How Coinbase Integrates AI Insights into Its Go-to-Market Strategy

NewtonX helped Coinbase reach 18 senior fintech decision-makers in just three weeks, uncovering what drives partner preferences for crypto capabilities.

Coinbase was interested in qualitative research with a hard-to-reach group: senior product and engineering decision-makers.

Coinbase aimed to understand what drives partner preferences for crypto capabilities and offerings. To shape go-to-market strategies and product positioning, they sought input from an elusive group: in-house product and engineering leaders influencing crypto capability decisions at top financial institutions in the US and the UK.

In a fast-moving space, speed mattered. With AI-moderated research, NewtonX recruited, interviewed, and delivered insights from 18 decision-makers in just three weeks.

Spotlight: What is AI-moderated research, and what are the benefits?

AI-moderated research is a new option for running qualitative research. Instead of a human, AI acts as the moderator: presenting the questions, asking follow-up questions, and requesting clarification when needed. On the backend, AI processes all of the study’s interviews, coding them and beginning the work of turning the responses into a storyline.

AI-moderated qual is a major tool in our research toolkit. It sits comfortably in-between how we think of traditional quantitative and qualitative methods, blending benefits of each.”
— Mat Duff, Lead Researcher: Developer & B2B Insights, Coinbase 

With AI-moderated qualitative research, you can obtain rich insights on a similar level to traditional qualitative in less time and budget. We’ve found you can complete five times as many interviews with an AI moderator than a human moderator with the same resources. These higher sample sizes allow for robust quantitative analysis and for surfacing niche themes that wouldn’t be otherwise apparent. 

Lastly, AI moderation can reduce the scheduling friction in international research; respondents can complete their interview at their convenience.

Coinbase tapped NewtonX to conduct qualitative research with fintech decision-makers

NewtonX quickly identified the exact target sample: 18 senior product and engineering decision-makers at financial technology firms (fintechs), banks, and financial brokers in the United Kingdom, the United States, and California. From ask to delivery, the project took three weeks. 

NewtonX quickly found the exact sample target, and the respondent UX was excellent, driving higher quality responses than traditional surveys.”
— Mat Duff, Lead Researcher: Developer & B2B Insights, Coinbase

Together, Coinbase and NewtonX developed a screener and interview guide to train the AI platform moderator. Participants completed their AI-moderated interviews asynchronously, recording video or audio responses at their convenience. The platform then coded their responses, delivered a quantitative readout and direct quotes, and generated a draft report that NewtonX and Coinbase refined into a cohesive insights deck.

Coinbase quickly gathered key insights to inform its strategy, drawing on feedback from potential clients.

The study revealed what influences fintechs and financial institutions to decide whether to build, buy, or partner for crypto capabilities, as well as the benefits and risks of these modes of expansion. This provided Coinbase with clear takeaways on what uniquely positions them to win further partnerships.

Three keys to pitch and plan your own AI-moderated qualitative research project

1. Emphasize opportunities

AI can help address things that have long been issues with traditional surveys and qualitative research. For instance, the AI moderation experience can help address widespread “survey fatigue” among survey respondents, which causes low quality quantitative data. 

When done at scale, AI can reintroduce opportunities for more dynamic conversations with respondents, including unaided questions. 

It can also allow for a more holistic interpretation of insights, moving beyond the limitations of discrete multiple-choice questions.

Lastly, AI moderation enables you to get larger samples for qualitative research. Larger sample sizes can give you more confidence in the findings, help surface valuable niche insights, and enable richer comparisons between sub-audiences.

By reframing AI as a way to restore depth and spontaneity to qualitative research, you’ll help stakeholders see it as an enhancement and complement to the traditional method.

2. Keep humans in the loop

In your pitch, emphasize that AI moderation doesn’t mean cutting humans out of the workflow. Rather, it frees them to focus on the parts of research that require creativity, empathy, and judgement, while AI handles the repetitive mechanics. 

Human expertise is still essential  for: 

• Understanding the richer context of your stakeholders’ requests
• Determining which topics are suited for AI moderation (and which aren’t)
• Interpreting the results within your organization’s unique context

Expect some trial and error as you find the best fit for AI moderation in your research toolkit. Embrace that process, because it’s how you’ll learn where automation creates efficiency and where human abilities are critical. 

3. Plan for internal resistance

As with any new methodology, expect to advocate for AI and be ready to explain both its benefits and its limitations. Common concerns include:

• The risk of fake respondents when a human moderator isn’t verifying identity.
• The potential for less authentic answers without live human interaction.
• Perceived weaknesses in the research process whenever automation is involved.
  

These concerns are why quality recruitment matters. In human interviews, moderators can root out unqualified participants early. An AI moderator can’t, which means that starting with verified participants—people who actually have the expertise and experience you need—is essential. By using NewtonX, Coinbase was able to reach the exact right audience by market, job function, seniority level, and decision-making authority. 

For all possible objections, prepare clear responses and frame the conversation around what you gain, not what you lose: larger samples in less time and at lower cost.

Go-to-market strategy and product positioning should start with rich insights from your target audience. If you’re in planning mode, get your project started with NewtonX.