What is Generative AI and how is it transforming Search and Paid Media in 2025

It is no understatement to say that the impact of AI in marketing is huge right now.
Here we take a look at some of the top uses cases that marketers are finding for AI, and consider how it is transforming both Organic and Paid Search and how marketers need to plan with this in mind for 2025.
In particular we consider:
- What generative AI is
- The 5 most effective use cases of AI
- How the AI Search arms race is upon us
- How AI overviews are now prominent in Google searches
- Google’s plans to add Paid Ads to AI overviews
- Changing the way we think about SEO optimisation for AI
- The BIG shift of control to AI on Paid Ads
What is Generative AI?
So, as a starting point, what is Generative AI? Here are a few definitions to set some context.
According to Forbes:
“Generative artificial intelligence (AI) describes algorithms (such as ChatGPT) that can be used to create new content, including audio, code, images, text, simulations, and videos. Recent breakthroughs in the field have the potential to drastically change the way we approach content creation.” Source: Forbes
But in the true spirit of using AI, here are also a couple of definitions which have been provided by 2 of the big AI players. First this is how Gemini (formerly Bard), Google’s AI offering describes generative AI:

And, for good measure, a definition provided by ChatGPT.

So, in the most basic sense, Generative AI is type of artificial intelligence which is able to create new content in a variety of formats from text to developer code – based on the prompts it gets from humans.
It is revolutionizing several functions, including marketing, by automating creative processes and enabling the generation of unique content at speed. Here are some of the most commonly used tools in the industry that you may recognize from your own work:

A quick caveat – there are literally hundreds of AI tools and models out there and the list is growing exponentially. So, the list above is a starter for ten only on some of the more mainstream tools out there.
How are we using Generative AI in marketing?
The simple answer on this is that marketers are using AI in a huge number of ways. But the data from Statista below gives us a good starting point for looking at this, by identifying the most effective use cases that they came across in their research.

Source: Statista AI Automation use cases
Let’s take a quick look at the top results:
Customer data analysis – one of the top results in these findings was generative AI insights. There is clearly huge potential for using AI to take performance data and train the model to uncover the insights from it. Uncovering patterns and trends in customer and revenue data. This can have huge efficiency gains by removing the need for manual data pooling and letting AI do the hard work.
Personalisation – AI can also be used to better customize and target content for individual customers based on their past interactions, preferences and behaviours. If offers the ability to collate and analyse vast quantities of data almost instantly and then use that to target content across email, website and ad content. Which is hugely powerful.
Content creation – improvements in Generative AI have spawned a plethora of tools that quickly and easily enable the ideation and generation of content across a range of formats including text, images, video and more.
Customer Service – there is huge potential and upside in using AI to improve customer support and this is increasingly something we are seeing in our own personal interactions with Chatbots on websites which are becoming the first line of automated defence in customer service strategies. According to research by Liveperson, 84% of businesses are using some form of AI to interact with customers.
Predictive analytics – and the potential for leveraging AI in analytics is huge.
At QueryClick we have also been using AI for a number of years to improve the quality of marketing attribution for our customers. One of the key attribution challenges here is that solutions like GA and Adobe, which rely on cookies for tracking, do a particularly poor job of tracking multi-session, multi-device journeys. Which results in around 80% of data being incorrectly attributed for attribution purposes.

This is why we developed our own Corvidae attribution platform which uses a combination of AI and Machine learning to effectively replace cookies and rebuild this data from the ground up. Giving our customers not only a much clearer view of what is and isn’t working in their campaigns – but also very specific predictive suggestions for how they can re-allocate underperforming spend to optimise their revenues.
The AI-powered Search arms race is upon us

There is also a scramble to apply AI to Search right now, and it is not just Google that is pushing the agenda here. There is a much broader transformation taking place as search engines race to integrate AI models and LLMs to help them provide a more contextual and conversational approach to search.
We have probably all seen the signs of that through things like Copilot, which many users have running on their Windows desktop now – or are coming into contact with on Bing searches. Google is also using AI to generate overviews in Search that are designed to reduce the work associated with researching topics on the web.

However, now a new player has entered the market. OpenAI which owns ChatGPT, potentially encouraged by the ongoing anti-trust issues which Google is experiencing in the US – and even the potential that they will be forced by the DOJ to sell off their Chrome browser – has launched SearchGPT. The AI-driven search capability is available to ChatGPT paid users right now – but is due for wider roll-out into its user base – and summarises information from websites along with an attribution link.
This is a clear statement of intent from OpenAI and it is a move that has the possibility to completely change the face of Search. Fundamentally altering traditional search behaviours and also bringing lucrative new opportunities for advertising and optimization of content.
AI responses are now prominent in Google searches
Google has also started to embrace its own use of AI in Search with AI overviews appearing in certain searches depending on the query.

