Foreword

For years, I’ve worked with and watched founders of insights and data companies chase the SaaS dream. The logic is seductive: research follows repeatable methodologies, so why shouldn’t it scale like software? Investors love the model, too. What’s not to like about recurring revenue, high margins, and proprietary IP?

Yet despite massive investment and a redefining of the industry’s perimeter, the work we consider to be traditional market research has stubbornly retained a service component. It is a tech-enabled services industry, and that’s not SaaS.

Why?

To paraphrase Clayton Christensen, it’s because the industry’s clients are not “hiring” market research to learn. Their jobs are to make decisions that make their companies money. When they are not sure what to do, they will seek manageable, defensible ways to minimize their uncertainty. What they really want is reassurance. One way they might get it is by commissioning market research. Another might be asking their pal Bob down in Sales. But whatever the solution, the conditions of manageability and defensibility must always be met.

Therefore, even if your platform can help them make a great decision, if it is not easy for them to use, then they are no longer in your addressable market. In fact, what you’re asking them to buy is a business transformation.

This may be changing, though. AI is fueling a new wave of SaaS optimism. Will AI-powered market research be good enough to sufficiently replace much of the human part of insights and provide a defensible and manageable solution for clients so that a SaaS operating model can finally take hold?

Possibly yes.

And will it give them the reassurance they are looking for?

That is a good question indeed.

This book is about why a SaaS transformation has failed to take meaningful root in market research, and what platform growth and scale actually require. It’s about why products don’t always succeed on capability alone, and why service often wins the day. I share the key factors that determine whether a product-driven or service-driven model is required, and thus what kind of customer success and support teams must be built. I also look at the growing use of AI and at the two emerging operating models of AI-forward companies.

If you’re running a SaaS business and you’re wondering why your growth has stalled—why adoption isn’t happening as expected, why customers aren’t renewing, why expansion is harder than it should be—this book will help you see the real problem. And more importantly, what to do about it.

If you are an investor in the sector, you will have a clearer understanding of how firms create and unlock value with clients, and how to minimize deal risk by recognizing whether a target company is truly a SaaS business or a tech-enabled service business.

Let’s dig in.

1. Why Market Research Appears Ready for SaaS

In little more than a decade, market research has undergone a profound shift. Digital platforms, AI, and automation have reshaped what was once a sector defined by human expertise and hand-crafted solutions. ESOMAR’s 2023 Global Market Research Report suggests that over half of the industry’s revenue now comes from digital data and SaaS businesses. Not only are these not traditional market research firms, but few would even identify as belonging to the sector. Instead, they position themselves as tech companies that blend analytics, automation, and software-driven solutions.

For investors, founders, and operators, market research, an industry long defined by manual and highly customized work, should be an obvious candidate for SaaS transformation. At a high level, it exhibits many of the characteristics that have made other industries like accounting and project management ripe for technology-driven reinvention. It is a highly fragmented industry, dominated by firms that have historically relied on costly labor-intensive processes and suffered from long project cycles to deliver value. And, critically, the work being done, at least in principle, appears repeatable: the same fundamental research techniques are used across companies, industries, and markets.

The logic of applying SaaS transformations to market research should therefore follow a familiar playbook. If research companies can develop software that automates data collection, standardizes methodologies, and delivers insights in real time, they should be able to shift from project-based work to a recurring revenue model with predictable margins and scalable growth. Rather than relying on teams of researchers to execute each project, companies could generate insights at scale through a self-service model, where clients interact with a platform with embedded research knowledge rather than a service team.

This vision has driven a steady stream of investment in research technology. Over the last decade especially, companies have attempted to “productize” market research through:

Survey platforms that allow users to design and execute their own research
Syndicated data products that monetize proprietary data assets
Self-service tracking and analytics tools that offer real-time insights.

While these solutions have gained traction, they have not fundamentally altered the structure of the industry. Traditional market research agencies—those built on human expertise and bespoke client engagements—continue to command a substantial share of spending. Even when SaaS solutions have made inroads, they have often been adopted as supplements, not substitutes for custom research services.

