Building an actionable marketing data-asset

Online behavioral data faces quality issues

As online behavioral data becomes an increasingly important source of customer insight, it is often not treated as a valuable asset, except in digitally native companies. Changes in business priorities and shifts in how online data is collected and used mean it’s time to take online touchpoints more seriously. With cookies becoming less important, rising privacy concerns, and stricter browser controls, businesses need to adapt to stay competitive in the digital landscape.

This blog post will cover:

1. The challenges of improving online behavioral data and why clear ownership is key.
2. Why digital analytics is more about conventions and interpretation than exact science.
3. Steps to enhance digital analytics and stay competitive in a changing landscape.
4. Key questions to ask about your digital analytics to gain better insights.
5. The real-world impact of the increasing complexity of data measurement.
6. How to adapt as data measurement becomes less precise.

Overcoming ownership and quality challenges in online behavioral data collection

The biggest challenge in improving online behavioral data quality is the lack of clear ownership. Data Controllers or Marketers often don’t have full control over their data collection, with external agencies taking the lead. This can result in websites collecting data for purposes that don’t align with the company’s goals, sometimes without the Data Controller’s awareness.

Another common issue is focusing on quantity over quality. Many businesses fail to define key performance indicators (KPIs) or prioritize data collection based on clear objectives. Instead, they gather all kinds of data, leading to complexity and quality problems, which limits the insights that can be drawn. Data collection should be guided by well-defined KPIs, with quality checks after each stage to ensure the data is useful.

Building a valuable marketing data asset, particularly with online behavioral data, takes time. It needs to reach the same level of quality as other data sources, such as customer or transactional data.

Digital analytics is not an exact science, it is a collection of definitions and conventions

Over the past two decades, investments in digital advertising platforms have surged, especially in performance-driven channels. This growth is partly fueled by the misconception that everything in digital marketing is exactly measurable.

While there is some truth to this, it’s often misunderstood. First, the ability to measure everything is gradually diminishing. Second, digital analytics has always been more of an art than an exact science. “Exact measurability” is really just a collection of rules and definitions that shape how to interpret KPIs, metrics, and reports based on online behavioral data. These definitions are often set by measurement tool providers, not by the Data Controller.

Despite digital channels being a primary way customers interact with companies, digital analytics has not gained significant importance as a discipline—except in digitally native businesses. In many companies, digital analysts are placed under marketing or communications departments, where they are focused mainly on marketing metrics or website statistics. Rarely is digital analytics combined with customer data to provide meaningful business insights, even though digital services are the preferred channel for many customers. Imagine a CFO placing the business intelligence team in a similar position!

Defining relevant KPIs that drive business outcomes is more challenging than it may seem, given the vast number of data points available. Digital analytics platforms make it easy to start collecting data, but this can lead to blind spots if not properly managed. Many commonly used metrics, such as sessions, unique visitors, time spent on site, traffic sources, bounce rate, and conversion rate, seem straightforward but are based on specific calculation rules that may not align with business needs.

Implementing a web analytics platform with default settings is simple, but the metrics provided out-of-the-box may not reflect what truly drives business success. Vanity metrics, which offer little business impact, can distract from deeper analysis of the behaviors that influence outcomes both in the short and long term.

Organizations with multiple brands, web domains, or operations across different regions face even greater challenges. When digital analytics is not coordinated in a systematic way, it leads to data quality issues and inconsistencies across the business.

Key actions to improve digital analytics capabilities

As the need to understand digital service usage grows, so must investments in analytical capabilities and skills. To avoid common pitfalls and gain valuable customer insights, digital services should be integrated with other channels, data sources, and analyst expertise across the organization.

Before diving into specific actions to enhance online behavioral data quality and digital analytics, it’s important to manage expectations—whether you’re building in-house competencies or using external consultants. The way online behavioral data is captured is shifting from the front-end to back-end applications, where event flows are tracked in real-time. This shift requires a new approach to how digital services are built, with data collection integrated from the start, rather than as an afterthought close to the launch.

Since developer resources are often limited and backlogs full, adapting digital services to meet analytics needs should not be overlooked. Both front-end and back-end resources are required to make necessary changes to customer-facing touchpoints.

Here are four key action to consider, when advancing your digital analytics capabilities:

1. Customer Perspective, Not Just Browsers or Devices
   Combining online behavioral data with other customer data sources shifts the analysis from browser and device statistics to a focus on actual customers or groups with similar characteristics. (For more on this topic, see our earlier blog post: Combining Behavioral Data with Customer Data Requires Privacy Compliance and Technical Enablers.) When merging different data sources, it’s crucial to establish common taxonomies that apply over the long term. Products, event types, conversions, customer definitions, and segments should be harmonized to avoid misinterpretations or misleading conclusions. Isolating online behavioral data from other sources diminishes its value and reduces actionability.
   
