How to think with data and share it ethically
Hello! My name is Larriza Thurler, and I’m excited to share my journey as a Research Fellow at the Open Data Institute (ODI). In this role, I have delved into the world of data, focusing on its power to help solve organisational challenges and deliver actionable insights. My mission has been to develop a framework that guides organisations in unleashing the full potential of data to improve decision-making and problem-solving across sectors and cultural contexts.
The heart of my research
As an ODI Research Fellow, I developed a prototype version of an online canvas called the Data Thinking Journey Canvas which empowers organisations to use data to address a range of challenges, such as the high absenteeism rate in a school or a store in need of improving its customer relationships. This framework emphasises ethical and responsible data usage and guides users through the process of data-informed thinking.
The theoretical background of my research is the Data Thinking Journey methodology, originally developed by myself and other researchers at the Strategic Intelligence Reference Centre at the Federal University of Rio de Janeiro (CRIE) in Brazil. This methodology emerged from a series of workshops and training sessions, proving particularly effective in facilitating discussions related to data. For instance, it has been instrumental in a non-profit institution dedicated to improving the quality of public education, where it guided stakeholders through processes that helped them make data-driven decisions to enhance educational outcomes and operational efficiency.
Through collaborating with researchers at the ODI, I was able to evaluate the most useful and appropriate ODI tools to be integrated into the methodology, such as The Data Ethics Canvas, Data Ethics Maturity Model, Assessing risk when sharing data: a guide, Data Ecosystem Mapping Tool, The Data Spectrum, and the Data Skills Framework. These discussions offered valuable insights that helped in formulating prompts and identifying additional resources, which are essential for promoting responsible data sharing and collaborative decision-making informed by data.
I envision the Data Thinking Journey Canvas as a framework that facilitates collaborative solutions. It fosters discussions focused on problem identification and data exploration, effectively merging the principles of design thinking with data science. Rooted in design thinking, it combines a user-centered, iterative approach with the Double Diamond model’s divergent and convergent thinking phases. Thus, it supports collective discussion and co-creation, sparking expansive idea generation in divergent phases like brainstorming, and guiding decision-making in convergent phases such as consensus building. The Data Thinking Journey Canvas also adapts the stages of data science, which include formulating pertinent questions, acquiring and exploring data, critically analysing the data, and communicating insights to transform them into actions.
Why the Data Thinking Canvas matters
The Data Thinking Journey Canvas is user-centric, collaborative, and iterative, and it helps teams structure conversations about data challenges and opportunities. It systematically guides users through identifying challenges, sourcing data, formulating hypotheses, planning communication, and creating action plans — crucial steps illustrated in Figure 1.
It does so by stimulating questions such as:
● What is the real problem we’re facing, considering data and evidence?
● How does our organisation use and access data? Does the organisation allow data to speak for itself, or is it in the habit of overlooking valuable data insights?
● Do we have the necessary skills to acquire, interpret, and present data effectively?
● Are we taking adequate steps to safeguard against biases and ensure that our use of data benefits all stakeholders without unintended harm?
It also provides a framework for documenting both divergent and convergent discussions, namely the brainstorming phase and the consensus-building phase, as well as the action plan. It is organised in the following steps:
- Identifying the Challenge or Opportunity: This step involves recognising and clearly defining the specific challenge or opportunity at hand. It’s about understanding the context and the problem or potential that needs to be addressed using data.
- Listing Data Sources: After identifying the challenge or opportunity, the next step is to identify potential data sources that could provide insights. This includes both internal and external data sources, and it’s important to consider the relevance, quality, and accessibility of the data.
- Proposing Hypotheses: Based on the challenge or opportunity and the available data, this step involves formulating hypotheses. These are educated guesses or assumptions that will be tested through data analysis to gain insights.
- Creating a Communication Plan: Effective communication is key in any data-driven project. This step involves planning how to communicate the findings and insights derived from the data to stakeholders, ensuring clarity and alignment with the project’s objectives.
- Creating an Action Plan: The final step is about turning insights into action. This involves developing a plan for how to apply the insights gained from the data analysis to address the identified challenge or capitalise on the opportunity. This plan should be actionable, measurable, and aligned with the overall strategy and goals of the project.
Let me illustrate the possible application of the Data Thinking Journey Canvas with some examples from two randomly chosen sectors for educational and explanatory purposes. For example, in the first step, a school may ask questions like ‘How can data help our school address the high absenteeism rate?’, or a toy store may ask something like ‘How can data help us enhance our customer relationships?’.
Moving to the second step, it’s time to identify relevant data sources. For tackling school absenteeism, the school would consider attendance records, student demographic data, academic performance, health records, and survey data. In the second example, for improving customer relationships,the toy store would list sources like CRM system data, sales figures, customer feedback, website analytics, and demographic and psychographic information would be invaluable.
The third step involves hypothesising. For example, the school might explore if higher absenteeism correlates with higher local crime rates or if socioeconomic factors like work commitments impact attendance. In customer relationships, the toy store might hypothesise that faster responses improve customer satisfaction, which in turn boosts loyalty and referrals. These hypotheses can be validated later, with the data on hand, time, and the appropriate team. It’s relevant to think about them now since a diverse group of people will be thinking together and providing different perspectives.