This raises a critical question: if market research is such an obvious candidate for SaaS transformation, why hasn’t the transition fully materialized?

The Business Case for SaaS in Market Research

The belief that market research should evolve into a SaaS-driven industry is not unfounded. The core functions of market research—data collection, analysis, and reporting—appear to align with SaaS principles. Several factors reinforce this belief:

Labor Costs

Market research has always been labor-intensive, relying on skilled professionals for survey design, data collection, and analysis. These human-driven processes make growth expensive. SaaS offers a compelling vision: replacing manual tasks with automation and AI to reduce costs and increase capacity and thus enable scale. Every technology platform leans into this promise of automating research workflows to make insights faster and cheaper. For the companies creating these platforms, automation is about more than saving the client money; it’s about unlocking their own scalability and high margins without adding headcount.

Inconsistency

Market research generally starts from scratch, which is naturally (and highly) inefficient. SaaS, by contrast, thrives on consistency and standardization. Syndicated data products or products that apply consistent methodologies to achieve repeatable outputs offer the promise of reducing redundant effort and bringing predictability and the potential for scale to what would otherwise be a highly variable and expensive process.

Speed and Efficiency

Speed has become a currency in a variety of client industries, and traditional market research feels increasingly out of step. Clients demand real-time insights that align with the fast pace of their businesses. Dashboards and automation tools promise to deliver just that. By enabling always-on access to data and reducing delivery times, these solutions align with SaaS’s value proposition of agility and responsiveness. There are innumerable platforms (each now, unsurprisingly, adding AI) attempting to capitalize on this demand for immediate business-critical data, hoping to position their platforms as indispensable tools.

Data Monetization

The digital economy produces vast amounts of data, from consumer behaviors to transactional records, creating opportunities for recurring revenue streams. SaaS models thrive on this kind of process-it-once, sell-it-often data productization. Panel companies, sample aggregators, and marketplaces are increasingly trying to monetize these assets in the broader customer acquisition ecosystem.

The Persistence of the SaaS Vision

Beyond immediate operational benefits, the idea that market research should naturally evolve into a SaaS-driven industry is reinforced by broader shifts toward technology-driven decision-making. Executives across industries have grown accustomed to software streamlining every aspect of their business, from finance and marketing to supply-chain management. In such environments, market research delivered via always-on platforms rather than traditional projects feels intuitively inevitable.

Investors find this prospect especially attractive because the economics of SaaS businesses are fundamentally superior to service-driven models. SaaS companies generate predictable recurring revenue and scale easily, making each additional dollar earned more profitable than the last. Consequently, investors reward these firms with higher valuations. Under investor pressure, market research founders increasingly position their businesses as SaaS-first platforms that promise automation, self-service, and scalability.

Yet despite this widespread belief in SaaS inevitability, the reality on the ground tells a different story. Self-service research platforms still represent only a small fraction of market spend. ESOMAR estimates their total size at just $3.19 billion—less than 3% of the overall market and only 7% of the size of traditional research. The limited inroads these platforms have made isn’t simply because they are ahead of their time; rather, it suggests a deeper structural misalignment between SaaS economics and market research practices.

To understand these challenges fully, let’s examine why SaaS has struggled to displace traditional operating models.

2. Why Market Research Has Resisted the SaaS Transition

For all its promise, the SaaS model has not fundamentally reshaped market research the way it has transformed other industries. Despite a surge in investment in research technology, self-service platforms, and automation, traditional market research agencies—built on human expertise and bespoke client engagements—continue to command a dominant share of spending. This reality is difficult to reconcile with the supposed inevitability of software-driven efficiency. If market research is so ripe for SaaS transformation, why hasn’t the transition fully materialized?

The answer, to paraphrase Clayton Christensen, lies in a fundamental misunderstanding of the job the industry’s clients are “hiring” market research to do. Investors and founders frequently assume that clients have an informational problem that is rooted in what the industry calls the “Iron Triangle,” where any supplier’s solutions effectively oblige the client to sacrifice one element of speed, cost, and quality. From this viewpoint, solutions that provide information better, faster, and/or cheaper should be a “slam dunk.”

