but I think probably the majority of time people will scroll. I think the question will be whether they overlook the ad or not. And probably the ad creative is going to be extremely important here because the majority of chat GBT answers unless you're asking it to create an image. The majority of answers are just text. They're essentially us reading very short blog posts. So the visual of the creative, if there's no images in the answer, it's going to stand out significantly. The only types of queries where I really really commonly get a lot of images are things like travel queries. If I'm asking about travel destinations, so for example, a couple of weeks ago, I was asking about um surf holidays, surf camps, and surf destinations. And Chhatbt quite often in that conversation would give a carousel of four images of surfing in unspecified destinations. So they're arguably kind of useless, but they look pretty and they make the answer look visual. Those are the only kinds of searches that I've had a lot of images come up on. And so I think those are exactly also the kind of searches where we're likely to see ads because in that kind of search it makes perfect sense for Chachi at the bottom to then actually have advertisers who are surf camps or surf schools in certain destinations. So I think it could actually make answers a lot more visual. At this stage, do we know if there are going to be varieties of ads or is it just going to be this one at the bottom? At the moment, they have just announced they're going to be placing at the bottom, but they've said this is their first test version. So, what I would assume is they're going to be tracking to see how those ads at the bottom perform first, but it could mean that ads are moved into different placements. And I would guess that they are definitely going to test different placements because you'd be kind of crazy not to. The ads would have to be so high performing and everyone so happy with the results. and chat gbt openai really happy with the income the revenue generated from the ads in order to not test um so I suspect there will be some testing over time when I do a product search in chatbt which I must admit it doesn't happen very often I I'm all in on Gemini at this point I'm surprised that actually that isn't the first place where they started to have ads like one of the three or four products that they show isn't an ad why do you think that they haven't gone down that route It's a good question and I wonder if it is because they are going to be one of the first platforms to actually have ads within it and they're worried about user trust because this is going to be quite a change for the average chatbt user that suddenly something that seems very informational, conversational, researchled now is very clearly an ad platform as well. So, I suspect they're concerned about the level of disruption and not wanting to lose user trust, but also like you say, loads of people are now switching back to Gemini since new releases, since Google's actually improved things with Gemini, and Chat GBT or OpenAI are going to be worried about losing some of that market share. So, this is maybe a little bit of a big and scary step for them to see what those ads look like. Yeah, it's Yeah, you could be right. Do you think that there's going to be any intent based around the searches that are happening? So, for example, it's only going to come up for maybe like a a more of a commercialbased conversation versus an informational one, or do you think they're literally going to drop them in at every conversation they possibly can? It remains to be seen, so it's my best guess, but I would at least hope that they're going to be do doing ads only on stronger intentbased searches. Similar to Google, if you put a very top offunnel search in there, the the likelihood of getting an ad is extremely low and arguably also a bit of a waste of time because those users are not actually looking to make a purchase at that time. So, I would guess, especially given that we know chatbt ads are going to be priced at a premium, they're going to be trying to target intent, middle and bottom of funnel type searches where the user potentially has an intent to buy. Higher intent, more likely to make a purchase. Therefore, OpenAI are more likely to actually make revenue from the ad being clicked as well. That would be my best guess, but it remains to be seen. It could be it could be really really messy, but I think users would hate to have ads if they were very top of funnel searching. And do we know if the you mentioned there about like the cost and the price of everything? Do we know if the model is based around uh you know cost per click or cost per impression or cost per um acquisition or do we know what it's going to be? Yeah. So, it's been reported so far that it's going to be $60 US per thousand impressions for CHBT ads. To contextualize that, that's about three times higher than a typical meta ad. Uh the other sort of I don't want to call it a kickback, but the other concerns is that advertisers in Chach are not going to have that higher level reporting going on. It looks like the analytics, the reporting is going to be fairly simple at the moment. And there was a lot of optimism and hope in the paid search community that there was going to be really good tracking and reporting, but that doesn't look like it's going to be the reality. I would assume that this pricing from Chatbt though, this premium pricing is because they think the intent is going to be a lot higher for the types