With the introduction of AI Max, many large-scale Paid Search accounts are facing a serious challenge: how do we marry a highly granular (and strong-performing) campaign structure with Google’s dynamic, flexible and explorative AI features? This question applies equally well to the application of broad match, which has been a persistent structural pain point for many of those same accounts.
A reasonably large advertiser may have hundreds of search campaigns (with thousands of ad groups and a vast amount of keywords) for a number of different and complementary reasons: to enable highly detailed reporting, to accommodate complex bidding approaches, or to ensure the most tailored ad copy possible. These accounts are often very mature and high performing, having been optimized and fine-tuned over the course of many years. Introducing broad match and AI Max into such a delicate ecosystem can be a minefield of cross-campaign cannibalization, confusing reporting and unstable performance; where the effort needed to police search queries and the need to apply exhaustive negative lists can end up directly countering the benefits these features aim to bring.
What is the answer then? For advertisers that want to remain at the forefront, avoiding AI Max and broad match altogether is hardly an option – now less so than ever, as ad placements in AI mode and AI Overviews look to become commonplace in 2026, and complex long-tail searches continue to explode in popularity. And yet, moving away from a tried and tested account structure, one that has been built with specific business needs in mind, can be a daunting, and potentially costly, proposition.
One potential solution to this conundrum lies in looking at the underlying data being fed into your accounts. Implementing a few key changes here can directly pick up some of the challenges your account structure has been designed to address, and help pave the way for a more seamless transition towards an AI-first future – without sacrificing insight or control.
Reporting and Insights
To illustrate this; imagine you’re a retailer with 100 sub-categories across your product range, and you currently have one campaign per sub-category. The prospect of reducing this structure down to, say, five category level campaigns that lean on AI Max (with Text Customization and URL Expansion enabled) and broad match keywords, might raise a number of immediate concerns – first of which would likely be the reduced visibility on sub-category performance.
This is where SA360 custom variables and reports editor can provide a practical solution. Appending the reporting information you need directly to your conversion action – such as purchased products and product sub-categories – allows you to continue to see where you are driving results, while allowing for more consolidated data, easier account management and all the benefits of an AI-first approach.
Custom variables work across all industries, and can report any information present in the website’s data layer at the point of conversion, provided it is not personally identifiable information (PII). This data can be surfaced in SA360’s reports editor and in custom columns directly across campaign, ad group, and keyword views. As such, they remain a valuable tool regardless of account structure.
For retailers specifically – also see Conversions with Cart Data for extra conversion level insight and reporting.
Bidding: Retaining Control
You might be thinking: “reporting insights are great, but if I don’t have granular coverage, how can I retain control over bidding?”
It’s a valid question. We know that not all conversions are created equal; most accounts require distinct bidding strategies for specific business areas, services, or priorities. Fortunately, this is still achievable within «AI-first» accounts without overcomplicating your setup, after all, consolidation doesn’t mean removing segmentation entirely.
However, this becomes a problem if the need for granular bidding control prevents consolidation, leaving your data heavily fragmented and your structure inefficient. This is where an advanced bidding solution – such as a server-side Pantheon approach, or using SA360’s value rules or CVAs – can come to the rescue.
Consider a lead generation business where user value varies significantly by location. Historically, this would likely have dictated a specialized account structure, where your campaigns are geo-targeted, keywords are geo-specific and several smaller bid strategies operate with their own regional efficiency targets. The challenges this presents for an AI-first approach quickly become clear. Fortunately, this is largely avoidable today by using the solutions above to create dynamic conversion values based on user data. For example, if a user is based in London, you could apply a 50% upweight to their lead value, while applying a 20% downweight to users in Cardiff. This ensures the algorithm optimizes based on the actual business value those locations provide.
With this type of data fueling the machine learning in your account, suddenly a segmented structure becomes less critical. You can still retain direct influence over how bidding algorithms approach your key business objectives, while aligning with AI-first best practice of modern PPC strategy.
Furthermore, advanced bidding approaches are applicable across industries, from a travel business wanting to prioritize key routes, to a retailer seeking to incorporate LTV factors alongside revenue or profit bidding.
A Daunting Prospect But With New Opportunities
For many, the prospect of moving away from a long-standing, granular account structure can feel challenging. This type of major change comes with risks and challenges, particularly during the transition period, but it shouldn’t become a blocker. Staying effective today means keeping pace with a rapidly evolving search landscape.
If you’d like to discuss how to approach a major account change, take a more AI-first approach to PPC, or explore any of the topics covered in this article, our team would be happy to help at SearchAdTech@incubeta.com.
Written by Gregory Rogers, Senior Search Consultant, Incubeta.
