The pipeline construct in B2B marketing has a dual nature. On the one hand, it is a true reflection of how buying groups move through purchasing hardware, services, and software. Concretely, it is true that companies, departments inside companies, and individual decisionmakers must be first made aware of a problem; then understand that a category of solutions to that problem exist; become aware of the vendors offering these solutions; at some point engage with the marketing and sales resources of one or more of those companies; and finally transact. Once they are a customer, they continue to update their experience of the company, perhaps adding services based on other perceived needs.

However, in most cases, the buyers and influencers who make up the customer buying group are indifferent to or unaware of whether they are interacting with a vendor’s “marketing,” “sales,” or “customer success” teams. To them, there is a brand, and that brand either meets or exceeds expectations, or does not. They simply want the best product and service at the best price, with the lowest risk (no one got fired for buying Company A) and do not want to jump through hoops to do so.

The reflection of this customer-centric pipeline inside the typical vendor is distorted but still relevant. For a typical B2B vendor—think Microsoft, Caterpillar, Oracle, Goldman Sachs, GE, etc.—the pipeline is divided into discrete stages, each made of either leads or opportunities, with different values and forecasted close dates. Typically, a “lead” is marketing’s responsibility, and an “opportunity” is owned by sales—but it’s critical to note that to a customer, these categories are irrelevant. This “lead / opportunity” split is a legacy of how B2B marketing and sales has typically functioned: Marketing “generates demand” and sales “closes deals.” The best way to think about “generated demand” in a software system is as a “hand raiser”—someone who has poked their head above water and can now be pursued. That hand raiser “becomes” an opportunity when they have been nurtured and developed, and at that point, the opportunity will gain momentum and hopefully turn into real revenue. Of course, leads and opportunities are both abstractions and simplifications of what is really going on.

We all want to measure marketing ROI

ROI (return on investment) continues to be a hot topic for B2B marketing and sellers, for obvious reasons. An accurate ROI (one that is non-duplicative, counterfactual, and based on a financial outcome) is extremely useful, because it allows all investments to be traded off against one another, particularly at the marginal or “last dollar” basis. If my marginal ROI for paid social is 1.1, and my marginal ROI for events is 0.9, then I should increase my paid social budget and decrease my events budget. Critically, ROI as an outcome metric allows marketing to be traded off against any other investment—at least in theory.

B2C companies are arguably closer to an ROI view of the marketing world. For large consumer brands like Coke, marketing mix models (MMMs) are constantly updated to provide ROAS (return on advertising spend) for various channels. The curves these models output are then used to remix dollars up, down, and across the funnel to maximize some objective—usually total revenue. However, MMMs are slow and prone to omitted variable bias—meaning that lurking, unknown variables, if left out of the model, can drive unrealistically rosy assessments of marketing’s performance.

B2B companies can’t generally use MMMs to measure marketing’s effectiveness (some try, and they “sort of” work, but that’s a topic for another day.) The same structural dynamics that lead to a pipeline view of the world make MMMs—which depend on large volumes of frequent time series data, including daily sales and marketing spend by region—ill-suited for B2B, namely:

• Long sales cycles (months to years, typically)
• Large transaction sizes, few transactions (chunkiness)
• Complex buyer groups
• Poor data quality when humans are involved (events, field sales, partner channels)

The pipeline, rooted in a database view of the world, is both a cause of and a solution to measuring ROI for B2B firms. It is the cause of the problem when it is taken too literally—that is, that the “lead” is a real thing that someone “generated.”

At some point in the foggy history of corporate marketing, “marketing attributed sales” became a commonly used term. This probably happened when someone in sales asked someone in marketing what value they were providing, which, by corollary, meant how many leads were being handed off.

Now, we commonly speak of “marketing attributed sales” as those opportunities that started with a marketing-generated lead. This means, concretely, that some individual at a buying group filled out a form, and was then “nurtured” until ready for handoff to sales as a “sales qualified lead.” In some cases, sales has to accept the lead for it to “count”—as a “sales accepted lead.”

There are three problems with this way of looking at marketing value. First, it assumes that marketing generated all of the “value” of the lead that it generated. This overstates marketing’s impact. However, this isn’t the biggest problem: All of the other value that marketing creates “under the water” is missed, because it’s not a part of the marketing software / CRM software that has largely come to define the B2B marketing organization. Finally, once a lead is “handed off,” marketing’s role is cut off, leading to both double-counting (marketing and sales both want credit for the deal), and a somewhat toxic “what have you done for me lately” adversarial stance between marketing and sales.

These dysfunctions have real negative impacts. Marketing’s insistence on taking full credit for leads—understandable given its typical fight to show value—drives a bias towards lower funnel behaviors that might not optimize long-run growth. The inability or unwillingness to understand how marketing drives value for all opportunities—known or unknown—makes assessing true ROI impossible. Finally, the “hand-off” concept itself creates an us-them duality that is nonsensical to a customer, and, once again, does not accurately capture marketing’s role in driving value.

Conclusion

Marketing and sales have a common goal: to drive revenue. Yet the most common marketing, sales and CRM tools today pit marketing and sales against one another to claim holistic credit for each sale. True B2B marketing ROI is achievable with the right measurement approach.

Stay tuned for Part 2: How to measure B2B marketing ROI and subscribe to our monthly Consulting newsletter so you don’t miss an insight.

The post The problem with B2B measurement appeared first on Marketbridge.

Economic predictions for a mixed year of nominal growth largely held true, supported by injections of tariff uncertainty and massive AI investments, leading to an unbalanced market that’s basically the magnificent seven vs. the rest of the economy, 2025 was a wild ride. Overall, AI dominated the GTM landscape, broadening the market understanding beyond generative AI to agentic integration across the GTM stack.

What we failed to anticipate was the stark, almost painful, misalignment between the pace of tech innovation and the inertia of the enterprise. Providers delivered tools at lightning speed, but customers couldn’t keep up. Much like the economy in general, AI predictions at least fared okay, and in most cases, pretty good!

Making the grade

To assess the veracity and reality of these shifts, I looked at consensus predictions against in-market performance and commentary from 2025, assigning grade based on proximity to how close predicted outcomes came to reality. To simplify, I bucketed the analysis on three dimensions that determined revenue success:

• GTM model alignment: Traditional GTM models will be replaced by blended Hybrid GTM approaches
• Data-driven profitability: GTM must transition to data-driven, Intelligent Pricing strategies
• AI workflow challenge: AI will successfully automate seller administrative tasks at scale

GTM model alignment

Blended GTM models were not fully adopted, but the economic climate necessitated the collapse of GTM silos and the adoption of more agile growth models, driven by predictions focused on unity and efficiency (Forrester, McKinsey). The core prediction here, the transition to Hybrid GTM Models, was a strategic success of the year, while implementation struggled.

Blending GTM models earned an A, signifying market adoption, but the underlying goal of Organizational Alignment (under the RevOps umbrella) fell short with a C+.