of searches and conversations going on in ChachiPT. I do wonder I do wonder just how much they were ready to actually release this product and how much they've been kind of scared into doing so because of various developments elsewhere because of you know they're not really generating a lot of revenue. They have a lot of investment from lots of different places. They're worth billions and trillions at this point, but potentially they're like, you know, we really need to start bringing in some money of our own that isn't just sub subscriptions. I have to say the more news I read, the more it looks like they weren't fully ready for this, but I don't know if a tech company's ever fully ready for a release that they make. At the same time, the the real nail in that for me is not having insight into whether the people who click these ads actually become purchases. That's a nightmare for advertisers and anyone who's working in marketing that then has to justify a very premium cost of ads charged on impressions without being able to very clearly tie that back to an ROI. It's going to be quite difficult to pitch that I think internally at a business. But on the other hand, I do I do think intent will be higher in a lot of these chat GPT searches that happen provided the ads are relevant to the intent and that that actually works well when the ads are set up and provided the ad creatives are really strong from what the advertisers actually put in there. I think they could perform really well, but it's going to be so difficult to get the data on that. It almost feels like this is in the reverse order of how you would do things if you were trying to get brands involved. Like I would first do something similar to Google in that you do you know what used to be web master tools or you know search console where you give people an idea of like these are the different conversations or queries that have brought people to your website or in this case have made people aware of your brand. I would first want to give people an idea of like you it's not super complex the analytics that you get but you know you're getting a bit of visibility for this kind of stuff and you access that you can sign up as a brand and okay if they want to monetize it which I think is not necessarily the right way to go about getting everybody on board but you know make this data available and then say right you've got this access to this kind of data you know which which rough topics are bringing people in how would you like to pay to appeal a [laughter] bit more for some of these queries that you know that you're appearing for or you you know you wish you appeared for more here's your opportunity pay for the ad for that topic and you'll appear for you know you can select the intent actually I do want to appear for informational because it's top of funnel and people are starting to understand okay what the solution may be or the product ranges maybe or something like that and then you know you you also offer them the opportunity to buy something later on directly within the platform like the the what's it called? I've forgotten the name already. That's how fantastic and memorable some of these releases have been. The the tie in with like Shopify and Etsy and stuff like that. You then have the advertising side of that, too. Yeah. I mean, yes, you're not you're not wrong here. And I guess Chachi PT just doesn't want to be held that accountable for how these ads perform on their first launch. Because what they don't want is they've launched some ads, they've placed them at the bottom, which if we think about that compared to the Google search results, if your ads are at the bottom rather than the top, you know that they don't perform as well as the ads that are sitting at the top of the of the Google search results. We know that advertisers know that. Chat GBT knows that they're not new. They're not completely new to this, right? So perhaps they don't want to be held accountable for that level of like, oh, I put in this much money and I only got this much back at this stage, at their early release stage, because these ads really are going to be for companies willing to invest, that have budget for brand awareness, that are happy to be a first mover on this, have a experimental play with how ads perform, and not too worried if it's actually a brand awareness exercise and there isn't anything specifically measurable coming out of it because they had XK already allocated just to brand stuff and actually if they see that it starts to work somehow amazing then for them it's a bonus otherwise they've put it in the overall brand awareness bucket so perhaps that's exactly what chatbt wants that's for those first movers premium pricing brand awareness type of marketing budget they are definitely angling here towards larger businesses that may not be holding them as accountable was a small or medium enterprise who really really needs to see ROI, who needs to see a return on their ad spend. Maybe they don't want that at this stage. So maybe they've done this on purpose. If you're working at a mid-market size business, maybe enterprise, and you do get an invite from Chachi BT to join the advertising side of things, you be a bit of a a guinea pig. Would you advise that businesses do, even if they may not necessarily have a particularly large advertising budget? This is a great question. If that business is already doing well in chatbt organically, as in they show in the types of results for queries that they care about that are relevant to the business, I