On episode 333 of PPC Live The Podcast, I speak to Nils Rooijmans, a renowned Google Ads script expert and top 10 PPC influencer, where she shares the experience of a costly error that serves as a valuable lesson for anyone managing paid search campaigns.

The Setup: A quick account onboarding gone wrong

The trouble began when one of Rooijmans’ existing clients acquired another company in the airport parking services industry. The acquired company was already running a small Google Ads account, and the client wanted Rooijmans to manage it without paying additional fees for proper onboarding.

Against his better judgment, Rooijmans agreed to a compromise: they would slowly migrate the new account to their existing setup over time. The new account would be left largely unmonitored during this transition period.

The fatal mistake: Ignoring Google’s warnings

After six weeks of minimal attention, disaster struck. Clicks and conversions dropped dramatically, eventually falling to zero. When investigating, Rooijmans discovered that Google had sent multiple emails warning that the consent management platform wasn’t implemented correctly. The emails threatened to stop allowing conversion tracking if the issue wasn’t resolved.

“We were very ignorant. We didn’t read the emails from Google, and we were relatively slow in responding to the issue,” Rooijmans admitted. The result? Google stopped processing conversion tracking data for that specific domain entirely.

The cascading effect of lost conversion data

Without conversion data, Google’s smart bidding algorithm made a logical but devastating decision: if clicks aren’t converting, reduce CPC bids to avoid wasting budget. Traffic gradually decreased, actual conversions were still happening but going unrecorded, and the campaigns scaled down to nearly nothing.

The website continued to receive bookings from other sources, which initially masked the severity of the problem. By the time the issue was fully identified, significant revenue opportunities had been lost.

The root cause: skipping proper onboarding

Through a detailed root cause analysis, Rooijmans identified the fundamental problem: allowing a client to bypass the standard onboarding process. Without proper setup, several critical safeguards were missing, including monitoring scripts for conversion tracking, assigned team members to check account emails, and standard processes for account health checks.

“During this root cause analysis, I always ask myself the why question five times,” Rooijmans explained. This technique, borrowed from quality management practices, helps identify the underlying cause rather than just treating symptoms.

The client conversation: managing expectations

Breaking the news to the client proved complicated. The business owner, Rooijmans’ primary contact, was relatively understanding since ad spend had also decreased. However, a meeting with the company’s CFO took a different turn.

The financial executive expected compensation for the lost revenue opportunity, even though actual bookings were still occurring and not all revenue was truly lost. To maintain the relationship, Rooijmans reduced his invoice, though he noted that the client had agreed to a slower migration process with limited oversight.

The technical fix: Working around Google’s limitations

Fixing the conversion tracking proved surprisingly challenging. Google support couldn’t resolve the issue despite multiple contacts with different departments. The problem was technical: Google had flagged the specific domain name, returning HTTP 400 errors for all conversion tracking requests instead of the normal HTTP 200 responses.

The workaround involved either importing conversions from Google Analytics (GA4) or setting up new conversion tracking through the manager account. “Don’t expect any help from Google,” Rooijmans warned based on this experience.

In the short term, switching from smart bidding to manual CPC bidding restored traffic levels while the conversion tracking issue was being resolved.

Key lessons for PPC professionals

Never skip onboarding

Regardless of client pressure or budget constraints, proper account onboarding is non-negotiable. Standard processes exist for good reason and protect both the agency and the client.

Monitor conversion tracking religiously

Rooijmans runs automated scripts that monitor conversions and conversion values, alerting him immediately to significant changes. In the era of smart bidding, conversion tracking is the foundation of account performance.

Don’t be arrogant about Google communications

While many Google emails are promotional or contain unhelpful suggestions, some contain critical compliance information. The challenge is distinguishing between noise and genuine warnings.

Implement the “Fail Fast, Fix Fast” culture

When mistakes happen, the priority sequence should be: take a deep breath and assess the situation calmly, fix the immediate issue to restore performance, communicate transparently with the client, perform root cause analysis, and document learnings in a post-mortem.