These overviews are:
- mainly on informational terms and searches like ‘how-to’ and ‘what-is’
- similar to the Gemini model where the AI is scouring sources and trying to formulate a response to your query
- taking the form of an almost conversational answer
The bottom line is that this fundamentally changes the way that Google is surfacing content for users:
- no longer do you have just simply 10 organic search listings and a few ads
- you now have different snippets which the user can navigate through
- the overview takes sources from the web
The practical implications for marketers? It completely changes the way that you need to think about Search and optimization of Ads and content.
Google plans to put Paid Ads in AI overviews

So, the integration of the AI overview has implications for Organic Search and the way that listings are being adapted to accommodate it.
But marketers should also take note of the fact that Google is already also rolling out Paid Ads inside these AI overviews. The first image on the right shows where Google is headed with this but the second image is an example of Paid Advertising that was seen in the US in November – something that will be released in the UK and Europe in coming months.

What this means for advertisers is that their ads and products are going to be part of a more interactive engagement where the user is able to ask follow up questions about particular products that directly impact the ad content shown. So the focus needs to be on understanding what it takes to get into those AI overviews and queries.
Which means that advertisers have to do the following to ensure it is their products and services that are surfaced in these AI overviews:
- have their product ad inventory thoroughly optimised
- ensure their feed is in good health
- make sure that they are using all of the attributes that are available to them
Changing the way we think about SEO optimisation for AI
Below is a recent practical example of how QueryClick was able to work with a beauty supplier to ensure their presence inside those AI overviews.
The focus here was on making sure that the content was authoritative and showed expert knowledge to ensure that it was cited as a source and had good presence within the overview. The question for us and the customers was how do we become expert or authoritative enough to get into those AI responses?

It is no longer a case of just publishing some content and hoping that it’ll rank on Google because queries will turn into a subset of ‘here’s the answer’ and ‘these are the sources’ – instead of the traditional sequential listings on a search listings page. The focus needs to be firmly on authoritative optimisation.
It also has the potential to create what we call ‘Zero-Click’ searches. Take the example of a user searching for ‘what is K18’ or ‘how do I manage dry hair’. In a world of AI overviews the answer may well be surfaced by the AI overview with the need for the user to click through on to your website. For performance marketers, that creates a very real challenge when it comes to measurement – so, ‘what impact does this actually have If nobody’s clicking on the site’ and ‘how is that journey happening’?
It is something we will return to later but it’s something to consider on both Organic and Paid Ads.
There is a BIG shift of control to AI on Paid Ads
Over the last 12 months we have seen a big shift of control to AI or Machine Learning models in Paid Advertising.

Lack of match type control and granularity in Google Ads
There’s a lack of match type control that we have probably all experienced if we’re Paid Search advertisers. Basically, Google is pushing us to try and put all match types into one campaign which is a particularly broad match that they favour. As Paid Advertising experts we feel that this is taking control out of our hands and placing it firmly with Google. And this has been frustrating when we have wanted to target specific keywords to specific audiences. This is something we see becoming more of a challenge with Google Ads going forward.
Consolidation of placements with PMAX

There is also some consolidation of placements with Google’s AI enabled PMAX. The reality is that Google is pushing a lot of advertisers to go towards the PMAX route – and while PMAX is good , and it has its benefits, the reality for advertisers is that it also results in lack of control for them.
Consolidation and lack of control in Meta Ads

We are also seeing consolidation and lack of control on Meta Ads right now. Again, if you run Facebook and Instagram ads they’re pushing you down the Advantage Plus route – Advantage Plus shopping and Advantage Plus Creative. What they are asking is to let Meta control all of the creative elements of the campaign. They are effectively saying – ‘we’ll overlay things, we’ll move that creative to different formats and different sizes’. The issue for marketers is that this is increasingly taking control out their hands and shifting it back to the platforms. And more specifically the AI that is being used to optimise these campaigns.
What makes QueryClick different?
Download our services brochure to find out.
Own your marketing data & simplify your tech stack.
Have you read?
Generative AI is transforming the way that marketers plan and assemble content for their Paid Ads. As big platforms like Google, Meta and TikTok increasingly build the tools needed to...
In a surprising move that has sparked heated debate, Mark Zuckerberg announced on his Instagram that Meta will be reducing its levels of censorship and in particular fact-checking on its...
It is no understatement to say that the impact of AI in marketing is huge right now. Here we take a look at some of the top uses cases that...