would probably take a gamble on doing this. I think it's a huge opportunity. It's different from what other marketers are able to do in other competitors and other businesses in your sector. And there's a lot to be said for taking a new opportunity and trying something completely different. Being seen in a place where your competitors aren't yet showing up is huge. The reason I say I would also tie this back to how the business is performing organically in AI results is because I think it will look very out of place if you show up very much at the bottom in just an ad and you're very rarely referenced in the wider answer then I think that creates a bit of a disconnect for the user. Whereas I think if it was a for for example one of the recent clients we've been working with Riverford organic veg boxes they deliver them every week. It's a subscription model. If you were having a conversation with chatbt you wanted to find a veg box delivery every week and then you get a list. So I'm going to get Riverford Ael and Cole the London Veg Box the Kent veg box etc in a list of the best veg boxes. And then I get an ad at the bottom which has a visual for Riverford. It's like a double reinforcement because I've seen them in the original text answer and then I see an ad and then there's also like a beautiful visual of their veg box with all these shiny lovely red peppers poking out of it. I definitely want that by the time I've seen it that way. So, in that kind of context, I think there is an opportunity and it's it's worth trying. You can still set a maximum budget that you're going to use. You don't you don't you don't need to throw all of the eggs and wait for them to break in in chat GBT ads, but if you had the opportunity, I would say have a crack. So, all about eggs. It's not Easter yet. All about eggs. [laughter] Okay. But at the end of the day, regardless of if you have the invite or not or access to to advertising, you got to make sure you're coming up in the organic results no matter what. I think it's much better if you're coming up in the organic results as well. Otherwise, it's it's like that's the biggest opportunity for all businesses right now. They have the opportunity to do that irrelevant of whether they are invited to the ad testing that's going on in chat GPT. And why wouldn't you want to be part of that search journey? Yeah, there's so much happening in terms of uh product education or business or brand education before anybody's really getting to, you know, midfunnel bomberunnel search, whether that's within JBT or most likely on Google. People are having conversations, learning more about your brand or your competitor's brands. And if you're not taking part of the conversation, that's it. You've got no hook. You've got no, you know, chance later on. And I I now almost always go back and forth between chatbt and Google while I'm searching. I use them concurrently to one another because I often actually don't have enough context and information to even know what I want to put into Google now. I need to have a conversation with chat GBT. So for example, when I was looking for these uh surf holidays and surf camps, I didn't even know like where I wanted to go in the whole world. I just kind of knew that I needed to go somewhere that was warm weather where the sea's not going to be absolutely freezing cold otherwise I won't want to get in there to go surfing. And so I actually rather than asking Google for you know the the water temperatures of various different locations of surf destinations around the world just had that conversation with chat GPT because it was much easier. I don't need to read every Google AI overview for 10 different searches. But then once I'd actually had that conversation and I'm thinking okay maybe I want to go to a surf camp in Sri Lanka then I'm just googling surf camps in Sri Lanka so that I'm getting like a top 10 list looking at the organic results on Google to see what there is and then I'm going back to chat GBT to ask for reviews. So I think that back and forth kind of search journey is really really common. Now, do you think that ads within chat GPT may act as a bit of an antidote to personalization within chat GBT where the chat is basically continuing a per a previous conversation that it knows about you, your preferences and so on. So, it may not recommend some of these destinations for you because of whatever. Whereas, well, I'm kind of answering the question myself. The ad can pop up and remind you about places you'd forgotten. Oh, potentially it will be really interesting to see that actually because chat GBT Open AI, they haven't announced specifically if the ads are only going to be relevant to the one conversation you're having or if they you would actually get ads relative and in relation to other past conversations that you've been having because already in organic answers, chatbt will reference back previous conversations that you've had if you have a logged in version of chatbt. So potentially you could just start to see ads all over. I could just be being advertised Sri Lanka surf camps in all sorts of conversations I'm having. Hopefully they would still be relevant conversation at least talking about holiday destinations. [laughter] Yeah. Do you think that this is going to create any immediate or further push of ads within Google as like their response or do you think Google have no interest in pushing their ads out any further? That's a great question. I think Google probably has a timeline on this internally considering that they've