Use the five whys technique

Asking “why” repeatedly helps uncover root causes rather than surface-level symptoms. This prevents the same mistake from occurring in different forms.

Additional Common PPC Mistakes to Avoid

Black Friday budget management

During high-traffic events like Black Friday, Google may spend significantly more than usual. While the platform allows spending up to twice the daily budget, exceptional circumstances can push spending even higher. Monitor spend closely during peak periods and be prepared to adjust budgets to capture opportunities.

Double-counting conversions

One audit revealed a freelancer had set up both imported Google Analytics conversions and native Google Ads conversion tracking, both marked as primary. This inflated reported results by roughly 100%, creating unrealistic expectations that had to be carefully managed when corrected.

Building a mistake-tolerant team culture

For agencies and teams, Rooijmans recommends several practices including implementing a second pair of eyes to review all work, encouraging experimentation with clear hypotheses and measurement plans, anticipating both positive and negative outcomes before testing, and maintaining detailed documentation of lessons learned.

“We learn through mistakes, and that’s part of the process,” he emphasized. The key is creating systems that catch errors quickly and turn them into learning opportunities.

The bigger picture: Remote work and PPC success

Interestingly, Rooijmans credits his PPC career with enabling his digital nomad lifestyle. Working from locations like Curaçao in the Caribbean or his Amsterdam houseboat, he’s built what he calls a “10-hour PPC week” through extensive automation and systematization.

This lifestyle was inspired by Tim Ferriss’s “The Four Hour Work Week,” which Rooijmans discovered at a conference in 2004. The lesson? Proper systems and automation not only prevent mistakes but also create freedom.

Final thoughts

Even experts who have built their careers on automation and efficiency make mistakes. The difference between a career-ending error and a learning opportunity often comes down to how quickly problems are identified, how transparently they’re communicated, and how thoroughly the root causes are addressed.

For PPC professionals at any level, Rooijmans’ experience offers a clear reminder that shortcuts in processes eventually become obstacles to success. Whether it’s ignoring certain types of emails, skipping onboarding steps, or failing to monitor conversion tracking, these seemingly small oversights can cascade into significant problems.

The good news? With the right monitoring tools, clear processes, and a culture that treats mistakes as learning opportunities, even serious errors can be resolved while maintaining client relationships and professional growth.

Search Engine Land is owned by Semrush. We remain committed to providing high-quality coverage of marketing topics. Unless otherwise noted, this page’s content was written by either an employee or a paid contractor of Semrush Inc.

Anu Adegbola has been Paid Media Editor of Search Engine Land since 2024. She covers paid search, paid social, retail media, video and more.

In 2008, Anu started her career delivering digital marketing campaigns (mostly but not exclusively Paid Search) by building strategies, maximising ROI, automating repetitive processes and bringing efficiency from every part of marketing departments through inspiring leadership both on agency, client and marketing tech side. Outside editing Search Engine Land article she is the founder of PPC networking event – PPC Live and host of weekly podcast PPC Live The Podcast.

She is also an international speaker with some of the stages she has presented on being SMX (US, UK, Munich, Berlin), Friends of Search (Amsterdam, NL), brightonSEO, The Marketing Meetup, HeroConf (PPC Hero), SearchLove, BiddableWorld, SESLondon, PPC Chat Live, AdWorld Experience (Bologna, IT) and more.

For years, marketers measured digital success through impressions, backlinks and clicks. If you ranked high in search results and won the click, you had visibility and control of the funnel. But that landscape is already shifting.

Large Language Models (LLMs) like ChatGPT, Claude, Gemini and Perplexity are rapidly becoming the first place decision-makers go for answers. These systems don’t return a page of links; they generate a synthesized response. Whether your brand is included, or ignored, in that answer increasingly determines your relevance in the buying journey.

This changes the marketer’s playbook. Visibility is no longer only about ranking on Google. It’s about whether you’re present in AI-generated responses, how you’re framed, and what sources are credited. In this new paradigm, being mentioned is the new click.