GTM successfully moved toward a hybrid operating model but underestimated the difficulty of achieving true organizational unity and structural alignment required to execute it efficiently.

Data-driven profitability

Maximizing margin and improving sales economics were paramount, requiring innovative intelligence-based GTM levers (Bain & Company). The single greatest failure of 2025: the inability to capitalize on advanced profit levers due to data deficiencies.

The most ambitious prediction, Intelligent and Dynamic Pricing, fell short with a C grade, directly contrasting the success of the foundational prediction: Data and RevOps as the Foundation, which earned an A+.

The ambitious revenue-driving predictions were entirely contingent on the fundamental work of RevOps and data quality, reinforcing that basic technical integrity is the prerequisite for innovation. While the assumptions are correct and the direction clear, much like organizational adoption, Data has a long way to go to achieve its profitability promises.

AI workflow challenge

The most compelling prediction for 2025 was the transformative impact of AI (Deloitte, Gartner). The critical question was whether organizations could translate this promise into measurable, scaled success. The results here were split: AI Co-Pilots earned a resounding A, while the mandate to Scale AI Across the Enterprise lagged with a C.

2025 proved AI’s effectiveness as an augmentation tool (Sales Support), but it revealed significant bottlenecks in process management and change adoption necessary for enterprise-wide transformation, especially as it comes up against entrenched teams, processes and vendors.

Final evaluation

A clear narrative coming at the end of 2025 is that while investors and technology providers move forward with AI-abandon, and talks of a bubble have dissipated, GTM leaders are taking more cautious approaches and investing strategically. The year demonstrated that while AI and market shifts are accelerating, successful transformation is ultimately limited by an organization’s willingness to address difficult, systemic, and people-centric challenges (Scale, Pricing, CX).

GTM organizations are entering 2026 leaner and smarter, having successfully prioritized operational efficiency and technology consolidation. However, the clarity gained from 2025 confirmed that the biggest blockers aren’t technological advancements—they are systemic and people-centric.

The success for your 2026 growth roadmap hinges on closing the adoption gap, turning C grades into A grades. This means tackling the fundamental human challenges. As you navigate this next phase of GTM transformation, we’d love to connect to help bridge the gap between technology potential and revenue reality.

The post Report card: Grading 2025 GTM predictions appeared first on Marketbridge.

Quality can differentiate brands, drive loyalty and increase revenue. We’ve been talking about this for a while on LinkedIn, on our blog, and at events.

Last week we co-hosted with the Association of National Advertisers a one-day conference that gathered marketing leaders and practitioners to discuss why quality matters and how to move toward quality marketing and analytics. Below are 5 key takeaways for marketing leaders to evaluate and discuss internally.

Build audiences offline to better control who you’re targeting

“One of our biggest dangers is thinking about people as datasets,” said Chief Analytics Officer Andy Hasselwander during his quality marketing analytics session, but multiple sessions discussed why thinking about your target audience as big numbers (and not individuals) is a problem.

Multiple sources say 252,000 websites are created daily, and the number of viewable impressions and IP addresses vastly outnumber human beings on earth. According to Truth{set}, any two given data providers agree on what IP address matches a postal address at most only 14% of the time. Privacy is not an excuse for bad data, but marketers are getting duped thinking they’re targeting one person but reaching another. Marketing needs to start policing itself on quality—potentially by bringing the identity spine into the open rather than relying on blackbox, outsourced providers.

If you want better results, leverage data and PII to build offline audiences and “stay in the PII as long as possible,” according to Mark Pilipczuk from The Industrial Arts. Segment your ICP within your own PII data, and leverage a vendor’s database to build lookalike models for your target audience, hashed audience is uploaded to the publisher or ad platform (audiences drops due to match rates), and then the files are delivered with partner cookies.

Image credit: Mark Pilipczuk / The Industrial Arts LLC, © 2025 — used with permission.

This offline, more targeted audience almost certainly will outperform the third-party interest and intent categories available in DSPs and ad platforms for customer lifetime value (CLV) and return on ad spend (ROAS).

One lever advertisers can pull to improve impression quality is to ask their DSPs and ad platforms to provide data on refresh rates (higher is worse), sites with multiple advertisers in the same consumer view, and sites with high ad-to-content ratios.

Marketing and analytics should embrace uncertainty

Understanding what we know and what’s still uncertain supports good decision making. Yet many in both marketing and analytics are hesitant to state when we don’t have a definitive answer.

In statistics and modeling, error bars show the variability of data. When MMM reports out a cost per acquisition (CPA) or ROAS, typically only the mean or median value is reported. But that’s where you get into trouble.

In the image below, the estimate for TikTok’s CPA is €41. If Instagram Reels’ estimate CPA is €60, the marketing team may decide to shift budget to TikTok. But then in the next readout (and with more data), TikTok’s CPA is €80. Now marketing’s mad and doesn’t trust the MMM. But in reality, TikTok’s CPA is still within the confidence interval—error bars would’ve helped marketing make a more informed decision.

Analytics teams must help educate their stakeholders about error bars and confidence intervals so the organization can make better decisions.

Delivering ROAS or CPA without error bars doesn’t breed confidence, it breeds distrust.

Cultivate curiosity and come with a solution mindset

Humans are wired to collect data, but creating knowledge, driving insight and providing wisdom don’t happen automatically. So what makes a great analyst? Curiosity, clarity and capability are the core skills of a great analyst, according to Sravanthi Konduri from Navy Federal Credit Union. Often we equate degrees with skill, but building experience and knowledge is needed to drive insight (and eventually wisdom).

Cultivating curiosity within the organization is another matter. Organizations with a growth mindset support an analytical environment and aren’t scared of data and learning. Analytics teams can fail because of analysis paralysis and wanting to have the perfect answer, rather than collaborating with internal stakeholders.

A solution mindset is key for marketing analytics teams to partner internally and offer alternative approaches and solutions. Analysts should think like the GM of a business unit—understand the problems, where the question fits in and who would care, and delivering an answer in the context of why it matters.

How to combat too high and too low ad frequency

Often marketers worry about capping frequency for individual views to prevent waste. This has long been an issue, especially if you’re running a campaign across multiple channels and platforms. David Riva from The Trade Desk pushes for unified frequency control—rather than capping each individual placement, DSPs should support capping frequency across channels.

Another area of waste, according to Ray Van Iterson from the United States Postal Service, is the large group of people who see 1, 2 or 3 fewer impressions than needed to achieve the goal.

And the key question is: do you even know who those people are? Can you identify and target them differently or with additional inventory?

Understanding consumer journeys is essential and MTA isn’t dead

The announcement of cookie deprecation was overblown and yet many organizations stopped trying to understand individual consumer journeys. But reporting focused on campaigns, channels or business units is inherently biased.

Marketers should know what combinations of channels and which sequences lead to the best outcomes. This can be done for known, trackable touches as well as likely touches using probabilistic mapping. Adding up the small probability of seeing an ad on a given day in a specific DMA across an entire campaign can give a better picture of how channels work together.