   
2. Identifying Value Drivers That Impact Business Outcomes
   The first step in improving analytics is gaining control over KPIs, metrics, and definitions to reflect your business reality, not just standardized vendor interpretations. This means identifying the key drivers of business outcomes and fully understanding how your KPIs and metrics are calculated. Data analytics platforms typically provide basic statistics, but real business insights come from custom dashboards built on raw data in tools like Google Data Studio, Looker, Power BI, Qlik Sense, or Tableau. For example, website visits alone may not correlate with conversions, but certain behavioral patterns can be identified that predict outcomes with high probability.The first step in improving analytics is gaining control over KPIs, metrics, and definitions to reflect your business reality, not just standardized vendor interpretations. This means identifying the key drivers of business outcomes and fully understanding how your KPIs and metrics are calculated. Data analytics platforms typically provide basic statistics, but real business insights come from custom dashboards built on raw data in tools like Google Data Studio, Looker, Power BI, Qlik Sense, or Tableau. For example, website visits alone may not correlate with conversions, but certain behavioral patterns can be identified that predict outcomes with high probability.
   
   
3. In-House Analytical Skills for Daily Insights
   To drive meaningful business results, core analytical skills should be kept in-house, as close as possible to those with P&L responsibility. As insights from online behavior become a critical source of competitive advantage, analysts need a deep understanding of the business context. This expertise is hard to outsource or purchase as a service. Keeping these skills within your team ensures that analysts are aligned with business goals and can provide insights that drive actionable strategies.
   
   
4. Turning Data into Actionable Insights
   To unlock the full potential of your online behavioral analytics, the data must be actionable. This involves three key areas:
   • Legal Grounds and Purpose: Ensure that legal requirements and usage purposes are integrated into the data set from the beginning, maximizing the extent to which the data can be used for both insight and activation.
   • Value Driver Identification: Early on, identify and cultivate the behaviors that lead to expected business outcomes.
   • Channel Integration: Ensure that your data is integrated with the channels and touchpoints where your customers interact with your company, enabling seamless activation and engagement.

    

Most important questions to ask about your digital analytics

In this section, key observations are translated as actions to improve online behavioral data quality into five actionable questions. If you can answer these questions confidently, you’re on the right track to unlocking value from your data and staying competitive:

1. Do you understand the correlation and causality of drivers for business outcomes?
   Have you developed a solid understanding of what behaviors drive tangible outcomes, like sales or leads, over time? If the answer is yes, you’re ready to draw meaningful conclusions and take action. This means recognizing the role of different touchpoints in the customer journey. On a strategic level, this insight helps leaders assess whether they are under- or over-investing in digital services. On a tactical level, it enables actions like targeted communication to accelerate deal flow or recognize when existing clients are in buying mode.
   
2. Can you identify user intent from a visit or series of visits?
   One of the most valuable insights from online behavior is understanding user intent during a visit or multiple visits. To serve customers effectively, you first need to identify why they are engaging with your site. For example, a user searching for contact information who finds it immediately via an organic search has their intent fulfilled in a short visit. On the other hand, for a user likely to convert within the next 14 days, driving them toward a purchase and increasing the sale value through personalization would be the ideal outcome.
3. Can you differentiate behavior across different personas?
   Analyzing online behavior through segments like existing customers, potential customers, or other stakeholders helps refine your understanding of user intent. This segmentation allows you to provide more relevant content to each persona. To separate behaviors effectively, combining different data sets is essential. (For more on this, refer to our blog post: Combining Behavioral Data with Customer Data Requires Privacy Compliance and Technical Enablers.)
   
   
4. What actionable insights do you gain from each visit, and how do you act on them?
   Every visit, except for accidental clicks, typically has a purpose. To make your data actionable, it’s crucial to understand user intent and turn it into a follow-up action. Doing nothing is also a decision, particularly when you recognize accidental clicks that don’t require further engagement.
   
   
5. What data is being collected, and for what specific purpose?
   If you can’t answer this question easily, it’s likely your data sets contain unnecessary noise, collecting data without a clear purpose. Clarifying the purpose of each data point has two key benefits: first, it simplifies analysis by reducing irrelevant data, and second, it ensures compliance with data minimization principles, keeping you on safer ground regarding privacy regulations.