But market research is not just about acquiring information; it is about structuring complex business questions, interpreting findings within strategic and operational constraints, and integrating insights into broader decision-making processes. The outcome the industry’s clients typically seek isn’t learning: it’s sufficient reassurance that they are going to make the right call. The problem they are trying to solve is therefore not just one of finding information, much less the best information. It is ultimately about obtaining information that is good enough to reliably guide decisions. For buyers, a good enough solution is one that allows them to take a decision and move on. While it may be true that this solution implies “Iron Triangle” tradeoffs, the way buyers will often view their constraints is through two conditions: defensibility and workability given their operating constraints. Defensibility is more than just using the right methodology or tools. In some companies, defensibility means “I won’t get fired if other people were to revisit my choice.” Workability means “given my company’s current resources and ways of working, I can adopt this solution.”

This distinction explains why SaaS solutions, despite their efficiency advantages, have not been fully and immediately successful in displacing traditional research models. Unlike software categories that thrive on repeatability, automation, and standardization, market research is a high-variability, expertise-driven, operationally complex function that resists a more templated or do-it-yourself (DIY) approach.

Variability of Needs

A core challenge for SaaS in market research is that while methodologies may be similar across industries, the application of research is highly variable. Companies in the same sector, using the same methodologies, will often have different objectives, data preferences, and approaches to decision-making. Moreover, research is not an isolated technical function: it is an embedded part of a company’s broader strategic and operational framework, shaped by industry-specific dynamics, competitive positioning, and internal corporate culture.

SaaS thrives in environments where problems are repeatable and solutions can be standardized. But many research questions are dynamic, shifting based on business priorities, resource constraints, market conditions, and stakeholder needs. A research platform that doesn’t allow for this will fall short.

The Expertise Gap

Collecting and analyzing data is the easy part of market research. The hard part, where real value is added, is designing a study and then making sense of the results. This requires methodological expertise, business acumen, and an understanding of industry-specific dynamics. SaaS solutions can automate many aspects of research execution, but they struggle to replicate the strategic judgment required to interpret findings, identify meaningful patterns, and translate insights into action.

Indeed, many clients outsource research not just for execution, but also to get access to expertise they do not have in-house. They rely on research partners for operational support and strategic guidance, as well as for advice on what questions to ask, how to interpret ambiguous findings, and how to position insights for internal stakeholders. Even in industries where SaaS has been widely adopted, such as CRM or marketing analytics, human expertise remains critical for implementation, optimization, and extracting value.

Shifting the Labor Burden: Who Does the Work?

One of the most consequential barriers to SaaS adoption in market research is who actually does the work. For there still is work, even if some of the elements are automated. Traditional market research firms provide a full-service offering covering study design, execution, analysis, and strategic reporting. SaaS solutions, by contrast, shift much of the workload directly onto clients. Without internal teams who have both the bandwidth and the expertise to handle this work, even the most sophisticated DIY tools are DOA. Simply put, a powerful research platform alone doesn’t ensure the client can—or will—effectively use it.

And this brings us back to the expertise gap. Even organizations with strong internal research teams want partners they trust to guide them in methodology, interpretation, and business application.

The Procurement and Budgeting Mismatch

Setting aside the operational concerns, SaaS pricing models are frequently at odds with the budgeting and procurement processes for market research. SaaS pricing is built around a licensing or subscription commitment, whereas market research is often project-based, with budgets allocated annually or tied to specific business needs. Unlike CRM or finance software where usage is ongoing and predictable, market research is frequently discretionary, dependent on internal priorities, economic conditions, and short-term business objectives. Moreover, higher expense levels annually (even if they result in reduced costs per project) can create hurdles arising from expense approval levels. Thus, even when companies offer clear advantages in automation and efficiency, they can find it very challenging to break down administrative barriers.