That individual journey data also can identify if someone is less engaged than we expect. Marketers can then deploy a higher impact channel to achieve the goal.

Understanding the consumer journey at the most granular level of data possible arms marketers with an understanding of the impact of particular platforms or partners within a channel. Identifying the high engagement or high attention platforms and partners can provide optimization opportunities and drive better outcomes.

Want more insights?

If focusing on quality in marketing and analytics feels like a challenge, let’s talk. We’d love to hear about the roadblocks, share best practices and brainstorm solutions.

Complete the form below and we’ll connect to schedule time.

The post 5 key takeaways on quality in go-to-market appeared first on Marketbridge.

Marketing is defined as all customer-facing functions of the enterprise. Decisions about brands are made by people; acquisition, retention, and advocacy are all human choices. Despite speculation that autonomous AI agents will be the customers of the future, the current reality is that people drive growth.

However, modern marketing is increasingly mediated by technology. Analysts and marketers interact with data, processes, and colleagues through screens, often reducing customers to digital avatars. This trend risks losing the human element in marketing decision-making.

Marketing analytics has followed suit, becoming technical and quantitative. This has enabled precision and accountability, allowing marketing to better communicate with finance, justifying and explaining marketing’s contributions. This is a good thing.

Marketing analytics is the de facto quality department of the go-to-market function. The analytics team—or the analytical resources in the marketing team—are ultimately responsible for identifying data problems, revealing invalid inferences, and uncovering inefficient programs.

However, quality-focused analytics team must beware of disembodiment. Our data represent people, and people are very complex. Our representations of them and their actions are crude indicators of their latent properties. Losing empathy is a road to poor quality, and poor quality leads to long term brand decline and de-growth.

What the Toyota Production System teaches us

In the 1950 and 60s, Detroit dominated the U.S. auto market. Their cars were big and beautiful. The factories of the big three automakers were high-tech—management science was born in Detroit, drafting after the mass production successes of WWII. However, they soon learned that data, technology, and automation do not necessarily mean quality. In fact, layers of abstraction can hide defects. By the mid-1960s, Detroit was shipping lemons, and buyers were quietly getting fed up.

Enter the Japanese importers, led by Toyota. Taking advantage of labor arbitrage, Toyota delivered less expensive cars. They also filled an unmet niche need for fuel efficiency. However, their value proposition didn’t stop there: Buyers soon discovered that their cars were also better built, had fewer defects, lasted longer, and consequently, kept their value over time.

Toyota did not accomplish this via technology. Toyota’s approach emphasized quality through human-centric principles:

• Andon cord: Stop processes when defects are found
• Hansei: Learn from mistakes
• Just-in-Time: Request or provide only when needed
• Kaizen: Empower the workforce
• Genchi Genbutsu: Direct observation of problems
• Nemawashi: Open information sharing
• Genba: Visibility on the factory floor

Quality improvement in marketing analytics requires a similar focus on people and process transparency.

Five Principles for Quality in Marketing Analytics

These five principles have emerged over the past decade at the Marketbridge team. They are best practices, proven through countless hours of focusing on delivering the best quality. They share many of the same values of the Toyota Production System:

Truth Seeking (Hansei)

The word “analysis” comes to use from Greek: Ana (untie) and Leuin (knot). Said differently, working to find all of the holes and loops in something complex, and getting to truth. For marketing, this means analysts should challenge strategies, not support them. There are many ways that this value can be practiced in real life. Some common use cases include:

• True up CPA: Demand capture channels’ last touch CPAs are usually much too low. Truth seekers can decompose lower funnel channels’ effects into accretive, distribution tax, and duplicative, and then report on CPAs taking each of these effects into account.
• Seek Outliers: The most common explanation for an unexpected result is a data issue. For example, if a tactic shows an extremely low CPA—say, two standard deviations below the mean—in almost all cases there will be some error in data collection. Of course, sometimes these are real results, but a good rule of thumb is that 9 out of 10 two sigma outliers aren’t real (unsurprisingly.)
• Eliminate Double Counting: It is rare that the marketing contribution attained from summing reports from individual channel owners adds up to less than total revenue, but very common that it is more. This is understandable; managers want to claim as much credit as possible. It is up to truth-seeking analysts to ensure that totals foot and sum. This will cause hurt feelings temporarily, but this is the cost of truth seeking, and in the long run, trust will be established.

Humanization / Embodiment

Customers are humans; we use data to represent them. A high-quality team should flip their default view, and consider what each customer receives, not the aggregate marketing mix going out the door. Humanization of prospects and customers is a consistent challenge for analysts who are almost always a few levels of abstraction away from people. However, technology also provides us with tools to humanize—if we choose and use them carefully.

• Center Data on People: A Longitudinal Human Record (LHR) aggregates all known and probabilistic data about prospects and customers, enabling empathy at scale. It also powers multi-touch attribution, MMM, and UMM, along with Customer 360 dashboards and reports.
• Go Small: The typical data science workbook flow is to chain data transformation steps together to get to an end result—typically a data frame or a model. However, this approach can miss detail. The best “data detectives” examine individual customer records and follow their journyieys. For example, following a journey from pre-sale to sale to customer—with all of the touchpoints in between—can unearth insight or quality issues that would otherwise be missed. When these findings are scaled up, they can yield big results.
• Build Audiences from Real People and Events: Too much abstraction can quickly yield muddled pictures. While advertising technology is amazingly sophisticated, real signal can be hard to find. Fortunately, addresses are still real, and so are purchasing events. Building audiences from addresses and events is grounding, and yields better results than models built upon models.

Transparency (Nemawashi, Genba)

Reproducibility is a term we borrow from scientific research. When a researcher reports a finding, it is important that other researchers can come back to the work, and at a minimum, follow the steps from raw data to results. Generally, transparency drives quality because more people can see “inside the box” and find problems.

Technology is often seen as impenetrable, but this does not have to be the case. Choiceful adoption of transparent tools and methods can make data simpler, not more complex.

• Use Version Control Tools like Git and Github: Code (R, Python, SQL) and documentation (markdown, readmes) are not just stored and available, but are change logged. In other words, the code base—a manifestation of an organization’s IP—can we seen evolving over time, and nothing will be lost in anyone’s desktop folder.
• Use Workbooks and Notebooks: Workbooks (for example, Databricks) and Notebooks (for example, RMarkdown and Quarto) are visual mash-ups of code, narrative, and results outputs. They facilitate inspection and intuitive understanding.
• Adopt a Medallion Architecture: A medallion architecture is a loose framework that acknowledges three states for analytical data. First, unstructured “dirty” data sits at the Bronze level. Second, taxonomized and QA’s data that is used commonly sits at the Silver level. Finally, use case-ready data sits in scrubbed data frames. supports both formal and innovative use cases, with ad hoc analysis (Bronze), centralized clean and governed data (Silver), and use case-ready data frames (Gold). The key is that all data end up in the data lake—even one offs (in the Bronze layer.)