already um experimenting with direct offers in AI mode for the US where you can actually buy make purchases directly in the Google AI mode search results, not the normal search results, the AI mode search results in the US. I think that they're making moves towards having ads as well. most probably they're not too concerned that OpenAI is going first with this rather than Gemini because they're going to have an interesting case study in what happens to the users. And I think the interesting thing I've heard a lot of people on LinkedIn saying that they're just going to jump straight over to Gemini and not be in chat GBT because they don't want to get any ads. I think those people are a little naive about what will be coming in the near future. Do you see a lot of people making like a bit of an exodus on chap GBT? Do you think that this could be enough to make people go, do you know what, I'll go with the free product or I'll go with the, you know, one of the many other options or do you think only the power users who are laying there a lot who are the most likely to complain and grumble might be the the first and maybe only ones to jump ship elsewhere. It's interesting. It's hard to break search habits, I think. So there's a lot of people with the chatt app downloaded on their phone and how willing those people would be to get rid of that app and then get a different app for a different AI chatbot, I don't really know. Whereas power users, the the people who really interrogate the type of answers they're getting and how useful those answers are and whether there's better models and they compare models and they've used different LLMs, I think they're much more likely to jump ship to an adfree version whilst those adree versions still exist. But for the average searcher, I tend to think people don't like to to jump on and off too often. Though over time, I think it's definitely possible that chat GPT is going to start to lose some market share here. It's going to be an interesting one to see how it develops and whether this will just be just a very minor bump in the road and chat will still increase its market share or you maybe not market share. Maybe the market share is going to stick to what it is, but like the total number of users because the market just expands. Um, it be interesting to see whether it changes at all. It will be interesting to see and I think a lot of users are going to hate that ads have come into it. But on the other hand, if ads are done well, I don't think there'll be such a push back. I personally don't mind ads when ads are done well. I know that some people absolutely hate meta ads. They find them disruptive. They think they're poorly targeted. My Meta ads tend to be excellent. I get very well targeted. I find them very useful. It saves me a lot of time because if I actually get an ad for something that I need, then I don't have to go off looking for it. Perfect. So, I think it's going to all depend on the user experience. And I think at the end of the day, we've had ads within Google's search results for what, 20 years, 18 years. It's still huge. It's still, you know, increasing every day. billions of daily searches. It didn't really impact people too much. Okay, we might all complain a bit every now and again when there's, you know, an abundance of ads, but still the usefulness of it outweighs the inconvenience or the frustration of being bombarded with ads left, right, and center. Yeah, absolutely. And I don't think that chat GBT ads are going to rival Google ads. Google ads have been such a mainstay. They've been here for so long. Google is dominant in terms of the number of actual searches that happen. I don't think that there's going to be some kind of mass exodus over to chatbt and advertisers moving there and forsaking their Google ads. I think anyone any business who is actually going to experiment with chat GPT ads is going to be doing it in addition to Google ads. If I'm a marketing leader and I have somebody internally who's responsible for paid search or paid advertising or I'm, you know, working with a a partner externally for paid advertising. Is there anything I should be doing uh as a next step or should I just wait for these people to come to me and tell me what to do? Oh, I never like to wait for people to come to you and tell you what to do even though that is nice that is a nice to have from your team. Um, I would ask them about what's happening with ads, what their opinions are. People hold very different opinions about what's happening with chatbt ads. And remember, for now, of course though, this is limited. You'd have to be invited to actually be using the tests. But what I would be doing if I was one of those senior stakeholders is is keeping track of what's happening with ads and when there will be a wider roll out so that you can decide if you would want to look at testing them once there's the opportunity for it. And the best way to follow with any of this stuff when it came comes to, you know, um, AI search optimization, whether that's the organic side or the paid advertising side, will be to go and follow Charlie on LinkedIn. Just hop over there and search with Charlie March and you'll be able to catch up with everything. You won't miss anything. Do not worry. Otherwise, head over to exposioninja.com/blog orodcast and we'll always keep you up to date with everything there as well. Well, thanks Charlie and thanks to everyone who's joined us this week and we will see you all in the future hopefully next week. We shall see. Cheers. Bye.