When SaaS Works, and When It Doesn’t

Despite these challenges, SaaS operating models are not inherently incompatible with market research. There are specific conditions where they thrive, particularly when research needs are highly consistent, structured, and repeatable. They work best when:

Research requirements are stable over time, such as brand tracking or customer satisfaction studies.
Clients have dedicated research teams with the skills to operate self-service platforms.
The client organization has a data-driven culture where insights are continuously integrated into decision-making.

By contrast, they usually don’t work when:

Research needs are variable and require regular customization.
Clients lack internal research expertise, making self-service tools impractical.
Budgeting is project-based rather than annual for recurring software expenses.
The work has broad strategic implications that SaaS platforms alone cannot address.

What suppliers invariably find is that revenue growth requires carefully identifying the right customer segments, qualifying prospects who are operationally suited for a SaaS approach, and ensuring that their solutions fit within the constraints of how market research is done and used within businesses. The implication couldn’t be clearer. If market research were only about delivering data, the SaaS transition would already be complete. But because research is a decision-support function and not just a data pipeline, its variability and expertise-driven complexity mean that pure SaaS businesses will struggle to get traction without sharp targeting of likely customers.

The next chapter examines the structural factors that determine whether a research business can operate exclusively with a SaaS model or whether a service layer is essential to long-term viability. Understanding this distinction is the difference between building a high-growth, scalable business and simply launching yet another competent and underutilized research tech platform.

3. Delivery Models

Market research companies and investors often assume that if a technology-driven solution exists to solve businesses’ informational problems, then that business falls within its addressable market. But as we have seen, market research is not just about access to insights; it’s about integrating that information into the customer’s own workflow and business processes. This means the biggest barrier to SaaS adoption in market research is not a lack of demand for insights, but rather a mismatch between how the clients operate and how the insight is delivered.

The Issue Is the Delivery Model, Not the Need

There is no question that companies prefer to make data-driven decisions. Since the dawn of the digital age, businesses have increasingly relied on market research and data more generally to test hypotheses that help reduce uncertainty and provide direction. The need for research is not disappearing, either. If anything, it’s growing. The real issue is finding the right way to deliver those insights.

Market research does not exist in a vacuum. Insights are not a static, standalone output. They are a means to an end, requiring analysis, adaptation, and application. Clients don’t just need data; they need to fit that data seamlessly into their decision-making processes, organizational structures, and operational constraints.

SaaS, in its pure form, assumes that customers can be self-sufficient—that they can log into a platform, extract value on their own, and integrate the solution into their business without significant external intervention. This model works well for tools that solve discrete, well-defined problems—like CRM, accounting software, or project management tools. But in market research, even companies that use SaaS solutions often find they still need services layered on top to make the insights actionable.

Who Actually Buys SaaS in Market Research?

There are really only two groups of companies that are naturally inclined to buy SaaS research solutions:

Small and mid-sized businesses that need a cost-effective solution. These companies can’t afford high-touch research services and are willing to trade depth, expertise, and customization for affordability and speed. SaaS provides them with value-creating capabilities they wouldn’t otherwise have.
Large enterprises that are already structured to produce, support, and integrate insights internally. These companies have mature research functions, embedded analysts, and established decision-making processes that allow them to extract value from software-driven insights without requiring external expertise. In short, they work in a way that allows them to adopt software.

The companies least likely to adopt SaaS research solutions are those that currently rely on full-service research. Notably, these are often large enterprises accustomed to outsourcing not just execution, but often interpretation and strategic guidance as well. These companies therefore aren’t in the market for data. They are seeking reassurance, most often in the form of a team of experts ensuring methodological rigor, identifying key insights, and packaging results in a way that facilitates decision-making. Yes, of course they want to pay less and have results quickly, but not at the expense or perhaps the risk of having to do it all themselves.

When Selling SaaS Means Business Transformation

There is a moment of truth when a SaaS insights provider pitches a client that is accustomed to full-service research. If they have qualified the client properly, they will know that the client is in the market to internalize at least some of what was previously purchased. But if they haven’t, what they will discover quite quickly is that they are no longer selling their product—they are selling a business transformation.