The challenge for marketers isn’t simply tracking this new set of KPIs. It’s knowing how to interpret the signals and translate them into action. Let’s look at four core AI KPIs: mentions, sentiment, competitive share of voice and sources. We will explore how each can directly shape strategy.

Mentions: The visibility test

The first KPI is the simplest: how often are you mentioned inside LLM responses? If you’re absent from common category or evaluation queries, things like “top SaaS tools for analytics” or “best project management platforms,” then you’re essentially erased from the conversation before it begins.

But mentions are more than a vanity metric. They are a diagnostic tool. Patterns in where you appear, and where you don’t, can tell you which parts of your content strategy are resonating and which areas need reinforcement.

• Making mention usable: Break mentions down by type of query. Are you showing up in broad “what is” or “how to” questions, or only in head-to-head competitor comparisons? Are you included in trend discussions but missing from buying-decision queries? That breakdown highlights where to expand your authority.

If mentions are low in early-stage educational queries, invest in thought-leadership content that positions you as a voice in defining the category. If mentions are absent in solution-oriented queries, build assets that explain your differentiators more clearly. Mentions are the first signal of where your brand is visible, and where it’s invisible.

For marketers, mentions are the equivalent of oxygen. Without them, everything else is moot. With them, you can begin to shape how buyers see you.

Sentiment: The market’s echo

The second KPI is sentiment. Being mentioned is good, but how you’re described is what really sticks. LLMs add qualifiers to their responses based on available information: “fast,” “trusted,” “expensive,” “hard to use.” These adjectives reflect the narrative that exists in the data the model has absorbed.

• Making sentiment usable: Capture the language used around your brand. Track whether descriptors skew positive, neutral or negative. Note recurring themes — are you consistently framed as “enterprise-grade” but also “complex”? Are you praised for “innovation” but dinged for “cost?”

Negative sentiment highlights messaging gaps to address. If you’re framed as costly, consider publishing ROI calculators, pricing comparisons or case studies that show value delivered. If you’re seen as complex, invest in content that simplifies onboarding stories or customer success examples. Positive sentiment, on the other hand, shows you what narratives to amplify. If you’re consistently described as “trusted,” weave that trust theme into campaigns, analyst briefings and customer storytelling.

Sentiment analysis transforms LLM outputs into a real-time market perception barometer. For marketers, that’s invaluable. It gives you a constant read on how your positioning is landing without waiting for lagging indicators like surveys or analyst reports.

Competitive Share: The benchmark that matters

Mentions and sentiment don’t mean much without context. The real question is: how do you compare to your competitors?

Competitive share of voice is about measuring your brand’s presence in LLM responses alongside peers in your space. If you’re mentioned in 30% of relevant queries, but your top competitor appears in 70%, you’re playing catch-up. If you both appear equally often but their sentiment is glowing while yours is flat, they’re winning the perception battle.

• Making competitive share usable: Track not only how often you appear relative to competitors, but also the nature of those appearances. Which types of queries favor them over you? Which attributes are assigned to them versus you?

These insights turn into a battle map. If competitors are dominating certain categories of questions, that points to content and messaging investments you need to make. If their sentiment is consistently stronger, it suggests you need to double down on proof points or sharpen your differentiators. On the flip side, if you’re leading in areas they’re weak, that’s a narrative advantage you can emphasize in campaigns.

For marketers, competitive share is a strategy guide. It shows where you need to defend, where you can attack, and where you’re already winning.

Sources: Who the AI trusts

The final KPI is sources. Mentions tell you if you’re in the story. Sentiment tells you how you’re framed. Competitive share tells you how you stack up. But sources reveal who the AI trusts to tell the story.

When an LLM cites a competitor’s whitepaper or an industry analyst’s report rather than your own content, it’s a clear signal: you’re not seen as the authority. Conversely, if your blog post or research study is the cited source, you’ve secured a position as the trusted voice.

• Making source insights usable: Audit which domains and documents are being cited when your category is discussed. Are trade publications showing up more than your own site? Are competitors’ research reports being favored?