Concrete consequences of measurement becoming harder

There’s no doubt that, in the post-cookie era, data collection and analytics are becoming more complex, requiring an evolving set of competencies. We’ll explore these changes further in a follow-up blog post, but to give you a concrete idea of the consequences, here are five key areas affected:

1. Marketing Effectiveness Analysis Becomes Less Precise
   Attribution analysis, traditionally based on clicks, referral paths, and visits, is fading. Instead, statistical and probabilistic methodologies will become more common as tracking individual user behavior across websites becomes restricted by default. As these new methods evolve, there may be an increased bias toward last-click channels in conversion reporting.
2. Web Analytics Becomes More Complex
   In addition to marketing analysis becoming less precise, web analytics as a whole is growing more complex, requiring greater technical expertise. Fewer visits, events, and conversions will be measured, making it harder to capture comprehensive data. Addressing these challenges will involve adopting new measurement methodologies, which will require different skill sets, including front-end and back-end development, raw data engineering, and data manipulation.
3. Online Advertising Targeting and Retargeting Loses Efficiency
   Targeting and retargeting are generally less effective in the post-cookie era. While this may improve the online experience for users by reducing annoying retargeting ads, it forces marketers to rely more on first-party data to personalize ads across different channels. Traditional metrics like view-through and click-through rates are becoming obsolete, making historical comparisons less relevant.
4. Alternative Technology Development Speeds Up
   The industry is constantly reinventing itself, with server-side tracking emerging as a hot topic. Although data clean rooms were previously discussed as a solution, they haven’t solved the problem and have mostly been used to justify increasing ad spending. Predicting the future of these technologies is difficult, but ethical concerns and corporate responsibility will play a larger role in the coming years. Collaboration between advertisers, technology vendors, and publishers to share data will also reshape the digital advertising ecosystem.
5. The Illusion of More Control for End Users
   The idea that users have more control over their personal data is somewhat misleading. While browsers, platforms, and websites now offer more choices—largely driven by regulatory requirements—there’s also an opposite trend of technical workarounds that bypass privacy controls. Whether users truly gain control over their data remains to be seen. For now, it’s becoming harder for the average person to understand who collects their data and how it’s used.

How to prepare for a world where measurement might be less exact

In the face of growing uncertainty, two things are certain for all businesses, regardless of industry or size:

1. The importance of customer insight and business intelligence from digital services is only increasing as more processes shift to digital channels.
2. The accuracy of measurement is becoming less precise, while data collection is growing more complex with each passing day.

Given these developments, there are logical actions businesses can take to remain competitive in the future, despite the changing landscape:

1. Build Trust and Direct Customer Relationships
   Earning and maintaining customer trust is crucial. This trust can lead to broader permissions and even expectations for direct communication with your audience. Treat data as a valuable asset—a form of currency in exchange for customer engagement. Contact information, such as emails or phone numbers, will become more valuable for both direct and indirect communication. Log-in services also offer an advantage for data collection, as they provide more reliable data while complying with privacy regulations. Taking data-driven sales and marketing seriously means having a clear strategy to encourage users to log in and maximize the number of identified online users.
2. Flexibility in Data Collection and Processing Logic
   Adopting a technology-agnostic approach to online behavioral data collection will allow your business to switch tools or methods with minimal disruption. Changes in data collection practices—whether due to new regulations or technological advancements—are inevitable. While these changes may come as a surprise to many organizations, they are often predictable for those closely following industry trends. In addition to agility in data collection, being able to quickly adjust consent and privacy mechanisms is crucial. As data collection becomes more challenging, businesses should develop capabilities for running statistical or probabilistic analyses on cloud platforms to compensate for potential gaps in data. A well-structured data model and taxonomy are essential for the agility and speed needed in this new environment.
3. MarTech and Analytics Architecture Design
   When designing your MarTech and analytics architecture, think in terms of solutions, not isolated channels. Leverage common IDs available in platforms like Adobe Experience Cloud, Google Marketing Platform, or Salesforce Marketing Cloud. Prepare to handle raw event-level data on cloud platforms rather than relying on aggregated data from tool interfaces. Whenever possible, pseudonymize customer data and personal identifiers and pair them with online identifiers to ensure privacy. Also, minimize vendor lock-in by identifying critical components that deliver the most value to your use cases.
4. Workarounds and Hacks
   Server-side tracking is one option to regain control over data collection and distribution or to bypass limitations such as ad blockers or Intelligent Tracking Prevention (ITP). Workarounds can be employed responsibly or less responsibly. For instance, relying on non-cookie identifiers, like fingerprinting, may raise legal concerns but is a common practice with a long history in the industry. However, it’s essential to balance the use of these hacks with ethical considerations and compliance with privacy regulations.

Conclusion

In conclusion, as the digital landscape shifts with the decline of cookies, rising privacy concerns, and evolving regulations, businesses must prioritize the quality and ownership of online behavioral data. Digital analytics, though often misunderstood as an exact science, relies heavily on conventions that may not align with business needs. To stay competitive, companies should focus on well-defined KPIs, integrate behavioral data with other customer data, and build in-house analytical capabilities. As measurement becomes less precise, organizations must adapt their data strategies, invest in customer trust, and develop flexible data collection methods to navigate the increasing complexity of the digital ecosystem.