Shifting from services to SaaS implies a significant change in the way a company approaches insights. For an organization used to buying research services, moving to a SaaS model means, at minimum:

Hiring, retraining, or reorganizing staff to manage insights internally
Building new processes around the contracting and delivery of insights
Developing internal expertise in areas like research methodology
Changing budgeting approaches

For some companies, this transformation may be worth it, but for many, it is a mountain too high to climb. Firms selling into this environment must overcome the inertia of the client’s existing operating model, manage the ensuing financial and organizational concerns, and hope the company has an appetite for change. A software platform, no matter how sophisticated, will never eliminate the effort required to achieve this transition. Without internal capability and alignment, even the best SaaS solution will struggle to gain traction. This is why many companies continue to pay for services. It is often the only viable way to ensure insights are gathered in ways they can effectively be used.

Linking GTM to Delivery

Because of these structural realities, the most successful research technology businesses will be those that recognize where services must complement software. Concretely, they will:

Segment their customer base effectively, understanding which of their ideal customer profiles can truly adopt SaaS and which require handholding
Build a flexible delivery model offering different levels of service depending on client capability and needs
Integrate technology with expert support under the recognition that insights require interpretation and usage will ultimately require assistance, and that this often means providing hands-on guidance alongside software tools
Wrap this up in an effective go-to-market strategy that helps clients qualify themselves, ensures salespeople know what to sell, and clearly delineates the service tasks the supplier will need to perform.

The question isn’t whether market research can benefit from technology, but rather how to integrate that technology seamlessly into how clients already operate. New insights businesses can’t blindly chase the SaaS playbook from other industries. Success will require building pragmatic delivery and pricing models that balance automation with expertise, scale with flexibility, and efficiency with usability for each client segment.

4. Will AI Change the Game?

Artificial intelligence has arrived in market research with a renewed promise of transformation. Automation, machine learning, and AI-driven analysis are rapidly reshaping the way data is collected, processed, and interpreted. For those who have long championed the idea of turning market research into a SaaS-driven industry, AI appears to be the missing piece—the technology that will finally remove the human bottlenecks and make insights fully scalable.

This thinking follows a simple logic: if AI can handle the design, execution, and analysis of research (the parts typically done by people), then surely market research can become an automated, software-driven industry just like so many others. SaaS solutions for survey design, social listening, and text analytics already exist, and AI is making them even more powerful. The investment thesis is even more compelling: with AI streamlining every stage of research, why wouldn’t clients embrace a fully self-service model?

Despite rapid advances, though, AI has not—not yet, at least—erased the fundamental constraints that have inhibited non-AI SaaS businesses from making significant inroads in market research. The same key reason stands out. Companies are not buying data. They are buying confidence.

The Limitations of Generalist AI

As we’ve discussed, businesses commission research because they seek reassurance to support decision-making that aligns with their market context, strategic priorities, and operational constraints. Confidence in decision-making almost never comes from more data, faster. It comes from knowing that the insights are valid and contextually relevant to the client’s market and specific business problem.

This is where generalist AI models fall down. While AI can surface patterns, summarize trends, and automate reporting, it lacks judgment. It cannot discern which research questions are most critical for a particular business problem. And, crucially, it cannot anticipate the internal dynamics of a company—the operational, organizational, and strategic factors that shape how information is used and decisions are made.

Thus, while AI can automate many of the mechanical aspects of research, without structured human oversight it risks generating insights that are technically sound yet devoid of strategic relevance. It will surface patterns but not know which ones truly matter. It will summarize findings but lack the intuition to challenge assumptions or spot inconsistencies. It won’t intrinsically possess the expertise required to ask the right questions, interpret ambiguous findings, and ensure insights are both accurate and decision-ready.

The Potential of AI Agents

While generalist AI may not meet a client’s defensibility condition, agent-based systems offer a more compelling path forward. These are AI models that can integrate proprietary datasets, follow structured research methodologies, and work within an organization’s specific workflows. Unlike general-purpose AI, which generates responses based on broad probabilistic patterns, AI agents can be trained specifically on industry knowledge, proprietary company data, and established research practices, allowing them to more naturally align with clients’ workflows and decision-making processes. But even agentic AI still needs people to do this—not just to design and train the system, but to ensure that outputs fit the client’s specific needs and can be trusted in context.

