Business visualizations are visuals that communicate insights drawn from data to business stakeholders. This is technique 3.2 of the IIBA Guide to Business Data Analytics, and it is different from technical visualizations, which help analysts understand the data [2]. On the CBDA exam it lives mostly in Interpret and Report Results, a domain worth 20% of 75 questions [1]. The exam tests one habit: choose the visual from the insight and the audience, not from what looks good.
What business visualizations are, by the guide
Business visualizations are fit-for-purpose visuals built to communicate key insights, while technical visuals have well-defined forms [2]. The guide places them in most analytics initiatives: dashboards and reports in descriptive analytics, and also visuals for exploratory data analysis (EDA) results, conclusions or the analytics process itself [2].
| Item | Business visualizations (3.2) | Technical visualizations (3.18) |
|---|---|---|
| Audience | Business stakeholders [2] | Analytics professionals [2] |
| Goal | Communicate insights | Draw and understand insights from data |
| Form | Fit for purpose, often interactive | Well-defined forms (for example ROC curve, residual plot) |
| Main domain | Interpret and Report Results [3] | Analyze Data |
The Analyze Data techniques table (2.3.6) says it in one line: technical visualizations are used to analyze the data, while business visualizations are used to interpret and report results. My take: if you remember only this line, you will already get the "which visual for whom" questions right.
Why the technique matters on the CBDA exam
The technique is listed in the techniques table of Interpret and Report Results (2.4.6), and the same domain discusses visuals in 2.4.4 and 2.4.5 and has a Visualization Best Practices box [3]. That domain carries 20% of the exam, the joint highest weight with Domain 1 and Domain 5 [1].
Last verified: November 2026 (exam facts on the IIBA CBDA page).
Where the technique shows up across the guide:
| Domain and task | How visuals appear |
|---|---|
| 2.2.2 Determine the Data Sets | Analysts use visualization skills for conceptual diagrams of sources, flows and feed frequency |
| 2.3.6 Select Techniques for Analyze Data | Contrast: technical visuals analyze, business visuals report |
| 2.4.4 Derive Insights from Data | A marketing stakeholder will likely be more interested in ad spend versus revenue than in an error residue graph |
| 2.4 Visualization Best Practices box | Text for one or two metrics, 2D can be more effective than 3D, a secondary axis generally not a good idea |
| 2.4.5 Document and Communicate Findings | Visualization and data storytelling work together; static charts can become dynamic models |
| 2.4.6 Select Techniques | Business Visualizations listed with Data Storytelling tools such as the 3-minute story and the Big Idea |
| 2.5.5 Case study | A word cloud to name product categories, spider charts to differentiate customer segments |
So the exam can touch business visuals in at least three domains. Most questions, I assume, will sit in Domain 4.
How the technique works: purpose, data type, message
The guide builds the technique from five guidelines and three elements [2]. I find it easier to read them as one decision path.
Five guidelines [2]:
- Know the research problem and the insight. The insight decides the visual, not the other way round.
- Know your stakeholders. Interests, motivation, technical acumen and availability set the level of detail.
- Present unbiased analysis. The guide's example: the same data on two y-axis scales gives two different impressions. A visual should help stakeholders "see" the insight, not push an opinion [2].
- Avoid complexity, function over form. Simpler graphics need less context. Focus on the message, not the aesthetics.
- Stand out from the rest. When two visuals are equal on intent and detail, pick the one that stands out, because similar visuals tire stakeholders.
Element 1. Purpose and type of visual. Two questions place the visual in one of four quadrants: is it declarative or exploratory, and is the information conceptual or data-driven [2]? The four purposes:
| Conceptual | Data-driven | |
|---|---|---|
| Declarative | Idea illustration (process models, infographics, frameworks) | Everyday dataviz (bar and pie charts, one message each) |
| Exploratory | Idea generation (mind maps, mental models) | Visual discovery (finding new insights, confirming hypotheses) |
The same 2x2 appears in a 2016 Harvard Business Review article, which asks "Is the information conceptual or data-driven?" and "Am I declaring something or exploring something?" [4].