Ownership (Kaizen, Andon)

There is a temptation to think that technology, data, and automation require less human ownership—indeed, this is the primary idea behind technology driving productivity. Toyota realized that people owning technology drove quality, and that hasn’t changed. In a marketing team, this means that everyone should understand data, have access to it, and be able to perform analysis.

• Own (at least partially) Martech: Marketers and marketing analytics are deeply dependent on Martech. Concretely, most data problems originate in source systems, and, by corollary, are easier to fix there than to band aid later. The people who use the data generated by these systems should at least have a large seat at the table for Martech implementation and customization—and should ideally own the systems altogether.
• Advanced COE: Analytics centers of excellence (COEs) should avoid the guild mentality, and continually focus on only the most advanced use cases—yielding simpler analysis to the marketers themselves. In other words, they should be an innovation lab, not a walled off set of protected jobs. Marketers owning their own data and analytics is a good thing.
• Celebrate QA: Data mistakes are scary. There is a human tendency to want to hide problems, particularly when they were your fault. World class organizations know that admitting failure and then correcting it should actually be celebrated. This is the basis of Toyota’s Andon Cord. The best marketing analytics teams have targets of “bugs found”—and give shout outs to those who find them (even if they created them.)

Probabilistic Communication

Because we are working with people, we can’t know everything about them. This is one of the things that makes marketing fun. When technology really started taking off in the early 2000s, some thought we would start knowing everything about audiences and customers. If anything, the opposite has happened. There is so much noise in the system that we know less, not more. However, leaders want precision—whether in terms of ROAS (return on advertising spend) or CPA (cost per acquisition.) It is up to analytics leaders to not given them exact answers—but rather to train them that everything in marketing is about probability and confidence, and to help them make probabilistic decisions.

• Always Include Error Bars: Error bars show stakeholders that the mean value, while the most likely result, is not precise. They can then make decisions that account for the chance that a result is higher or lower than reported.
• Forecasts that Get More Uncertain: Weather forecasts end at 10 days, but generally don’t include a range around a chance of rain or temperature. Marketing forecasts should not make this mistake; confidence definitionally falls further into the future, and confidence bands should follow suit.
• Beware Extrapolation: Diminishing response curves—which are typical in marketing—are sometimes fit with a small range of stimulus (x) data. The curves look sharp all the way out on the right, but in truth we have no information to predict those points. So, if we are recommending spending a lot more, we have to be honest about what is likely to happen. Flipped on its head, marketers should seek to add variation to spend levels to “soft test” on the outer parts of curves.

Quality Pays

Most readers will nod their heads to the points made above, but might wonder if they are worth doing. Companies, after all, care about profit—and focusing on quality might sound like an expensive nice-to-have. Well, Ford adopted its “Quality is Job One” tagline after they got their clock cleaned by Japan, and if you went back in time and asked their executives, they would have told you that not focusing on quality probably cost them hundreds of billions of dollars over several decades.

But what about marketing? Marketbridge has been collecting data on quality and precision among its clients since 1997. Specifically, we have been interested in its impact on four key metrics: customer lifetime value (CLV); net CPA; long-run growth; and the percentage of go-to-market dollars that are “working” vs. “non-working.” In all four cases, quality-focused organizations have achieved superior results.

• CLV is particularly affected; on an industry-by-industry basis, CLV in quality-focused marketing organizations averages about 10% higher—mainly result of a scientifically-determined, less sugary acquisition diet
• High-quality digital targeting yields around 20% better 6-month retention than cookie-based, spray-and-pray approaches
• Organizations that adopt high-quality approaches have better working / non-working dollar ratios than those that do not, a result of better efficiency and less waste that can be translated into higher media spend
• Overall, high-quality marketing organizations grow at healthier long-run rates than those that are more reactive

A focus on quality isn’t sexy or flashy—it is a cultural shift, and a long-term commitment. However, it might just be the highest ROI marketing mix decision an organization can make.

Wondering how to ensure your organization is focusing on quality in marketing and measurement? Reach out!

The post Quality marketing analytics appeared first on Marketbridge.

Our unique approach

Marketbridge takes a consultative approach to building MMMs, MTAs, and “UMMs”—which integrate the functionality of both. We start with our extensive econometric, inference, and optimization libraries, and then build a bespoke solution for each client. But across projects, our core marketing effectiveness principles remain the same: the open-source measurement consultancy, complex measurement specialists, and actionable brand measurement.

The Open-Source Measurement Consultancy

• Built in your infrastructure:
  We build inside your domain, keeping data first-party and your measurement code in your version control. This also means that data engineering pipelines are native. API calls come from your environment direct to platform, publisher, and Martech sources.
  
  
• Radically whitebox:
  Both custom elements and Marketbridge libraries are viewable and modifiable at the source code level in Github, ensuring auditability and reproducibility.
  
  
• Near real-time data connectivity:
  Direct APIs wherever possible allow rapid updating of source data and re-estimation of model coefficients on a daily basis.

Complex measurement specialists

• Bespoke complex builds:
  Over and above simple business use cases, we model the most complex go-to-market activities across considered purchases, financial services and subscription businesses.
  
  
• B2C to B2B flexibility:
  Our methods handle small-n, long transaction cycle businesses as well as high-n consumer brands.
  
  
• Right-size performance marketing:
  We use systems of equations to avoid over-attributing value to branded paid search and affiliate “capture” channels, and then redistribute value to driving channels.

Actionable brand measurement

• Quantify your brand’s long-run impact:
  We model advertising’s impact on brand strength, and brand strength’s corresponding impact on sales—insuring accurate ROAS up- and down-funnel. The common question “should I be optimizing on ROAS because it doesn’t take brand into account” is now obsolete.
  
  
• Considers both paid and earned media:
  With our strong heritage in PR, we weigh investments in syndication, influencer marketing, and media relations. This will be increasingly important in the era of LLM discoverability.
  
  
• Measures the right brand attributes:
  Identify the upper-funnel KPI that does matter to drive true long-run growth.

Learn more

We’d love to meet with you to share some case studies, learn about your organization’s current stage on its measurement journey, and discuss potential pitfalls for MMM and MTA. Contact us to get in touch.

Footnote: Forrester does not endorse any company, product, brand, or service included in its research publications and does not advise any person to select the products or services of any company or brand based on the ratings included in such publications. Information is based on the best available resources. Opinions reflect judgment at the time and are subject to change. For more information, read about Forrester’s objectivity here .

The post Why Marketbridge for MMM or MTA: Principles of our open-source approach appeared first on Marketbridge.

In today’s environment – where skepticism towards healthcare institutions is at an all-time high – the gap between consumers and the health care providers feels wider than ever (John Hopkins Carey Business School, Edelman). This mistrust doesn’t just affect perceptions of a given brand or provider; it can directly impact health outcomes; when trust is lacking, patients may delay care, avoid preventative services, or disregard treatment altogether.

