How connected data powers better consumer insights

Brands have more consumer data than ever, yet they still have questions. 

Marketers can see the number of consumers who clicked, engaged and bought. But even with more marketing analytics at their fingertips, it’s still not easy to answer the big question: Why did the campaign perform this way? 

Zappi’s 2025 Connected Insights Imperative found that 41% of marketing and insights professionals cite data fragmentation as the biggest barrier to using insights effectively. 

There’s more information than ever and marketing analytics can answer important questions, but learning from one question isn’t always connected to what the organization learned last month. 

An ad test can tell marketers how consumers respond to a particular piece of creative. But when that research becomes part of a larger, connected body of evidence, marketers can ask a different question: What have we learned across our advertising about what works, for whom and why?

That accumulated knowledge gives marketers something to work with before the next campaign begins. They can use previous evidence to develop a hypothesis, then look for evidence that supports or challenges it.

In this article, I'll look at what connected data is, the different sources brands can bring together and how connected data can help turn those signals into better business decisions. 

Marketers rarely make decisions based on a single source of consumer information. Connected data makes it possible to examine those different sources together, so teams can understand how one piece of evidence relates to another and build on what they already know. 

Definition

Connected data is data from multiple sources that is integrated with enough context to identify relationships between the signals found within that data, which can be used to answer shared business questions. 

For insights or marketing teams, consumer data integration can bring together different sources of consumer insight, including innovation research, advertising research, social and digital data and brand health tracking. Instead of analyzing each source in isolation, teams can compare what they're learning across them.

Context makes the connections meaningful. Marketers need to know whether they're comparing consistent measures and audiences, what each signal represents and which business question they're trying to answer. 

Say advertising research shows consumers respond strongly to a brand character or celebrity choice that’s appeared across a series of ads. That repeated, positive signal could indicate an approach worth repeating — and worth including that character in the next big campaign. 

Common data sources

A connected data ecosystem can include several types of consumer evidence. 

Innovation research Insights from consumers that focus on the development of new products and concepts. Think: New flavors, packaging or entirely new products.

Advertising research Examines how consumers respond to ad creative and campaigns. Think: Emotional response, brand recall, message clarity. Social and digital data Captures how consumers behave and respond outside structured research settings, including conversations, engagement, search behavior and other digital signals.

Brand health and tracking 

Shows consumer perceptions of a brand and how they may change over time, including measures such as awareness, consideration, preference and brand associations.

These sources become most useful when learning travels between them. Each source adds evidence marketers can use to test, challenge or refine what they think they know.

Good marketers don’t expect a dashboard to tell the story. They come to the data with a question and use the data to learn from it. 

Connected data gives marketers more evidence to work with and lets them build on what another team learned before. 

This flips the script on what consumer insights was traditionally used for:project-based testing. On Zappi's Inside Insights podcast, CIO and founder Steve Phillips describes the traditional model with an old adage: "companies can use insights like a drunk uses a lamppost, more for support than illumination. Insights teams are brought in after an idea has been developed to test it rather than using existing consumer knowledge to help shape the idea from the beginning.”

Connected data creates an opportunity to reverse that process with:

Better decisions

While a single research project can answer one question, connected data helps marketers understand that answer in context. 

Suppose innovation research shows that consumers have leaned towards more spicy flavors across past research. This could indicate that their specific target audience would like to see more of that in the brands portfolio, and be something worth exploring across their line of products. 

Now the team has a hypothesis worth testing. 

That question leads to a specific decision. Rather than abandoning ideas consumers like, marketers can test whether it’s something worth exploring further.

Faster answers

With connected data, not every new question starts with a blank page.

When research is standardized and connected over time, marketers can use the existing knowledge base to establish a starting hypothesis.

McDonald's calls this “getting a head start.” Rather than evaluating every new innovation in isolation, the team mines its historical research first to understand what it has already learned from previous successes and failures.