This is where content engineering comes into play. If you want your sources to be cited, they must be comprehensive, structured and credible. Think FAQ-style pages, data-driven reports, or clearly attributed expert commentary. By publishing content that AI can recognize as authoritative, you shift from simply being mentioned to being the foundation of the answer.

For marketers, this is the ultimate form of influence. When your resources are the citations behind the AI’s output, you control the conversation.

From signals to strategy

The temptation with any new metric is to build elaborate frameworks and dashboards. But the value of AI KPIs lies less in the infrastructure and more in the insights.

Mentions highlight visibility gaps. Sentiment exposes how you’re really perceived. Competitive share shows you where rivals are winning ground. Sources reveal who has authority.

Together, they form a compass. They help highlight performance and point you toward action:

• Fill gaps with new content.
• Reframe narratives with stronger proof.
• Defend share with sharper positioning.
• Earn trust by publishing resources built to be cited.

Marketers who use AI KPIs this way will be able to get ahead in the AI era, and they’ll actively help shape it.

Why acting now matters

It may feel early. The tooling isn’t standardized, and there’s no polished dashboard that marketers can log into and get all this in one view. But that’s precisely why early movers have the advantage.

Think back to the early 2000s, when SEO was still experimental. The brands that learned to optimize before the playbook was written ended up owning search visibility for years. We’re at the same moment now with AI KPIs. Waiting for the tools to catch up means letting competitors set the baseline while you play defense.

The actions don’t have to be complex. Even a lightweight process like running a set of prompts, logging responses and looking at mentions, sentiment, share and sources over time yields intelligence that can shape marketing and content strategies right now.

Conclusion: Mentions as strategy

The rise of LLMs doesn’t eliminate the value of clicks, impressions or backlinks, but it does redefine what visibility means. Increasingly, your brand’s story is being told inside AI-generated responses long before a buyer reaches your website.

That’s why these KPIs matter. Being mentioned is the new click. But the real advantage comes not from counting those mentions, but from using them to make smarter decisions, closing visibility gaps, reframing perception, benchmarking competitors, and owning citations.

For marketers, this is about translating AI signals into strategy. The brands that learn to do this now will have a better chance to survive the shift to AI-driven search.

At Brightspot, we’re helping organizations navigate that shift — turning AI insights into actionable strategy that keeps their brands visible, trusted and ahead of change. Learn more at brightspot.com.

Automation has shaped PPC for decades, and the landscape keeps shifting.

I’ve seen that evolution firsthand, from helping build the first AdWords Editor to developing early Google Ads scripts and writing about automation layering.

Now we’re entering another major transition. 

As AI changes how we search and get answers, it’s also transforming how automation itself gets built. 

And this time, the momentum isn’t coming from ad platforms like Google – it’s coming from AI companies like OpenAI.

Until recently, AI mostly helped with human language tasks like writing ad copy, summaries, or reports. 

But the latest generation of LLMs can increasingly generate computer language too, including the software and workflows that streamline how we work. 

At OpenAI’s DevDay in San Francisco, the company introduced AgentKit, a new way to build AI that can take action.

It marks the start of a phase where the automation mindset that powered PPC optimization can extend far beyond campaigns and into entire workflows.

Imagine if AI could handle your everyday busywork

Picture this:

• A client sends a CSV with weekly results, and before you’ve even opened the email, the file is saved to the right folder and added to your dashboard.
• A client asks for a meeting – AI checks your calendar, drafts an agenda, and schedules it.
• You start writing new ad copy with AI, and the system automatically pulls your brand guidelines and checks for tone and compliance.

This is all possible today, and you don’t need an engineering degree to make it happen. 

If you can define how your work is broken down into distinct tasks, you can create an agent that does those steps for you.

Dig deeper: 4 ways to connect your ads data to generative AI for smarter PPC

What agents really are

An AI agent is a smart helper that can figure out what needs to happen and then take action using connected tools.