While not necessarily its primary job, marketing should be used as a tool to bridge the gap and rebuild trust. Effective techniques do not just promote services; they build credibility, demonstrate transparency, and convince consumers that their needs and values come first. Moving to a quality-driven, trust-building marketing approach benefits all parties by fostering connections, driving engagement, and ultimately, improving outcomes for both the consumer and the provider.

Defining quality: What to say, and how to say it

In the context of marketing, “quality” means more than grammatical correctness or an aesthetic design. It’s about delivering a specific message to a specific audience in specific places at specific times, with each touch curated to resonate and to either drive a specific action or build trust and brand equity. This requires understanding the target audience, and crafting each communication to be honest, accessible, and most importantly, personal.

Consider John, a rural Virgina resident who hasn’t seen a doctor in years. A generic email about low-cost scheduling might be cheaper to send, but it won’t necessarily convince him to act. A tailored direct mail piece highlighting proximity of care or a two-way text encouraging conversation with an agent might be the experience that would work better with John. Send both and explain how a doctor’s visit is likely to improve John’s quality of life, and you’ve transformed a generic outreach into a personalized invitation—one that feels relevant and actionable.

Now consider the type of insurance shoppers that would likely respond to a TV ad promoting a “$0 Premium” “No Risks” plan versus members that would be more swayed by a billboard stating, “Insurer A has been serving the Charlotte community for 15 years. We are committed to bringing the highest level of care to our members today and for years to come.” Will the TV ad outperform the billboard on applications? Probably. But will the quality or value of the members that enroll from the TV ad be as high as the members that enroll from the billboard? Probably not. And which ad goes further to rebuilding trust between the provider and the consumer?

The need for measurement: Quality marketing requires quality data and KPIs

High-quality data is a pre-requisite to running personalized marketing campaigns. In the John example, the team wouldn’t have been able to deliver the right message if they hadn’t known he 1) hadn’t been to a doctor in years and 2) lived in a rural area. And how did they know that? Data. Or how could the team even determine that the tailored direct mail plus 2-way messaging drove a response, while the generic email did not? Again, they had the tracking in place to tie marketing touch points AND desired outcomes back to John.

However, what if the team noticed that, in aggregate, the generic email campaign drove a greater number of responses than the tailored direct mail + 2-way SMS? What would they deem the better campaign? Just like the gimmicky “$0 Premium” TV ad versus the Local billboard, to say which is “better,” an objective or goal metric must be in place. What was the goal of the “get to your doctor” campaign? Appointments scheduled? Appointments kept? Repeat visits to the doctor? For the health insurance ad, was the goal to just drive as many new members as possible, to build brand equity, or drive high future tenure enrollments? It is critical that marketers set a campaign objective prior to launching, and that the campaign objective aligns with the overall business strategy.

Conclusion

Marketers in the health care industry face real headwinds in an environment where individuals are increasingly skeptical. Those who seek to foster trust through the messages and creatives they deploy should consider this simple list:

1. Determine what business goals a campaign is meant to align to and set KPIs accordingly (i.e. value vs. volume)
   • For acquisition marketing, “quality KPIs” could include total lifetime value (LTV) of new member sales, specific membership profile target, or even a longer term KPI of 12-month retention of new members.
     
2. Use messaging that is personalized to the consumer and/or that gives the consumer a reason to trust their company
   • Steer clear of relying solely on quick-win tactics that are effective at eliciting an immediate response from customers, and shift more into creatives that are meant to build credibility and consideration of the brand and product over time
     
3. Track the right metrics (defined in #1) to determine if campaigns have the desired effect
   • Remember, the desired effect will almost always take longer to observe within a value paradigm vs. a volume paradigm. When value is the goal, it typically also means that measurement will be over a longer time horizon than if measurement only required counting appointments, clicks, applications, etc. Be patient.

For many, this may be a significant shift in strategy that will require alignment from multiple parts of the organization and therefore take some time. But committing to a plan like this will pay off in higher brand equity, more satisfied customers, and almost certainly, more sustainable growth.

The post From quantity to quality: Rethinking healthcare marketing for long-term value appeared first on Marketbridge.

A common problem with media mix models (MMMs) is over attribution of down-funnel, demand capture channels, such as paid search and affiliate. These channels have become, in some cases, distribution channels rather than marketing channels. Said another way, many actors have naturally sought to extract economic rent by inserting themselves in a buyer’s discovery and purchase process.

Importantly, there are two different kinds of down-funnel rent seekers: Duplicative and Distribution Taxing. They are defined by what would happen if they weren’t in place, in other words, the counterfactual case.

• If Duplicative channels were removed in the counterfactual case, nothing would happen to sales.
• If Distribution Tax channels were removed in the counterfactual case, sales attributed to these channels last touch would disappear.

Google search is perhaps the best example of distribution taxing. Given Google’s ubiquity, it is essentially impossible for the average consumer to get around the search interface, and in a competitive market, money must be paid via a search bid to be in the list of considered companies. Hence, removal of the bid term will likely result in lost sales. It is generally impossible to disintermediate well established distribution taxers, but it is important to understand where they sit and the amount of economic rent they demand, typically best understand as a percentage of revenue—just like the margin discount a manufacturer provides to a retailer.

Figure 1: Three basic types of down-funnel channel; two are truly “demand capture”

Certain Affiliates, on the other hand, are often duplicative. In the case of a toolbar like Honey, the Affiliate is essentially cherry picking, waiting until the consumer is ready to buy to take credit. In these cases, removal of the bid to these platforms will likely have little effect.

The third type of channel, accretive, generates demand that would otherwise not have happened. This is often taken into account in a taxonomy determining the channel’s objective funnel position—usually called demand generation—but any channel can be accretive. For example, a demand capture channel that nudges an on-the-fence buyer to purchase could be partially accretive.

A helpful way to think about the degree of duplication is downside elasticity. Using our counterfactual example, this is simply the quantity of sales lost divided by what we would have predicted the loss would be based on the channel’s assumed ROAS—usually reported on a last touch basis. For example, say that an affiliate channel reports a ROAS based on last touch of 5 (an investment of $1M drives $5M of sales, e.g.) In the counterfactual case of no investment in the channel, say we only lose $1M of sales. The downside elasticity would be 0.2—not a great result, implying that the true ROAS is about 20% of that reported.

Downside elasticity: The total revenue lost when deinvesting in a channel (all other things being equal) divided by predicted lost sales based on assumed (usually last touch) ROAS

Figure 2: Downside elasticity is a helpful way to think about duplication vs. incrementality.

Measurement

In reality, all channels have some mix of duplication, distribution taxing, and accretive behavior. It is the job of the marketing analyst to estimate this ratio and keep it fresh. There are two basic ways to do this: econometrically and via testing.