Element 2. Type of data. The guide maps four data intents to charts [2]:
| Intent | Charts the guide names |
|---|---|
| Comparison | Bar charts, line charts, butterfly charts |
| Composition | Stacked bar charts, waterfalls, area charts |
| Distribution | Line charts, histograms, scatter plots |
| Relationships | Scatter plots, heat maps, chords, Venn diagrams |
Element 3. Key message and supplementary texts. The visual is not always enough. The context, data element, key statistics and legends are explained, and key messages are embedded without crowding the visual [2].
A worked example: choosing one regional sales chart
This is a made-up example, built on the guide's own regional sales scenario in 3.2.2 [2]. The steps follow the guidelines and elements above.
The data (hypothetical, in $M):
| Region | Last year | This year | Growth |
|---|---|---|---|
| North | 4.0 | 4.6 | +15% |
| South | 5.8 | 6.1 | +5% |
| East | 6.2 | 6.4 | +3% |
| West | 5.5 | 5.9 | +7% |
Step 1. The insight. North grew fastest year over year, but it is still the weakest region this year. That is two comparisons at once, the same shape as the guide's example.
Step 2. The stakeholders. A sales VP who decides on regional budgets. She has two minutes and no interest in the model behind the numbers.
Step 3. The quadrant. Declarative and data-driven, so everyday dataviz: one message, a few variables.
Step 4. The data intent. Comparison by items and over time. A grouped bar chart fits: four regions on the axis, two bars each.
Step 5. Bias check. Bars start at zero. If the axis started at 4.0, North's gap would look much bigger than it is.
Step 6. Complexity check. No 3D, no secondary axis for the growth percentage, no pie of this year's shares. Growth goes into a short label above each pair of bars.
Step 7. Key message. A headline above the chart: "North grows fastest but is still the smallest region." One line of text under it with the source and period.
A pie of this year's sales would look nicer. It would also hide the first half of the insight, because a pie has no "last year". This is exactly the guide's point: a visual that is unsuited to the insight fails, however good it looks [2].
The guide also has a real-style case in Domain 5 (2.5.5). A UK online gift retailer, GiftRonline, got a clustering result: five product groups and eleven customer clusters, labelled "cluster no.0", "cluster no.1" and so on. Executives liked the collaboration but were unsure how to use the result. The business analysis professional on the team worked out how to communicate it, and the team created key visuals. A word cloud of keywords from product descriptions helped recommend easy names for the product categories, for example "Jewelry & Luxury". Spider charts then described each customer segment, with product categories as axes and attributes such as the number of purchase visits and average price paid per order. Together with additional analysis and a later CRM feasibility study, the visuals turned "cluster no.2" into something marketing could act on.
Strengths and limitations
The guide lists four strengths and four limitations [2]:
| Strengths | Limitations |
|---|---|
| Powerful when combined with storytelling | Can lead stakeholders to an incorrect conclusion through biased representation |
| More effective for discussions and for understanding insights | Require processing a large amount of data and are prone to data errors |
| Versatile across contexts of interpreting and reporting results | Mostly summaries, so minor but relevant trends can be missed |
| Most stakeholders are familiar with visuals, so they can have a big impact with decision-makers | Need strong storytelling and data orchestration skills, which limits standalone visuals |
Two of the four limitations are about trust. I think that is why the exam likes "biased representation" as a wrong-answer theme.
Experts disagree on chart rules
Practitioners agree that the chart should serve the message, but they argue about almost every specific rule. Five positions, with the guide's view next to each:
1. Pie charts should go. A data visualization consultant argued in 2007 that our eyes compare length and position well but angles and areas badly, and closed with "save the pies for dessert" [5]. The guide is close to this: the Visualization Best Practices box in section 2.4 suggests replacing a pie that needs arc lengths and angles with a simple horizontal bar chart [3]. Yet it still lists pie charts as everyday dataviz [2].
