The Secret of Getting Stakeholders to Trust Your Numbers
A Simple Framework for the Most Complicated Part of Data Analysis.
Data analysis is a two-part process. You perform technical work to find patterns and insights in data, and it is half the job.
Explaining it is the other half. You turn raw data into clear messages, and effective communication is a link here. It connects numbers to people who use information in order to act. This process relies on your ability to turn numbers into understandable findings.
Communication is the final stage of the data lifecycle, where you transform findings into actionable insights.
Understand Your Audience:
The success of your project depends on collaboration with stakeholders. Identify your stakeholders at the beginning of every project:
Primary stakeholders use findings to make decisions.
Secondary stakeholders depend on your work for their own tasks.
Executives focus on high-level business impact and return on investment.
Product teams want to see customer behaviors and market insights.
Technical teams want details on methodology and model performance.
The general public needs simplified and engaging narratives. Use analogies and infographics to explain trends. Use plain language for non-technical staff.
Answer these four questions before you prepare a presentation:
Who is your audience?
What do they already know?
What do they need to know?
What decision do they need to make?
Look at the preferences of your audience. Some prefer visuals, while others prefer tables. Adjust your delivery to fit the needs. If people know you and trust you, they will use your work.
Take your time at the beginning of every project to identify stakeholders and their goals. Identify who else is on your team and what their roles are. Know who is managing the data. If you communicate with other analysts, you avoid spending time collecting or analyzing data yourself.
Define the Problem:
Identify the problem you solve and determine which business outcomes you impact.
Align your approach with business strategy. Ask how your work helps stakeholders. Defining success involves measurement planning. Ask what a successful outcome looks like for the business. Is it more revenue? Is it better employee retention? Identify the metrics to quantify the success.
Do not skip a problem definition. It will lead to a lot of trouble later. So, connect with the person and ask questions about their needs.
Know the Data:
You should know your data to describe the business. Understand why this data is relevant to the field.
Exploratory data analysis helps you learn about a problem. Your stakeholders will listen if they feel you understand their business. Expertise in your field is necessary. If you work in banking, you should know economics basics. If you work in medicine, it is better to understand the medical field at some level. Audit your data for quality issues. Make limitations explicit.
Explain why you performed an action or made a certain choice.
The Three C strategy:
In communication, you can implement a strategy called the Three Cs: Connection, Curiosity, and Clarity.
Connection:
Connection involves the human element of your work. You acknowledge the importance of the insights to the stakeholder. You have a conversation to build a relationship.
Trust is built through reliability. You should be empathic with your stakeholders in order to understand their attitudes and frustrations. User experience personas represent the needs of people who will use your work. You focus on human pain points in every deliverable.
Curiosity:
Curiosity means that you want to ask questions. You look for understanding of the business need behind a request.
Identify the question the stakeholder wants to answer. Think like a business owner to find the problem. Look under things and ask questions before you start an analysis. You find out what people care about and what causes them pain. Curiosity helps you understand how data flows through a system, and you use the “so what” test to verify the relevance of your metrics.
If a metric does not lead to an action by the third time you ask “so what,” you remove it from the report.
Clarity:
You confirm your understanding of the stakeholder’s answers. You define what you will deliver and provide a deadline.
Being clear about your needs is a part of effective communication. You set realistic goals to manage expectations. You provide a high-level schedule with project phases. You define success metrics to set a roadmap. Summarizing key insights upfront respects the audience’s time.
Structure the Message:
Lead with the answer. Start with an executive summary or conclusions first to grab attention. Most leaders do not need to know the process before they hear the answer.
Avoid technical details in presentations. Business units work with estimates.
Focus on what is relevant for business decisions.
How will the project improve the business?
What are the main takeaways?
What is the plan for the future?
Put everything else in a technical document.
Use plain language to be sure that your recommendations are understood.
Engage in two-way conversations during your delivery. If you face conflict or mismatched expectations, reframe the problem. Ask stakeholders how you help them reach their goal.
Effective Data Visualization:
Visuals help people understand information faster. Approximately 65% of the general population prefers visual learning styles. Graphs let audiences digest large amounts of information quickly, so use the right graph for the job.
Use line charts to show trends over time.
Use bar charts to compare categories.
Use scatter plots to show relationships between variables.
Keep visuals simple.
Remove useless gridlines or decorations to decrease the ink-to-data ratio.
Do not clutter graphs with unnecessary details. One main idea per chart is enough.