In marketing, econometric estimation of cause and effect is called MMM (media mix modeling or mixed media modeling, depending on who you ask.) In this approach, stimuli (marketing promotions and earned media) are used to explain response (sales or some proxy). In the case of down-funnel channels, it is critical to model these as an intermediary between demand generation and sales. In other words, the modeler must create multi-stage models to allow a channel like search both “receive credit” from other channels, and “create demand.” This can be a challenge for the modeler, as multiple “second stage” channels are generally required. This ends up looking like a system of equations, and visuals are extremely helpful to remember what is going on.

Figure 3: A system of equations is helpful to visualize what’s going on with channels in an MMM.

After the modeling is complete, it is possible to list each channel’s CPA (cost per acquisition) and / or ROAS (return on advertising spend, essentially the inverse) in three ways: last-touch (what Google or the Affiliate will take credit for); one-way (the causal impact of the channel not taking other contributing channels into account); and multi-stage (removing credit and reassigning it to contributing channels.) Seeing these three metrics side-by-side allows marketers to understand the trade-offs between channels, and to better interpret the often misleading results reported by platforms and agencies.

We sometimes call this a “systems expansion” view of marketing contribution. In the figure below, which for simplification’s sake does not include last touch data, each channel’s single-level regression spend, return, and ROAS are listed in the upper-left table. In the top-right table, each “upper funnel” channel’s contribution to paid search and affiliate are added to its last-touch contribution, and then subtracted for paid search and affiliate. Once ROAS are adjusted, an “MTA effect” (in the bottom left table) is calculated—essentially the degree to which each channel is taking credit from or giving credit to other channels.

“Single Level” MMM Return and ROAS:

Multi-Level MMM Return and ROI:

“MTA Effect”:

Figure 4: The MTA effect can be calculated by dividing a channel’s true “driver” contribution by its last touch or single-level MMM contribution. A positive effect means the channel is more accretive than it appears, and a negative effect means it is over-crediting on true incrementally.

Of course, testing is the gold standard way to calculate a channel’s incrementality or downside elasticity. There are generally two options: Geo-based holdouts or time-based reductions.

Geo-based holdouts using synthetic controls have become common in modern marketing. In this approach, several test markets are chosen for a treatment—either a positive (upside) treatment or a negative (downside) one. At the same time, a synthetic control—essentially a weighted grouping of the remaining markets—is set aside to run at “standard” levels. Then, a causal inference Bayesian analysis is performed to understand the difference between the experiment and the control-the counterfactual.

The challenge with geographic tests is that they can be difficult to execute. In many cases, it is simply impossible to persuade an affiliate to shut off bids geographically (for obvious reasons—they don’t want to be tested.) In other cases, algorithmic optimizations interfere with test purity. In these cases, a whole market reduction in spend can be used. In this case, a channel can be “dimmed” by, say, 50% to understand reduction in sales. This is not as statistically easy to read—there is no same-time-period counterfactual—but they provide the natural variability that can be read in an econometric time series (MMM) model.

Key Takeaways

• It is important to think about demand capture channels as distribution taxing, purely duplicative, and accretive
• Distribution taxing channels cannot be avoided, but should be understood strategically to potentially disintermediate with long-run go-to-market strategy changes / routes-to-market
• Analysts can be model duplicative and distribution-taxing channels via a multi-stage econometric modeling approach inside of an MMM
• Last-touch, single-level, and multi-level CPA and ROAS should be reported side-by-side in output
• The gold standard to understand downside elasticity is a geo-based holdout

The post Down-funnel channels: Duplicative, distribution taxing, or accretive? appeared first on Marketbridge.

Customer Data Platforms (CDPs) were introduced to address the need for the comprehensive, multi-channel data necessary for modern marketing. For many organizations, these solutions provide helpful structure around audience segmentation and campaign targeting. But traditional CDPs are built with fixed logic. They assume a degree of centralization and integration that most financial institutions simply do not have – and often require marketers to adapt to the software, rather than the other way around.

That’s why I believe composable CDPs (sometimes referred to as go-to-market data lakes) are a better fit. They allow marketing, analytics, and technology teams to assemble a flexible data foundation that works across existing systems. Instead of being forced into someone else’s box, you get to design the system around your own business needs. And in an industry with complex products, legacy infrastructure, and heightened regulatory expectations, flexibility matters.

Here are five ways a composable CDP can help:

1) Align Marketing, Analytics and Tech Around Shared Goals

One of the biggest challenges I see in financial services is that marketing, analytics, and tech teams operate in their own ecosystems and still speak different languages. They’re doing good work, but often on different timelines, with different priorities and different definitions of success. Marketing focuses on strategy and outcomes, analytics is buried in reporting and data engineering, and tech is managing capabilities and infrastructure. When those groups aren’t working from a shared roadmap, priorities get misaligned quickly.

A composable architecture helps bring those teams together. When you organize around specific use cases – like onboarding new customers or identifying upsell opportunities in the advisor channel – it’s easier to stay aligned. Everyone understands what they’re building and why. That cuts down on back-and-forth, reduces wasted effort, and improves speed to market.

A composable architecture also helps reduce cost. Anyone who’s worked through multiple rounds of rework knows how expensive it can be to get it wrong. This approach minimizes that risk.

2) Tame Data Complexity from Mergers and Legacy Systems

Most financial institutions aren’t starting from a clean slate. They’ve grown through acquisitions. They manage multiple product lines and deliver through multiple distribution channels. And they often rely on infrastructure built over decades – which means customer data lives across dozens of antiquated systems, none of which were designed to talk to each other. Add in a wide range of state- or account-level variations and compliance requirements, and you’ve got a perfect storm.

Trying to shoehorn all of that into a single CDP can be painful and expensive. Composable CDPs work differently. They allow you to connect the systems you already have, extract what matters, and standardize the data just enough to activate it. You don’t have to rebuild everything. You can move forward with what’s useful and gradually evolve from there.

This is particularly helpful when you’re trying to deliver consistent experiences across business lines or channels that weren’t originally designed to coordinate. A composable approach makes that achievable.

3) Protect Customer Trust While Meeting Regulatory Demands

Another big reason this matters in financial services? Regulation. Privacy and compliance are non-negotiable and a marketing data strategy that doesn’t fully account for them will eventually fail – if not operationally, then reputationally.

A composable CDP can help on both fronts. It provides structure for managing consent preferences, documenting data lineage, and making sure sensitive data isn’t used out of context. It gives compliance teams the transparency they need, while still giving marketers the ability to move with speed.

You don’t have to choose between responsible data practices and effective marketing. With the right setup, you can do both.

4) Move Beyond Guesswork and Test Like Scientists

Many marketing teams want to build a culture of experimentation; however, in financial services, it can be a struggle to run tests that meet both business and regulatory standards. Whether you’re optimizing retirement planning campaigns or fine-tuning service reminders for lapsed policyholders, experimentation can feel risky without the right controls.