“I always call that getting a head start. Why wouldn't you want a head start with all this knowledge that we had before? It's going to make your outcome better”

- Matt Cahill, Senior Director, Consumer Insights Activation, McDonald's

That historical knowledge gives the team a starting point for its next hypothesis. Instead of asking only “will consumers like this new idea?” marketers can start with what previous research already taught them and build on that research. 

Fewer silos

Data fragmentation doesn’t support this type of learning. 

Phillips gives a simple example. In a traditional project model, a marketer might ask, Will this new ad campaign work for young men in Thailand? Insights conducts the research, answers the question and the data may end up sitting in an Excel file or PowerPoint after the immediate decision has been made.

This challenge extends beyond consumer research. Nielsen's 2025 Annual Marketing Report found that only 32% of global marketers measure media spending holistically across traditional and digital channels. Nielsen points to data issues, weak tools, too many vendors and a lack of transparency among the challenges marketers face when calculating cross-media ROI.

Fragmented evidence makes it harder for marketers to understand how one result relates to another. 

But if five years of advertising research has been collected systematically, marketers can ask a more powerful question before developing the campaign: What type of advertising works for young men in Thailand?

Existing evidence can inform the brief. The team can develop a hypothesis based on what it already knows, test the new work against it and add what it learns back into the system.

That's the difference between accumulating research and accumulating knowledge. 

As I’ve touched upon throughout this article, analyzing connected data can help teams identify relationships across a larger body of evidence. 

Sometimes that means comparing results across campaigns or concepts. Other times it involves meta-analysis or predictive models that identify relationships across hundreds or thousands of data points. 

These patterns give marketers evidence to interpret the data and invite new questions. Here’s a few examples of connected data in practice:

Identifying patterns

PepsiCo uses connected data to look beyond the performance of individual ads. Jane Wakely, then Global CMO at PepsiCo, described using years of Gatorade campaign research across multiple markets, along with competitive data, to identify what makes the brand's strongest advertising distinctive. She shares:

“The real value in raising the creative bar is when you begin to get to that meta insight, when you connect the dots, when you see a body of work.”

For Wakely, learning is one of the advantages of scale: a large organization can draw on years of accumulated evidence rather than treating each campaign as a fresh start.  

Predicting future outcomes

Historical patterns can also help marketers develop better expectations about what may happen next.

McDonald's, for example, can analyze previously tested concepts to understand which attributes have historically driven interest and purchase. Its connected dataset doesn’t guarantee that consumers will respond the same way to the next shake or McFlurry, but it does give clues for what the team should investigate next.  

“For a new shake flavor, I analyze the drivers of interest and purchase in all the shakes we’ve tested before. I can see how consumers play those concepts back, and what they want us to do differently. There’s a lot I can do easily with the data set.”

- Matt Cahill, Senior Director of Insights Activation, McDonald’s

Predictive analytics takes this further by using historical relationships to estimate likely outcomes. Models can identify combinations of signals associated with specific outcomes and use these relationships to estimate how new concepts, campaigns or ideas may perform.

Prediction doesn’t eliminate uncertainty, but it does help marketers decide what evidence they need before making a decision. 

Improving marketing performance

Connected data analytics creates a feedback loop between what marketers expect and what consumers do. 

Imagine that your previous advertising research shows a specific message that consistently produces a strong consumer response. The team hypothesizes that emphasizing that message in its next campaign will strengthen a specific brand association. Then, the creative tests well, but additional brand tracking shows little movement in that association. 

The team now has a discrepancy to investigate. Did consumers take away a different message? Was the brand association already strong? The answers can shape the next round of research and creative development. 

Connected data analytics becomes more valuable over time as each test builds on the last.

A connected insights system has the bigger job of helping an organization build on what it's learned.

Technology enables that connection, but the most mature organizations also embed insights into everyday decision-making through cross-functional collaboration and continuous feedback loops. 