Use short, readable labels for axes.
Place labels next to data to reduce eye movement.
Provide context with reference points like trend lines.
Use color with purpose only.
Use ordered bar charts and clean layouts. Items arranged on a line or curve seem related to each other. Things that look alike seem like part of the same group.
Use the same color for all bars in a chart unless you highlight one point. Place labels next to data to reduce the eye movement.
Storytelling:
Storytelling connects data to real situations, and it makes findings memorable.
A successful data story follows a path:
Start by setting the scene.
Explain the problem or question.
Guide the audience through the analysis.
End with recommendations for action.
Use the three-act structure:
Act one is the setup. Define the conflict and describe the data set.
Act two is the development. Escalate the conflict until there is no way out.
Act three is the resolution. Solve the conflict to show what business decisions could be made. Without change there is no story.
Identify a protagonist in your story.
The character should be someone the audience will root for. It can be you, your team, or your clients. Describe the struggle you went through to make the project work. Mention the victories as well.Use story-driven titles for your charts. Instead of a generic title, use a headline that relays the main insight. A good story has a hook to pull the audience in.
Demonstrate why they should care about it at the beginning.
The Reality of Imperfect Data:
Data does not come out perfect. It contains errors and exceptions. You should spend time with the data to understand how it works.
You provide your opinion, perspective, and what you inferred from the numbers. People have their own views when looking at data. Your responsibility is communicating the meaning behind the numbers. Highlight the significance of findings and state your key assumptions. Data has limitations. You sometimes lack access to data you need. Data sources are not aligned, and data is unclean. These issues affect your communication with stakeholders.
Balance stakeholder expectations with what is possible for a project. Set realistic, objective targets. Set expectations for a realistic timeline.
Cognitive Biases:
Confirmation bias occurs when you focus only on data supporting your belief. Anchoring bias happens when people rely on the first piece of information they receive. Historical benchmarks skew how someone sees current data. People overestimate the importance of easy-to-remember information.
Survivorship bias involves focusing on things that survived a process while ignoring those that did not.
Reduce bias to guide the decisions based on facts:
Highlight strengths and weaknesses.
Present both positive and negative trends.
Provide broad context with industry benchmarks. Encourage debate and seek opinions.
Biases affect how information is delivered in meetings. The curse of knowledge occurs when you assume that the audience knows as much as you do. But this is a wrong direction.
Acronyms and Jargon:
Use clear, simple language. Avoid jargon and complex sentences when talking to non-technical audiences.
The audience loses attention when you present a very complex topic with a long speech, so concise messages are more effective. Explain your terms. Do not assume people know common acronyms. Even terms like “average revenue per user” need explanation sometimes. Tell the audience how you calculated a specific value. If you use variable names from a database, explain what they mean.
Ask questions if you do not know an acronym. Every team has different expectations for communication.
Expectations and Conflicts:
Balance stakeholder expectations with what is possible. Setting realistic, objective targets is important.
Flag problems early. If data issues delay your work, let your stakeholders make necessary changes. Outline the pros and cons of requested changes. Present reports on how new requests affect the project timeline.
Conflict is normal in work life. Mismatched expectations and miscommunications are common causes. Try to be objective and stay focused on team goals. Reframe problems to be more productive. Ask how you can help a team member reach their goal.
Communication is the most important part of your analysis. Focus on the WHY behind your figures.
Accuracy and credibility are essential. Be honest about uncertainties and limitations to build trust with decision makers. Disclosure of data sources and methodology is necessary here. Always connect data to the decisions people must make. Your job is to help the business make better decisions.
Always ask “so what” to ensure your metrics are relevant. If a metric does not lead to an action, then remove it from your report.
💡Subscribe to get better at the part of data analysis nobody teaches you.
What is the hardest part of communicating with non-technical stakeholders?
Let me know in the comments👇
This is a fantastic blueprint for integrity, especially your point that "accuracy and credibility are essential." Many today trust AI as an ultimate source rather than a tool, but that is like trusting a physical structure built on a bad foundation; it is destined to fail.
Your insight that "disclosure of data sources and methodology is necessary" is a critical reminder that when AI is used, the background sources themselves must be transparent and credible. My audits of cross-language logic failures confirm that without this rigor, we lose the very trust you advocate for.
I commend this article for its brilliant focus "on the WHY behind your figures"!
Such a great blueprint. A lot of this is applicable whenever you're trying to influence leaders and decision makers too.
Thanks for putting this together!