Software has historically been built around deterministic steps. If X, do Y, else do Z. It’s predictable, but inflexible. 

And it requires humans to define every possible scenario that should be covered, which makes writing a helpful program time-consuming and difficult.

But just like an LLM is flexible in how it answers your questions, it can use that flexibility to automatically figure out a reasonable next step to complete a task. 

Instead of replying with text, agents can reason through steps, call APIs, and perform tasks.

I’ve explained early versions of this before: 

• You ask ChatGPT for restaurant ideas while planning a trip.
• It suggests a few places.
• It then uses an app like Resy to book the reservation.

That’s what an agent does: it can understand your intent and take a real-world step.

This concept builds on earlier OpenAI features, such as GPT Actions and function calling, which gave models controlled access to outside data. 

Agents are the next evolution – they combine reasoning with execution, meaning they can plan and act in the same flow.

Now, think about that in PPC terms. 

An agent could pull campaign data, summarize results, and even reference brand or policy docs before generating compliant creative. 

That’s a big step up from traditional “AI writing assistants.”

Dig deeper: AI agents in PPC: What to know and build today

From coding projects to five-minute builds

AI agents aren’t a new idea. 

Many marketers, myself included, have experimented with them for more than a year, but it used to take a lot of technical work. 

About a year and a half ago, I built an agent based on the two books I’d written that could answer questions in my tone and reference my ideas. 

I used LangChain, one of the first frameworks for connecting large language models to data and tools. It worked, but it wasn’t quick. 

I had to learn vector databases, RAG, and several other moving parts to get it working – not something most PPC pros want to tackle on a Monday morning.

Since then, several companies have made it easier to build agents like these, and some even feature them with a digital clone of a person, such as HeyGen. 

But when OpenAI introduces a way to create agents, I pay attention – and that’s what they did with AgentKit. 

It brings a visual interface for building agents directly on the platform of the most used chatbot.

What used to take hours or days of development can now be done in minutes, and you don’t need to know how to code.

AgentKit: ‘Zapier for AI’

AgentKit is OpenAI’s new toolkit for creating agents that can connect to tools and take actions through those tools. 

It’s a visual builder where you link services like Gmail, Dropbox, or Slack, and describe what the agent should do using tools you already use every day.

If you’ve ever used Zapier, n8n, Make, or Rule Engine, the concept will feel familiar: you connect blocks in sequences that represent what you want to happen. 

But because a flexible AI model sits at the core of these flows, AgentKit is different – it can use reasoning instead of rigid rules. 

If that sounds scary, you can add a simple human-in-the-loop approval step to any flow.

Instead of “If X happens, do Y,” you can say, “If a client sends a campaign report, summarize it and save it to the right folder.” 

The AI figures out how to do that by making reasonable requests that help it understand what you mean by vague instructions like “the right folder.”

For PPC marketers, this opens the door to automating work around campaigns (think reporting, documentation, and creative preparation), without waiting for a platform feature or a developer.

The unsung hero: Model Context Protocol (MCP)

Under the hood, much of the power that enables agents to take action comes from the Model Context Protocol, or MCP. 

It’s not brand-new, but it’s the key piece that makes all of this work.

MCPs are the connectors that let agents talk to your tools or data in a structured way. 

If you think of APIs as the connectors of the web, MCPs are similar, but built as a standard that any LLM can use. 

Some are built by OpenAI, like the connectors for Dropbox or Gmail. 

Others come from third-party developers, like Box. 

And you can create your own to connect private data or internal systems.

You can think of it this way: MCPs are the plumbing. AgentKit is the faucet.

The plumbing defines what data can flow where. The faucet is how you turn that into something usable.

Without MCPs, an agent would be like a brilliant intern with no logins to any of the systems they need. 

With them, the agent can safely use your data and tools with clear permissions.

Dig deeper: How Model Context Protocol is shaping the future of AI and search marketing

MCPs in plain terms

If this still sounds abstract, think of an MCP as a menu of what an AI can do inside a given flow.