The goal is to bring consumer understanding into the process early enough to shape the brief, creative or innovation, rather than asking insights to validate decisions at a stage gate. 

Types of insights to include 

A connected insights system brings together learning from several areas, including: 

• Innovation insights help teams understand consumer needs, which ideas resonate and what drives interest in new products or propositions. 

• Advertising insights help marketers understand how consumers respond to creative, which messages resonate and why some advertising performs better than others.

• Social and digital insights add evidence from consumer conversations and behavior outside structured research. They can surface emerging needs, changing language or unexpected responses that give marketers new questions to explore through research.

• Brand health and tracking insights show how brand perception evolves and provide further evidence about whether marketing activity is associated with the outcomes teams expected. 

A question can start in one area and continue through the others. Social data might surface an emerging need; innovation research can explore it; advertising research can test how the idea is communicated and brand tracking can show what changes over time. 

Stakeholders who benefit

Connected insights go beyond insights teams. 

While insights teams teams can use previous learning to develop and evaluate new ideas, marketing leaders can also use that interpretation to decide where to invest and what may be worth promoting next and business stakeholders can gain greater understanding of their customers and what works for the business. 

That cross-functional element also matters because insights teams aren’t always centralized. Zappi's 2025 research found that only 23% of organizations have a dedicated insights department or function, while one in three had centralized or restructured the function in the previous year. 

Fortunately, teams don’t need to sit in the same department to build on the same consumer knowledge. When previous learning is accessible across the organization, insights can shape work earlier rather than being called in once a decision is nearly made. 

Common mistakes

Connected data can give marketers a stronger foundation for decision-making, but not all of it is relevant. The system has to produce learning marketers can trust and use. These three mistakes can undermine that goal. 

1. Looking for patterns without a hypothesis Large connected datasets can reveal plenty of interesting correlations, but that doesn’t mean it’s useful. Start with a business question, use the data to test it and be prepared to adjust. As experiments stack up, you’ll find what matters.

2. Connecting data that isn’t comparable Finding meaningful patterns across research depends on consistency. 

If teams use different audiences, methodologies, questions or measures from one project to the next, it's difficult to know whether differences reflect a real change in consumer response or simply a change in how the research was conducted. Standardized frameworks make it possible to compare like with like and build reliable learning over time. 

3. Treating every project as a series of one-off projects

Zappi's 2025 research found that only one in three organizations manage consumer insights systematically. 

An ad hoc project isn't inherently a problem; sometimes marketers need an answer to a specific question. The lost opportunity occurs when that learning disappears into a deck or spreadsheet rather than contributing to the organization's knowledge.

Phillips' lamppost analogy captures the difference. Insights can be brought in at the end to support a decision that's already been made, or accumulated consumer understanding can illuminate the path before the decision is made.

Connecting data creates the foundation, but data alone doesn’t offer insight. Metrics tell you what was measured and analytics can reveal patterns or relationships. The insight comes from interpreting that evidence in context: why it matters, how it changes your understanding of the consumer and what the business should investigate or do next.

Zappi's connected insights approach connects consumer learning across advertising, innovation and brand tracking so teams can build on previous research rather than treating each study as a standalone answer. Standardized frameworks make results comparable and centralized access gives teams a growing body of research to learn from. 

PepsiCo's Stephan Gans describes the result simply: 

“The organization gets ‘smarter & smarter over time’ by connecting data across brands, countries and categories to generate meta-learnings.”

The real value of connected data isn't having more information in one place. It's about making sure what your organization learns today informs the next decision.

Connected research gives marketers a starting point. Previous learning can shape the hypothesis, new research can challenge it and what the team learns becomes part of the evidence available for the next decision. 

That's how individual research projects become organizational knowledge. 

Connected insights make that learning accessible and reusable, so consumer understanding can shape the next brief, idea or campaign rather than simply validate it after the fact.

In short: The advantage isn't having more data. It's getting smarter every time you use it.