Hurry up⚠️ Only 100 tickets left in this pool WaysAwardsSubmit your WaysAwards project — enter the 2026 nominations

Research & Analysis

The Power of Atomic UX Research

September 20, 2024 11:10 AM

Long Lecture

🇬🇧 English

Bratysława 2

About

Atomic Research is a simple but powerful process which allows you to get richer insights from your research, combine with existing learnings, and makes it easy to share and discover knowledge in a large organisation.

The creator of this process, Daniel Pidcock, will be talking through the process, how it came about, and some tools you can use to implement it in your own work. With time for Q&A at the end.

Talk by

Photo of Daniel Pidcock, Ccreator of Atomic Research & Founder of Gleanly

Ccreator of Atomic Research & Founder of Gleanly at Independent

Watch the full talk

Watch this WaysConf session, then continue with related talks or explore the current programme.

Explore the 2026 agenda More WaysConf recordings

From the recording

Talk in brief

In this WaysConf 2024 presentation, Daniel Pidcock, the creator of Atomic UX Research and founder of Gleanly, explains how breaking down user research into its smallest components can solve the "Google Grave" problem where valuable insights are lost in traditional reports. Drawing inspiration from Brad Frost's Atomic Design, Pidcock outlines a four-part framework consisting of Experiments, Facts, Insights, and Recommendations. This structured approach separates objective evidence from subjective opinions and actionable steps, making research highly scalable, collaborative, and accessible to non-researchers. Pidcock illustrates the practical application of this methodology with real-world examples, including a fashion brand's analysis of consumer preferences and a criminal investigation team's evidence tracking. Ultimately, he argues that research repositories must be designed primarily for decision-makers rather than researchers, ensuring that data directly drives faster and more accurate business decisions.

Key takeaways

01
The Google Grave Phenomenon
Traditional research reports stored in static folders like Google Drive or Confluence often go unread and are lost to the organization, leading to wasted resources and repeated studies.
Watch from 1:10
02
The Four Atoms of Knowledge
Atomic UX Research structures information into four distinct, connected elements: Experiments (how we learned), Facts (objective quotes, observations, or statistics), Insights (opinions on cause and effect), and Recommendations (actionable next steps).
Watch from 7:51
03
Separating Evidence from Opinion
Forcing teams to define an Insight before jumping straight from a Fact to a Recommendation prevents biased decision-making and encourages deeper, more creative problem-solving.
Watch from 9:39
04
Prioritizing Quantity Over Quality
Allowing non-researchers to contribute to a repository is vital because poor-quality research is already happening and being acted upon; bringing it into the open allows researchers to identify and correct it.
Watch from 35:14
05
Repositories Exist for Decision-Makers
The ultimate value of a research repository is not to serve researchers, but to enable business stakeholders to make better, faster, and more evidence-backed strategic decisions.
Watch from 38:52

Video chapters

0:10The Challenge of Lost Research
Daniel Pidcock describes his experience at Just Eat where valuable research was buried in Google Drive, leading to duplicated efforts and lost organizational knowledge.
4:41The Genesis of Atomic UX Research
Pidcock explains how collaborating with other industry teams and drawing inspiration from Brad Frost's Atomic Design led to a new way of structuring research knowledge.
7:51The Four Core Components of the Framework
The talk breaks down the atomic model into its four fundamental building blocks: experiments, facts, insights, and recommendations.
12:08Visualizing the Stakeholder and Synthesis Views
Pidcock demonstrates how the atomic model can be mapped out to show decision-makers the exact chain of evidence supporting or disproving a recommendation.
22:12Implementing Atomic Research with Tools
The presenter reviews various tooling options for adopting the framework, ranging from simple whiteboards and Air Table templates to dedicated repositories like Gleanly.
33:21Applying the Framework Beyond UX
Pidcock discusses how the atomic model is actually a universal knowledge framework, sharing examples from fashion retail and criminal investigations.
39:37Q&A: Context and Automation
The session concludes with an audience Q&A addressing how to preserve context during synthesis and how to manage automated quantitative data feeds.

Read edited transcript highlights

These concise notes were edited from automatic captions and checked against the talk structure. They are not a verbatim transcript.

The Context Problem in Traditional Repositories

When analyzing why traditional research repositories fail to scale, the issue of context is paramount. Traditional, long-form reports contain a wealth of context, but this very detail often buries the individual, actionable insights. Conversely, many standard insight repositories strip away too much context to make the data searchable, rendering the findings incomprehensible to anyone outside the immediate team. To solve this, a successful repository must find a way to isolate individual nuggets of information while still preserving the essential background that makes them valid and understandable across different departments.

Watch from 3:00

The Power of Forcing an Insight Step

It is incredibly common for teams to want to jump directly from a factual observation to a product recommendation without defining the underlying reason. However, forcing contributors to explicitly state an insight—explaining the cause and effect behind the facts—is one of the most valuable aspects of the atomic process. Even when a recommendation seems completely obvious, taking the time to articulate the insight often reveals hidden complexities, prompts more creative solutions, and provides a clear historical record for future team members who want to understand the original rationale.

Watch from 9:39

A Real-World Lesson in Defining Facts Versus Insights

A French fashion brand using the atomic process initially miscategorized a survey finding—that sixty-four percent of respondents preferred green clothing—as an insight. In reality, this statistic was simply a fact. By treating it as a fact and combining it with sales data showing a massive spike in green clothing purchases, the team was able to hypothesize on the deeper insights regarding fashion trends and branding. This creative synthesis led them to run a homepage experiment with green imagery that successfully boosted their conversion rate by over four percent.

Watch from 29:12

Why Repository Quantity Trumps Quality

While researchers often worry that opening a repository to non-researchers will dilute the quality of the data, restricting access is actually more dangerous. If untrained employees are conducting poor-quality research, they are already using those flawed findings to make business decisions. By lowering the barrier to entry and allowing everyone to contribute to a single repository, professional researchers gain visibility into these activities. This transparency allows the research team to step in, offer constructive guidance, and help improve the organization's overall research maturity.

Watch from 35:14

Preserving Context Through Clear Taxonomy

During the audience Q&A, a participant asked how to prevent losing the original context of an experiment during the synthesis process. While some tools have attempted to use complex algorithmic scoring to weight evidence based on age or business area, the most practical solution lies in human taxonomy and clear writing. When documenting atomic elements, researchers must write them so clearly that a colleague from a completely different department or country can instantly grasp the boundaries and relevance of the data.

Watch from 39:52

Explore WaysConf 2026

View the agenda

Get tickets