This suggests a future where AI does not replace human expertise but rather changes where and how it is brought to bear. Where many research professionals today spend time on operational tasks, in a near and very probable future they will design, train, and prompt AI-enabled frameworks that produce structured, valid insights that fit seamlessly into their clients’ workflow. AI can (and should) take over much of the heavy operational lifting, but the intellectual scaffolding—the definition of the business problem, the definition of good outcomes, the contextualization of findings, and the final oversight which ensures that outputs align with business needs—will remain human-driven.

Challenges to AI Adoption

As always, though, we must view adoption through the client’s eyes and their journey from consideration to purchase to repeat usage. The use of AI for substantive elements of the research process—whether the AI is used as a substitute for researchers or for research respondents—creates new ways in which clients must satisfy themselves that the work they are hiring out will be defensible and workable. There are three distinct moments where the AI solution will be tested:

Pre-Sales Doubts

Trust issues will first emerge in the sales process. Companies selling to sophisticated clients with complex needs will surely face early skepticism. Will synthetic respondents provide “accurate” information? Is an AI research bot “smart” enough to support strategic decision-making? Will AI succeed when even talented humans can struggle to understand and incorporate the nuances of a client’s needs and business context into the work product? Businesses that cannot convincingly respond to these questions will fail to close deals, indicating that they do not have product-market fit or have not properly qualified the client.

Onboarding and Setup

Can the client realistically adopt this solution? What data and training need to be provided? How is the output structured? Which people and teams need to be involved from the client’s side? How does the system fit in the client’s process? Is setup “one and done,” or is there ongoing non-trivial maintenance of the inputs?

Ongoing Use

Even if onboarding is successful, delivering value over time is far from guaranteed. Clients must still know how to interpret results, how to apply them, and when to challenge them. They may need to update the system’s data and training. Strategic oversight is required—not just to keep the system producing outputs, but to ensure those outputs lead to better decisions.

And this is where traditional approaches to Customer Success (CS) will fall short. Most CS teams are built to manage operational tasks like onboarding and support tickets. They are not usually designed to deliver the kind of strategic guidance that clients need to extract defensible, high-value outcomes from AI-based solutions.

The Emergence of the Strategic Enablement Role

To address the challenges, suppliers will need to create a new role—a human capable of bridging AI capability and business relevance. Let’s call these people strategic AI partners, who will be one part applied AI architects and one part industry consultants. They will understand how agentic AI models are built, trained, and interpreted. They will also understand the client’s domain well enough to know which inputs are needed to deliver decision-ready insights.

We can expect these strategic AI partners to become critical to a firm’s GTM as well. Their knowledge can be used to create mid-funnel buyer- and sales-enablement materials. This also implies that the firm has clearly defined its ideal customer profiles. Clients with complex strategic needs will benefit from a sales-led conversation around trust and credibility, whereas other clients may not need (or be able to afford) the same handholding. These strategic AI partners will serve as trust-builders and enablers. Perhaps most importantly, they will be a firm’s competitive differentiator.

Two Operating Models for the AI Future

By walking through the client journey, a pattern emerges. There are two operating models that can succeed, depending on who provides the human scaffolding:

Model 1: Platform + Strategic Support

In this model, the supplier delivers an AI-powered platform but retains a layer of strategic human support. The client is expected to do more of the execution, but the supplier provides expert guidance during pre-sales, onboarding, and usage. This model works when clients are somewhat capable but still require interpretive help. The strategic AI partner becomes the linchpin.

Model 2: AI-Powered Agency

Here, the supplier uses AI internally as a force multiplier to enhance its own delivery while continuing to provide a service-led experience. The client doesn’t need to change how they operate. They still receive consultative, bespoke work—but it’s delivered faster and more efficiently thanks to the AI under the hood. The client hires reassurance; the supplier uses AI to scale it.