The fourth step focuses on communication planning. Here, strategies would be devised to present findings effectively depending on the target audience, such as creating visual data representations for school boards in the school case or developing clear, actionable reports for marketing teams in the toy store case.
Finally, the fifth step is about action planning. For school absenteeism, this could involve developing targeted intervention programs for at-risk students or community engagement initiatives. To enhance customer relationships, actions might include implementing a new customer service protocol or revising marketing strategies based on customer feedback analysis.
The benefits of using the Data Thinking Journey Canvas
● Improved problem-solving: By structuring and analysing challenges with data and evidence, we gain deeper insights into root causes and can devise more effective solutions.
● Growth in impact and trust: Ethical data use is paramount. The canvas emphasises privacy, confidentiality, and bias avoidance, ensuring that data use respects individual rights and has a positive societal impact.
● Increased accuracy and efficiency: Data-driven decisions reduce guesswork and enhance precision, helping spot trends and anticipate risks.
● Enhanced team engagement: The canvas exercise fosters collaboration, bringing together diverse perspectives for holistic and innovative outcomes.
The Canvas can be used across an array of challenges. For instance, in digital transformation, it can be used to guide companies in integrating new technologies and managing change effectively by leveraging data-driven insights over assumptions or opinions. In the context of market expansion, the Data Thinking Journey Canvas may aid organisations in analysing consumer behaviour and market trends, allowing them to make data-informed decisions about new market entry strategies and product positioning. Regarding supply chain optimisation, it enables businesses to analyse logistical data and supply chain operations, helping them identify bottlenecks, predict disruptions, and optimise inventory management through insightful data analysis.
How to implement the Data Thinking Journey Canvas
The process involves a series of steps, from problem framing to actioning insights, ideally completed in groups of 4 to 8 people. We recommend involving professionals with a range of perspectives, from diverse backgrounds and expertise, that have a common challenge to solve or opportunity to explore. It’s not about answering every question but using them to spark insightful discussion. We start with divergent thinking, noting ideas on sticky notes, before moving to convergent thinking, where we synthesise these ideas. The suggested timeframe ranges from 2 to 4 hours, ensuring a thorough exploration of each step.
You can find the Data Thinking Journey Canvas here on Miro. You can make a copy of it and use it with your team; just remember to give credit for the inspiration. Here are some recommendations.
● Start with Step 1: The journey through the Data Thinking Canvas is sequential, moving from Step 1 to 5. Each step is critical and builds upon the previous one.
● Engage in thoughtful discussion: For each section, it’s important to delve into the instructions and engage in rich discussions with your group. The questions provided in the cards are not mandatory but serve as catalysts for deeper thinking and debate.
● Divergent phase: Here, all ideas are welcome. Use this phase to brainstorm and jot down thoughts freely, using virtual post-its in the Miro template. This is about exploring all possibilities without judgement.
● Convergent phase: This phase is about distilling the ideas from the divergent phase. Summarise the discussions and articulate the conclusions drawn, marking the transition from broad thinking to focused solutions.
As you navigate the journey, you’ll encounter various recommendations of relevant ODI tools and resources, each shedding light on crucial aspects of data, such as data ethics, data ecosystems, data skills, and open data. If you have time, I recommend using the ODI tools suggested; they will improve your data analysis and use.
An essential element in navigating the Data Thinking Journey Canvas is the role of the facilitator. This individual is the guiding force of the journey. The facilitator’s expertise is not just in knowing the canvas but in bringing it to life, making the process dynamic, collaborative, and productive. It does not have to be someone from outside the organisation, it can be someone from your team. An external facilitator, however, brings fresh perspectives and impartiality, often serving as a catalyst for innovation and challenging biases.
What comes after
Completing the Data Thinking Journey Canvas is just the beginning; it’s the first step towards creating a tangible action plan (see the board in Figure 3). This plan, born out of the insights and conclusions drawn from the canvas, will serve as a roadmap for implementing data-driven strategies within your organisation.
By continuously promoting data literacy, engaging in technological training, and involving leadership in data-driven decision-making, the insights gained from the canvas can translate into concrete, actionable steps.
Another approach to use the Data Thinking Journey Canvas is through a specialised GPT I’ve developed. Those with access to the premium version of ChatGPT, can experience this guided journey, where the GPT offers step-by-step instructions and insightful explanations for each stage of the process in a fluid conversation. Here is the link. In the initial tests made, the insights provided were really inspirational. It’s important to say, though, that it’s an experimental tool and the accuracy of recommendations may vary. Moreover, we don’t recommend submitting confidential information, since we are not sure how OpenAI deals with it.
Conclusion
My journey as an ODI Research Fellow has been an enlightening experience, uncovering the immense potential of data to transform how organisations think and act. Embarking on the Data Thinking Canvas journey is more than a methodical process; it can help us approach problems and make decisions in our organisations.
Bridging data science with design thinking can open up a world of possibilities — smarter decisions, more efficient processes, and a deeper understanding of the challenges. My team and I are now in the final stretch of the research, and I would love to hear about your experience using it. Please feel free to reach out to me at larriza@gmail.com for feedback, inquiries, or if you are interested in receiving training. In exchange, I would request your permission to report the results anonymously for inclusion in an upcoming academic paper.