For example, the Google Ads MCP currently includes actions like:

• Search for entities.
• List connected customers.

That’s it for now. It can read data, but it can’t change bids or create ads yet. 

That limitation is a good illustration that MCPs don’t open the door to entire systems for an LLM to go wild. 

Instead, they provide a defined set of capabilities created by the MCP developer. 

It’s an important guardrail. And even with MCPs that offer broader capabilities, you still control exactly which actions your agent can access when you integrate them into a flow.

Even in this early state, it’s a clear preview of how AI might eventually interact with Google Ads data through well-defined, secure interfaces.

Example: A ‘brand-safe ad assistant’

Here’s what this looks like in practice. 

Imagine you want an AI assistant that writes Google Ads while automatically following your brand voice and legal disclaimers. 

In AgentKit, you could create an agent with two connected tools: 

• Dropbox, where your brand guidelines live.
• A vector store with your agency’s tone and policy docs.

You could then ask the agent to “write new RSA headlines for our fall campaign using our style and disclaimers,” and it would connect with the right data to complete the task. 

Behind the scenes, it reads the files, extracts the rules, and generates compliant ad copy. You still approve the final version, but the prep work is automated.

It may sound simple, especially since you can already do this with a custom GPT, but it shows how these building blocks can be expanded. 

For example, you could integrate an MCP for your email platform and have the agent send a client an approval request for the creatives it generated.

Connecting data sources in AgentKit

Here are the steps to create an agent connected to the two data sources mentioned above. 

In Agent Builder, click the + icon next to Tools to give your agent a new capability, such as connecting it to an MCP.

Choose an existing MCP, like the ones shown here, or connect a custom MCP by clicking + Server.

You can also add a file search capability and select the files to include directly in the pop-up dialog.

Now you can interact with the agent to see how it uses its new abilities to produce better answers and, where enabled, how it uses other tools to take actions.

Dig deeper: How to get smarter with AI in PPC

Why this shift matters for PPC

If you’ve been in PPC for a while, you’ve seen this script before. 

We went from manual optimizations, to automated rules, to scripts, to automation layering – and each wave changed the skill set needed to stay ahead. Agents are the next wave.

Instead of writing scripts or building workflows with APIs, we’ll soon describe them in plain English and let AI generate the logic. 

That amplifies what marketers can do. 

The core skills stay the same – strategy, measurement, and judgment – but the way we build automation is about to get much faster, more flexible, and far more accessible.

The current tools for building AI agents are still early. 

Setting up an MCP takes some configuration, and the Google Ads connector is limited to reading data. 

But the potential is clear: AI will move beyond generating text to running workflows, checking rules, and getting work done.

If you want to stay ahead of this shift, start small. 

Experiment with simple automations that connect your email, files, or reports. 

Learn what agents can and can’t do yet. 

Just as marketers who adopted scripts early will be the ones setting the standard later, those who learn this now will be the ones setting the standard later.

Dig deeper: Agentic PPC: What performance marketing could look like in 2030

Search Engine Land is owned by Semrush. We remain committed to providing high-quality coverage of marketing topics. Unless otherwise noted, this page’s content was written by either an employee or a paid contractor of Semrush Inc.

Anu Adegbola has been Paid Media Editor of Search Engine Land since 2024. She covers paid search, paid social, retail media, video and more.

In 2008, Anu started her career delivering digital marketing campaigns (mostly but not exclusively Paid Search) by building strategies, maximising ROI, automating repetitive processes and bringing efficiency from every part of marketing departments through inspiring leadership both on agency, client and marketing tech side. Outside editing Search Engine Land article she is the founder of PPC networking event – PPC Live and host of weekly podcast PPC Live The Podcast.

She is also an international speaker with some of the stages she has presented on being SMX (US, UK, Munich, Berlin), Friends of Search (Amsterdam, NL), brightonSEO, The Marketing Meetup, HeroConf (PPC Hero), SearchLove, BiddableWorld, SESLondon, PPC Chat Live, AdWorld Experience (Bologna, IT) and more.