Which Model Will Clients Choose?

AI does not eliminate the need for human judgment and expertise, but it does impact how and where that expertise is brought to bear. Rather than removing humans from market research, AI will change their role. On the supplier side, human expertise will provide essential domain knowledge from which specialized agentic solutions can be developed. On the client side, humans will surely oversee and guide AI-based systems and then provide a strategic lens to translate outputs to the business.

Will buyers of market research continue to value external expertise and therefore purchase from companies that offer it, or will more companies internalize these capabilities, leveraging AI directly within their own teams? Both outcomes are possible, and AI will undoubtedly produce valuable outputs in either scenario. But the assumption that AI alone will turn market research into a purely software-driven, service-less industry remains flawed—yet that may be a good thing. Because what clients ultimately seek from suppliers goes beyond automation and insight. They want judgment they can trust and insight they can use. Successful firms will be those that combine human expertise with AI capability to help clients confidently make decisions given their specific business contexts and needs.

5. The Investment Case for AI-Driven Market Research

AI will change, but not replace, human expertise in market research, namely where and how this expertise is applied. The firms best positioned for long-term success will be those that blend powerful AI with deep domain knowledge, embedding themselves deeply in client workflows to ensure insights aren’t just generated but actively used to drive confident decisions.

For investors, this creates a significant opportunity. AI-driven research firms promise increased efficiency, improved scalability, and a broader addressable market by democratizing access to sophisticated insights. However, capturing this promise will not be as simple as copying a SaaS playbook from other industries. Instead, businesses will need to rely on a hybrid model, combining technology with a lightweight, targeted human service component to facilitate adoption, integration, and sustained value creation. This means success—and therefore the investment thesis—depends as much on a firm’s customer success structure as it does on the quality and scalability of its technology.

Evaluating AI-Driven Research Businesses

The investment rationale for AI-driven market research firms hinges on three key dimensions:

1. AI as an Organizational Asset

Successful firms will go beyond general-purpose AI models, embedding proprietary domain expertise, unique datasets, and specialized algorithms. Investors will therefore need to carefully evaluate how proprietary and differentiated a firm’s AI really is, as this will contribute to its moat and adoption.

2. Scalability through Land-and-Expand

Long-term viability and growth depend heavily on how seamlessly AI solutions can be integrated into client workflows and become part of the client’s operational DNA. Firms capable of driving expansion within client organizations—moving beyond insights teams and into broader decision-making areas such as marketing, product, and executive strategy—will show stronger growth, better retention, and more profitable client relationships.

3. Economic Viability Beyond ARR

Annual Recurring Revenue (ARR) is an imperfect indicator because many clients simply do not purchase insights in subscription-based models. Project-based budgets, procurement practices, and organizational constraints often mean that ARR is only relevant for a SaaS Ideal Customer Profile (ICP). Investors should therefore pay close attention to other metrics, like:

Net Revenue Retention (NRR): A stronger measure of client satisfaction, reflecting expansion and deeper integration over time.
Contribution Margins by Segment: Understanding unit economics is crucial—especially since high-profile enterprise clients often require significant service, potentially reducing margins without careful management.
Users per Client / Share of Client Spend: Indicates how deeply embedded the product becomes across departments beyond the insights team, signaling both scalability and integration.
Funnel Metrics and Acquisition Costs: Particularly the progression from Marketing Qualified Leads (MQL) to Sales Qualified Leads (SQL) and eventual conversion. Poor funnel conversion indicates misaligned targeting, unclear value propositions, or underestimated implementation challenges.

Aligning GTM Strategy with Client Segments

One of the most interesting aspects of the AI revolution will be whether, finally, market research will become broadly accessible to the long tail of the industry. Accessing these unserved and underserved clients will require a distinctly different go-to-market (GTM) approach compared to traditional market research. One would expect smaller and mid-sized customers to have lower average tickets but higher volumes. This would suggest a product-led or marketing-led GTM strategy: one that is low-touch, scalable via automation, and driven primarily by product experience and digital marketing channels, where success requires streamlined user onboarding and self-service adoption.