A composable CDP changes the game. It gives you access to real-time data across systems, supports test design, and makes it easier to track and optimize performance in a way that stands up to internal scrutiny. This doesn’t just improve outcomes – it improves credibility and trust with the rest of the business. When marketing shows up with results instead of opinions, it becomes easier to justify budget, ask for resources, and lead with confidence.

5) Scale Personalization That’s Actually Useful

Personalization is important, but only if it’s meaningful. Sending someone their first name in a subject line doesn’t move the needle. However, a needs-based approach that allows you to recognize that a young family is saving for college, or that a retiree is reevaluating their drawdown strategy, actually might.

A composable CDP helps you make that leap. It enables you to respond to intent-based behaviors, engagements, signals, and life events—so that you can serve the right message at the right time. And because it’s connected across systems, you’re not guessing. You’re making decisions based on what people are doing, not just who you think they are.

Done right, this builds trust. Customers begin to expect, and appreciate, that your outreach makes sense given their situation.

Getting a handle on this is 100% doable

I’ve worked in financial services long enough to know how hard all of this can be. The systems are fragmented. The expectations are high. And the time to show results is always shorter than anyone would like.

But I’ve also seen what’s possible when marketing, data and tech teams come together around a common strategy. Composable CDPs don’t eliminate the complexity, but they make it manageable. They provide the architecture to move faster, plan smarter and execute with greater clarity.

At Marketbridge, we help financial services organization build these kinds of systems. We’ve got both the technical and industry expertise to help connect strategy to architecture, marketing to analytics, and data to decisions.

If you’re navigating disjointed infrastructure, dealing with legacy or disparate systems, exploring how AI fits into your stack, or just trying to modernize the way your team operates, we’d be glad to share what we’ve learned. Let’s talk.

Download the whitepaper, “Building a composable go-to-market data stack”​

Rethink your data foundation and lead the next era of AI-ready, insight-driven marketing.

Download

The post Five ways a CDP can help financial services marketers appeared first on Marketbridge.

Whether you’re focused on B2B or B2C, this challenge is widely understood. Most marketers don’t lack awareness; they’re already making moves to fix it. Unfortunately, that’s where many are encountering their next problem.

In an effort to unify customer data and power smarter campaigns, many teams invested in Customer Data Platforms (CDPs), viewing them as the silver bullet. But the promise hasn’t matched the reality.

CDPs were built to activate, not to analyze. Most struggle to provide true cross-channel visibility, AI-ready insights, or the depth of performance tracking required by modern go-to-market teams and c-suite stakeholders. We’ll explore these needs in detail in just a moment, but first, it’s worth understanding why the CDP model is breaking down.

Traditional CDPs promised a single source of truth, but delivered it in a rigid, vendor-controlled box. They’re often expensive, difficult to adapt, and optimized for short-term execution rather than long-term agility. As marketing stacks evolve, these singular platforms become bottlenecks, limiting integration, innovation, and insight.

Introducing Composability: A Smarter, More Strategic Path Forward

Composability offers a fundamentally different approach. Instead of relying on one vendor or platform to do it all, composable data architecture lets marketers build and evolve their own stack—piece by piece—based on changing needs.

Think of it like LEGO bricks. Marketers can connect best-in-class tools for campaigns, analytics, and CX while maintaining a centralized, AI-ready data layer. Composability empowers marketers to innovate without ripping and replacing systems, while still ensuring their data is unified, structured, and usable across teams.

At Marketbridge, we call this centralized foundation the Go-to-Market Data Lake (GTMDL): a composable environment that integrates marketing, sales, CX, operations and even clinical data to fuel attribution, personalization, and growth.

And while the architecture matters, strategy matters more. That’s why we help healthcare marketers not only build GTMDLs but also design the right data strategy to make them valuable.

Three Essential Components of a Modern Go-to-Market Data Strategy

To unlock the full potential of a composable architecture, you need more than the right tools, you need a strong strategic foundation. These three elements are the building blocks of a future-proof data strategy for healthcare marketers:

1. Marketing Attribution Models

Attribution remains one of the most challenging aspects of healthcare marketing. At its core, it’s about understanding what’s working, what’s not, and why. In today’s complex, multi-touch, multi-channel customer journeys, that clarity is hard to come by.

Every interaction matters. From paid search to sales outreach to out-of-home campaigns, each touchpoint influences the decision-making process. Yet most marketers still can’t answer critical questions like:

• Which campaigns are actually driving conversions?
• Are we over-investing in one channel and ignoring others?
• What is the true ROI of our brand or media spend?
• Which messages resonate with specific audiences?

The absence of visibility leads to misallocated budgets, inconsistent performance, and decisions based on intuition rather than data. Although improved dashboards can assist, they are insufficient on their own.
This issue becomes even more critical when marketing leaders must demonstrate their impact on the C-suite. Executives primarily focus on business outcomes rather than specific channel metrics.

Without attribution models that connect marketing investments to clinical engagement, member retention, or revenue growth, marketing efforts are often viewed as a cost center rather than a driver of growth. The pressure to prove ROI in measurable, financial terms has never been higher.

Solving attribution requires a unified data foundation and models that reflect real-world behaviors, not just last-touch conversions. Tools alone cannot connect every part of the customer journey or account for external variables like competition, compliance, or seasonality. That is why advanced models, such as Marketing Mix Modeling (MMM) and Multi-Touch Attribution (MTA), are essential to a modern marketing strategy. They move attribution from reactive reporting to proactive planning and provide the kind of defensible, business-aligned insights that resonate with senior leadership.

2. Taxonomy

Taxonomy may not be flashy, but it’s the quiet engine behind every successful data strategy.

Simply put, taxonomy is the consistent naming, tagging, and classification of data across systems. Without it, even the most sophisticated models and tools fail. In healthcare, where data flows from CRMs, marketing automation platforms, claims databases, and EMRs, inconsistency is the enemy. The same channel might be labeled “DM,” “Mailer,” or “Offline Touch” making measurement, automation, and analytics nearly impossible.

A clean, governed taxonomy enables:

• More Accurate Attribution: Consistent tags let you connect touchpoints to outcomes with confidence.
• Stronger AI Models: Clean, labeled data is essential for machine learning and predictive analytics.
• Better Collaboration: When Sales, Marketing, CX, and IT speak the same data language, you eliminate misalignment and confusion.

Ultimately, taxonomy turns raw data into structured, queryable insights. Without it, your composable stack cannot function effectively, no matter how advanced the architecture.

3. Data Integration and Accessibility

Even with strong models and a clean taxonomy, your strategy stalls without seamless integration and access to data.

In healthcare, valuable insights are often buried in disconnected systems. A claims system captures utilization, a marketing platform tracks member engagement, a CRM tracks sales outreach, and none of it talks to each other. The result? Manual workarounds, data gaps, and missed opportunities.