2. Pie charts are misunderstood, not useless. Visualization researchers summarize their earlier study as finding "no clear evidence" that people read pies by the central angle; area and arc length appear to be at least as important [6]. A 2019 follow-up concluded area is the most likely cue [7]. A storytelling coach's team grants that "there are still a few cases where a pie works", while other charts usually win [8].
3. Always start at zero, or not. Visualization researchers note that truncated bar charts exaggerate effect sizes and that advice is "equivocal" for other chart types; in their own experiments the exaggeration held across designs [9]. A charting tool team blocks truncated bar charts, while "line charts can but don't need to start at zero" [10]. The guide's y-axis example sits on the safe side: the scale changes the opinion [2].
4. Never use two axes, or sometimes. The guide calls a secondary y-axis "generally not a good idea" [3]. A chart designer made the same case in 2018 and proposed indexed charts instead [11]. In 2026 the same designer wrote that she changed her mind, and her company launched dual-axis charts [12]. So even one expert can stand on both sides.
5. Dashboards for executives. The guide expects dashboards and reports in descriptive analytics [2]. A data practitioner writes that "senior management doesn't like to look at dashboards" and prefers pushed, targeted reports [13]. An analytics blogger quotes a peer: "Good dashboards are opinionated" [14].
On 3D the experts agree with the guide. A professor's data visualization book says our brain only partly corrects for 3D distortion and calls gratuitous 3D "unequivocally bad" [15]. The guide says it more softly: 2D graphs with appropriate colours and labelling can be more effective than 3D [3].
Common misconceptions
- "Business visuals are just simpler technical visuals." No. The audience and purpose differ. A residue graph helps choose the number of predictors; a marketing stakeholder will likely be more interested in ad spend versus revenue (2.4.4).
- "A good visual speaks for itself." The guide adds key messages and supplementary texts on purpose (3.2.3).
- "More interactivity is always better." The guide says an interactive or a static graph may be more suitable depending on the context (2.4, Visualization Best Practices).
- "A graph is always better than numbers." With one or two metrics, simple text or a small table can be more effective (2.4, Visualization Best Practices).
- "Standing out means decorating." Standing out applies only when two visuals are otherwise equal on intent and detail (3.2.2).
Business visualizations vs similar CBDA techniques
Four techniques sit near business visualizations in the guide. The difference is mostly in what they deliver.
| Technique | What it delivers | When the exam wants it |
|---|---|---|
| Business Visualizations (3.2) | Charts, graphs, infographics for stakeholders | Choose a visual for an insight and audience |
| Technical Visualizations (3.18) | Analyst-facing plots such as ROC curves or residual plots | Analyze model fit or data patterns |
| Data Storytelling (3.7) | Data, visuals and narrative together | Drive a decision with a story |
| The Big Idea (3.19) | One statement with the central insight and call to action | Summarize the "so what" |
| 3-Minute Story (3.20) | A short narrative of the insight | A busy stakeholder with three minutes |
The guide notes that business visualization is powerful "when combined with storytelling" [2]. So in a scenario, the visual is rarely the whole answer.
Check yourself: CBDA practice questions
Four short questions, checked against sections 3.2 and 2.4 of the guide.
1. A data scientist wants to show the head of marketing a plot of prediction error versus the number of predictors. What should the analyst suggest?
A. Present it as is; it is accurate B. Replace it with a business visual of ad spend versus revenue C. Convert it to 3D for clarity D. Add a second y-axis with revenue
Answer: B. The error plot is a technical visualization; a marketing stakeholder needs a business visual (2.4.4). C and D go against the Visualization Best Practices box in section 2.4.
2. A visual will show how total sales split across five product lines this quarter. Which data intent is this?
A. Comparison B. Composition C. Distribution D. Relationships
Answer: B. A whole made of different items is composition, with stacked bars, waterfalls or area charts (3.2.3).
3. A sponsor only needs to know overall budget spent versus budget allocated, with no daily tracking. What does the guide suggest?