Larger enterprise clients will continue to be won by a sales-led GTM strategy, characterized by higher-touch, consultative processes with dedicated sales and customer success resources. This GTM will mean longer sales cycles, deeper client relationships, and investment in experienced sales teams who are also capable of navigating complex procurement processes.

Each GTM strategy thus implies fundamentally different organizational structures, resource allocations, cost bases, and investment profiles. Investors must clearly understand how an AI-driven research firm aligns its GTM model with its target customer segments to evaluate its long-term growth and profitability prospects accurately.

Final Thought

AI is fundamentally changing how market research firms create value—but it doesn’t eliminate complexity. For investors and founders, the critical task is understanding precisely how AI fits into the practical realities of client organizations. The most valuable AI-enabled firms will be those that thoughtfully align their products, service offerings, and go-to-market strategies with how their customers actually work and buy, rather than assuming AI alone can overcome these operational challenges.

Conclusion

The allure of turning market research into a scalable, SaaS-like business has captivated investors and founders for years. The vision of predictable, high-margin revenue streams, automation replacing expensive human labor, and frictionless adoption across enterprise clients is compelling. And yet, as we’ve seen, this vision is largely a mirage.

Suppliers haven’t failed to gain traction because their products are bad, but because they have misunderstood what clients are looking to buy. Market research is not simply an informational problem; it is a high-variability, expertise-driven function that sits at the intersection of data, business context, and operational realities. While technology can accelerate processes and improve efficiency in ways clients appreciate, clients need solutions that are defensible and workable in their environments.

This is where AI presents a new and more promising path forward. Unlike traditional SaaS platforms that rely on consistency and self-service delivery, AI—and particularly agent-based approaches—offers a way to blend automation with expertise, enabling research firms to scale while still delivering high-value, bespoke solutions. But for AI-driven businesses to succeed in market research, they must be built on the right business model, one that acknowledges the client’s inherent complexity rather than trying to bypass it.

The Businesses That Will Win

The market research firms that thrive in the AI era will not be those that attempt to force SaaS onto a service-heavy industry. Instead, they will be those that embrace a hybrid model, leveraging AI to reduce inefficiencies while retaining the necessary human expertise to deliver true value.

Winning businesses will be structured around:

AI as an Asset – Not just generic AI capabilities, but proprietary, differentiated technology and data that enhance decision-making rather than just accelerating data processing. With foundational AI models becoming increasingly accessible, differentiation will depend on unique datasets and specialized tuning for industry- and client-specific issues.
Market Fit and Buyer Alignment – The best businesses will design their offerings to fit how clients actually work, ensuring adoption by the right customer segments and an ability to “land and expand” rather than attempting to force SaaS adoption where it cannot take root. This implies a flexible economic model, as well. Where clients are unable or unwilling to make annual revenue commitments, net revenue retention, usage rates, and service-enhanced pricing models will define the most valuable businesses.
Strategic Enablement as a Capability – Winning firms will build teams with the dual expertise to make AI work: an understanding of how AI systems operate, and a grasp of the client’s business and industry context. These roles will be essential to aligning outputs with decisions and embedding AI into client workflows.

Parting Thoughts for Investors and Founders

For investors, the most attractive businesses in market research may not be pure SaaS. They will be those that overcome clients’ skepticism and genuine operating concerns because they know which clients they serve, how those clients operate, and where their product adds real value. The rest will be burdened by fragile operating models that inevitably limit their growth and profitability.

For founders, this means resisting the urge to pitch a SaaS story unless the model truly fits. The real work is understanding how clients operate, building a solution that meets their needs and ways of working, and then qualifying them carefully. Without this discipline, you may still find growth, but it will be harder and more costly to deliver value at scale.

The future of market research is neither purely SaaS nor purely services. Instead, it will be defined by firms that skillfully blend AI-driven automation with human expertise and build solutions tailored to their clients’ operational realities. The next era will belong to companies whose products not only deliver insights, but reassure clients by helping them make smarter, faster decisions in the way their businesses already operate.

Summary

The full text