Composable architecture solves this by connecting systems through APIs and microservices. It allows you to:

• Centralize and Normalize Data in Real Time: Build a unified view without replacing every tool in your stack.
• Act on Data Immediately: Marketers can analyze and optimize campaigns without waiting weeks for reports.
• Personalize at Scale: Trigger outreach based on real-time behavior, clinical activity, or enrollment milestones.
• Minimize IT Bottlenecks: Provide governed access to marketers and analysts while maintaining compliance and security.

Integration is the connective tissue of a data strategy. Without it, even the best models and insights stay locked inside silos.

The Next Era of Healthcare Marketing Starts with Data Strategy

Healthcare marketers are navigating one of the most complex data environments in any industry. The stakes are high, and so is the pressure to deliver measurable outcomes. But as tempting as it is to chase the next big platform, lasting success comes from something deeper: a modern data strategy that aligns teams, tools, and tactics around shared goals.

A composable approach gives you the flexibility to evolve as your business grows, the clarity to connect action to outcome, and the control to move fast without breaking compliance or collaboration. But the real power lies in how you use it. In healthcare, the ROI of better data isn’t just improved marketing and sales performance; it’s healthier members, reduced churn, and stronger care engagement. A strategic approach to data is what turns fragmented insights into meaningful action that drives both clinical and business impact.

At Marketbridge, we help healthcare organizations move beyond disconnected tools to build integrated, insight-driven systems that support real business outcomes. From attribution models and clean taxonomy to full data integration, we bring strategy, structure, and execution together—so marketers can finally do what they’ve always wanted: make smarter decisions with confidence.

Because in the end, it’s not about collecting more data. It’s about putting it to work.

Download the whitepaper, “Building a composable go-to-market data stack”​

Rethink your data foundation and lead the next era of AI-ready, insight-driven marketing.

Download

The post Healthcare marketing needs a data strategy reset appeared first on Marketbridge.

This idea originated from customer frustrations with bloated Marketing software platforms as major vendors acquired dozens of point solutions over the past decade. Implementations became notoriously complex, and modules lacked strong integration or simply underperformed compared to best-of-breed alternatives. Composability shifted more control back to the client and focused on creating building blocks and better integrations. The approach has been welcomed across the industry, and many major vendors have modernized their offerings around this concept.

Evolution of Customer Data Platforms

The demand for composability becomes particularly evident when examining how customer data platforms have evolved and where they’ve fallen short. Marketing platforms have been embedding customer databases since the ’90s. The functionality started as storing basic customer lists and has grown to include a sophisticated set of attributes including preferences, behavioral signals, order history, and more. This was coined in 2013 as the “Customer Data Platform” (CDP) and the vision promised a 360-degree view of the customer that could be built inside marketing platforms without the help of IT. This vision resonated because customers were already using these platforms to orchestrate across an array of digital channels, and enriched data enabled additional use cases like personalization, cross-selling, and retention.

However, several limitations emerged with CDPs. First, they required pre-cleaned data, necessitating that transformations and cleansing be handled before import. Additionally, CDPs were not designed for advanced modeling and analytics, requiring data exports to dedicated data warehouse systems for complex analysis. Cost management also posed challenges, as most CDP pricing models were based on the number of stored records, which could inflate costs quickly. While CDPs are a great tool for digital activation and campaign execution, they fall short as comprehensive solutions for end-to-end go-to-market data management.

As organizations recognized these constraints, the prominence of CDPs has started to shift in the marketplace. The State of Martech 2025 survey revealed that CDPs fell from 27% to 17% and Cloud Data Warehouse ticked up from 21% to 24% as the center of the Martech Stack amongst B2C and B2B organizations. This trend indicates that a growing number of organizations are demanding more from their marketing data infrastructure.

Data as a First-Class Citizen

The solution lies in treating data as a foundational infrastructure layer rather than an embedded software feature. We call this solution the go-to-market data lake (GTMDL)—a strategy that places customer data in a modern cloud data warehouse such as Databricks or Snowflake. This provides greater flexibility, cost management, and analytical capabilities compared to using embedded databases within CDPs.

Importantly, the scoping and prioritization of a GTMDL should be guided by marketing use cases, not IT milestones.

This graphic illustrates key use cases the GTMDL supports.

The power of the GTMDL can be illustrated with a common direct-to-consumer (D2C) acquisition use case. The first step is to acquire a prospect list from a data provider and import it into the GTMDL. Next, data pipelines need to be created to augment, clean, and enrich the data with additional sources.

Once the data is prepared, a propensity model is created by training on historical customer sales data. This model identifies the optimal attributes to target when finding “look-alike” prospects in the new list. Each record on the prospect list is scored against the trained model, assigning a probability between 0-100%, indicating the likelihood that the prospect would convert as a customer.

The scored list is then segmented into testing groups (typically deciles) and exported to the activation platforms before being pushed out to the market. As the offers run in the market, performance data including clicks, impressions, engagement, and sales flows back into the GTMDL. This drives real-time campaign performance reports showing metrics like return on ad spend (ROAS) and enabling advanced measurement like multi-touch attribution models.

This entire workflow can be handled within the GTMDL, demonstrating why this approach is the preferred solution for marketing measurement and activation. Storage is cheap, and millions or even billions of rows can be imported without concern. Multiple languages such as Python, SQL, and R can be run in the platform to perform advanced data transformations and modeling. Compute resources can be provisioned on-demand, with costs incurred only during active processing. Finally, activation platforms integrate seamlessly to streamline audience deployment and engagement data collection―and you have full control of the data layer, which provides resilience from perpetual vendor upgrades and platform changes. Leading vendors like Databricks, Snowflake, and AWS Redshift all provide robust capabilities.

Conclusion

Marketers need the ability to both measure and activate their customer data, and the GTMDL makes this possible within a single platform. This solution enables measurement capabilities that CDPs have failed to deliver and that are essential for proving marketing ROI. The composable approach establishes GTMDL as the central hub, proving superior to an embedded CDP database.

The transition to this approach is more accessible than many organizations realize. Unlike the lengthy, resource-intensive CDP implementations that can drag on for months or years, GTMDL projects start with a use-case driven approach. Organizations can begin by focusing on their highest value measurement or activation challenge, such as multi-touch attribution or retention segmentation to demonstrate clear business value. This approach allows marketing leaders to build confidence while minimizing risk and resource commitments, ultimately enabling them to prove their ROI with real-time results and accurate measurement that drives budget allocation and business credibility.

For go-to-market leaders, the choice is becoming binary: evolve toward a data-centric, composable data architecture or risk being outmaneuvered by competitors who can measure, test, and optimize at greater speed and scale. In today’s marketplace, data-driven marketing is no longer a competitive advantage; it’s table stakes for survival.

Download our GTMDL whitepaper for a comprehensive guide to implementing this approach.

Download the whitepaper, “Building a composable go-to-market data stack”​

Rethink your data foundation and lead the next era of AI-ready, insight-driven marketing.

Download

The post The rise of composability in marketing stacks appeared first on Marketbridge.