A. A daily line chart B. A heat map by day and team C. An aggregate percentage visual D. A scatter plot of spend per invoice
Answer: C. The guide's own stakeholder example favours an aggregate percentage visual (3.2.2).
4. Which is a limitation of business visualizations listed in the guide?
A. Stakeholders are unfamiliar with visuals B. They cannot be used with storytelling C. Minor but relevant trends may be overlooked D. They only work for predictive analytics
Answer: C. Visuals are mostly summaries, so small trends can be missed (3.2.4).
Conclusion: the gap between the guide and real dashboards
The guide's principles hold up well in practice; the gaps are in how strictly practitioners apply single rules.
| Area | Guide | Reality | Delta |
|---|---|---|---|
| Pie charts | Listed as everyday dataviz; replace when angles must be read | Debated both ways [5, 6, 7, 8]; the guide already sits in the middle | No gap |
| Y-axis scale | Scale can bias the audience | Strict for bars; tools let line charts start above zero [10]; research finds the effect across designs [9] | Added by practitioners |
| Secondary axis | Generally not a good idea | Debated, now offered by tools [11, 12] | Reduced to a default, not a ban |
| 3D charts | Prefer 2D | Strong agreement against gratuitous 3D [15] | No gap |
| Dashboards | Typical output of descriptive analytics | Executives often prefer pushed, opinionated reports [13, 14] | Reduced to pushed reports for executives |
So for the exam, take the guide's rules as written. At work, I would read them as strong defaults with known exceptions.
Related topics
- CBDA data visualization questions: a set of chart-choice drills.
- CBDA Interpret and Report Results quiz: the domain where this technique matters most.
- Business simulation, technique 3.1: the previous technique in this series.
- CBDA exam guide and the 6-week CBDA study plan.
- Practice timed on the CBDA exam simulator.
FAQ
What are business visualizations in the CBDA guide?
They are technique 3.2 of the Guide to Business Data Analytics: visuals used to communicate insights from data to business stakeholders. They differ from technical visualizations (3.18), which help analysts understand data.
Which CBDA domain uses business visualizations most?
Interpret and Report Results (Domain 4, 20% of the exam). It lists the technique in 2.4.6 and discusses visuals in 2.4.4 and 2.4.5, with a Visualization Best Practices box. Visuals also appear in a Domain 5 case study.
What are the four types of data intent in business visualizations?
Comparison, composition, distribution and relationships. For example, bar and line charts for comparison, stacked bars and waterfalls for composition, histograms for distribution, and scatter plots or heat maps for relationships.
What is the difference between business and technical visualizations?
Business visualizations communicate insights to stakeholders and are fit for purpose. Technical visualizations, such as ROC curves or residual plots, have well-defined forms and help analysts analyze data and models.
Does the CBDA guide ban pie charts and dual axes?
No ban. Pie charts are listed as everyday dataviz, but the guide suggests a horizontal bar chart when the audience must read arcs and angles. A secondary y-axis is described as generally not a good idea.
Sources
- IIBA: Certification in Business Data Analytics (CBDA)
- IIBA: Guide to Business Data Analytics, 3.2 Business Visualizations
- IIBA: Guide to Business Data Analytics, 2.4 Interpret and Report Results
- Harvard Business Review: Visualizations That Really Work (2016)
- Perceptual Edge: Save the Pies for Dessert (2007)
- eagereyes: Judgment Error in Pie Chart Variations (EuroVis 2016)
- eagereyes: Evidence for Area as the Primary Visual Cue in Pie Charts (IEEE VIS 2019)
- storytelling with data: forty-five pie charts? never say never (2019)
- arXiv: Truncating the Y-Axis: Threat or Menace? (CHI 2020)
- Datawrapper Academy: Why our column and bar charts start at zero (2021)
- Datawrapper: Why not to use two axes, and what to use instead (2018)
- Datawrapper: Why I changed my mind on dual-axis charts (2026)
- The Art of Data Science: Dumbest thing in data science (2024)
- Conor Dewey: Making Dashboards Suck Less (2022)
- Fundamentals of Data Visualization: Don't go 3D