Seeing through the Marketing Data Mirage - and how to avoid it!

Marketing performance looks healthy on the surface, but most marketers face a serious challenge: the data they use isn’t connected to real buyer behaviour. In DemandScience’s 2026 State of Performance Marketing, 750 senior marketing leaders were surveyed on their Performance Marketing strategies. 

The top-level takeaway is that most Marketing organizations operate within a “Marketing Data Mirage.” What is this mirage? How is AI making it worse? And how can Marketers avoid it? It seems that in a Marketing Data Mirage, the data signals are wrong. Marketers think things are good, but the signals they are relying on aren't tied to actual buying behavior. The biggest performance gaps, the report says, don’t come from media choices but from upstream inputs such as signal quality, content effectiveness, tool integration, and measurement reliability.

In other words:

• There are lots of clicks but no conversations.
• Intent signals look good, but don’t reflect buyer readiness.
• Impressions are high, but not for the actual ICP (ideal customer profile).

Also, Marketers are dealing with too many tools, some connected, others not. Getting a clear view of the customer across systems is challenging.

According to the study, 99% of Marketing organizations have experienced at least one mirage, and 70% have experienced two or more.

Here comes the AI

Here’s the not-so-surprising part: AI is amplifying this mirage. Most Marketers talk about being data-driven; making decisions on campaigns and content based on what the data tells them is important to their customers. But for many marketers, that’s not really what happens, and this study shows that.

Seventy-six percent of marketers said they create content that is not data-driven. Instead, they rely on assumptions, generic personas, and mimicking competitors. 72% said AI-generated content is hurting brand distinction.

There are two things to consider here. First, Marketers are using AI to create their content with little understanding of buyers beyond basic research. Yes, you can use ChatGPT or Claude to research your ideal customer, and you’ll get back some good information. But it’s general information that everyone is using to create similar content and campaigns. You won’t stand out.

You have to connect that research to real-world research on your current customers or by talking to people you know you want to reach. Which leads to the second point: understanding your buyers takes a lot of work.

You understand your buyers by talking to them, and by going back to listen to sales conversations, support calls, emails, and every other channel your customers use to communicate with you. This takes time, and it does require marketing to look at data beyond marketing systems.

Plus, as Sangram Vajre said in a recent LinkedIn post, it’s not solely marketing’s job to define your ICP. It’s a company-wide go-to-market (GTM) job, so leaving it to marketing, or to AI, isn’t going to work.

When you don’t have a clear understanding of who you are trying to reach, any content you create will be hit or miss (AI-created or not).

AI-driven systems or functionality within existing martech systems is another challenge. Again, though, it really comes back to the underlying data being used. For example, it’s critical to understand how a solution that scores intent signals using AI works. When you talk with martech vendors, AI capabilities can’t be black boxes; they must explain how the AI is working with customer data to identify the right signals across the buyer journey.

Cutting through the mirage

When Marketers are under pressure to drive pipeline, they often default to the same things: create more content (44%), add attribution or analytics tools (55%), or increase paid media spend (43%). These investments fall into categories where ROI is unclear, and they don’t help fix data problems if they are simply added to the pile of Marketing work without resetting strategies and activities.

To address the Marketing Data Mirage, Marketing must fix the underlying problems of tracking the right signals and creating content target customers are looking for across the right channels.

In this study, 86% of Marketers are chasing phantom “buyer intent” signals, and only 26% of signals convert to qualified leads.

In a LinkedIn post, Kerry Cunningham, Head of Research and Thought Leadership at 6sense, said:

Signals are dynamic. They're probabilistic. They can be strong or weak, clear or noisy. They require interpretation. And they demand action.

He means Marketers can’t take signals at face value; they have to understand where the signals are coming from and how they connect to give you a full picture. Cunningham said individual signals are not reliable indicators of buying journeys, especially given that B2B purchases are made by buying committees.

The report said signals are often considered phantom because of where and how they are collected: AI-powered scoring systems, behaviour tracking platforms, and intent data providers. But it’s not necessarily the systems themselves. It’s how they are used.

If 81% of respondents said that half or less of their content drives buyer engagement that leads to measurable outcomes, the answer isn't to implement another martech system or hire a content creation agency.

Best practice

There are a few things Marketers can do reduce or eliminate this data mirage. The report offers several best practices revealed by top performers.

Marketers first need to get a handle on the right data to truly understand their customers. This is about data quality and confidence, so verified sources and transparent collection practices are key.

It’s also about quality over volume. More data doesn’t necessarily equate to a better understanding of the buyer journey. If the data isn’t good, it leads to incorrect analysis or recommendations from systems. The same goes for content. Create content that resonates with customers based on quality data and signals, not on what you think they might like or what your competitor is doing.

The final one I’ll note here is to reduce the amount of tech in the martech stack. The less tools you use, the less work to tie the data they use and produce together, and the less chance of mistakes.

My take

More tools or content will not fix the problems marketers deal with today. This is a data issue that needs to be resolved before tools are added or content is developed. It’s a hard thing to say to marketers who are under so much pressure to deliver qualified leads to Sales.

Stepping back and taking the time to understand the customer and what data and intent signals will truly drive engagement and conversation is required. To do this, marketing needs to work with other members of the GTM team, including Sales, because there is a lot of customer data already available that can help drive better decision-making.