FP&Ai Podcast Ep. 17: Mastering Data Visualisation for Real Insights
In this episode of the FP&Ai Podcast, Roger Knocker shares 10 practical techniques for turning data into meaningful business insights, using live Power BI dashboard demonstrations.
Roger explains how an encounter with a CFO, while working on a business improvement project, changed how he thought about explaining data analysis. Although he was already an experienced data analyst, what struck him was how clearly the CFO walked through his approach to the management accounts.
That experience eventually led to his 10 Tips to Get Insights from Any Set of Data, a practical framework for understanding performance, identifying patterns and uncovering opportunities or problems that might otherwise be missed.
The first principle is to understand who the analysis is for and what they're trying to achieve. From there, Roger demonstrates how to examine targets, growth, trends, rankings, mix, comparisons and statistical measures, using different visualisations to make the findings easier to understand.
The demonstrations use Power BI, but the techniques can also be applied in Excel, Tableau, Qlik and other reporting tools.
Frequently Asked Questions
1. What is the 80/20 rule in data analysis?
The 80/20 rule, or Pareto principle, helps identify whether a large share of results comes from a relatively small number of items.
In one dataset Roger uses, 80% of sales came from just seven out of 254 countries, approximately 3%. Finding those few items shows you where to focus. The remaining "long tail" may be worth investigating to see whether a different operating model makes sense.
Podcast timestamp: 34:50
2. Should management reports be sorted alphabetically?
Usually not. Executives want to see the most important items first, so sort by value or impact rather than by name.
Sometimes the biggest loss maker is the most important line, and that needs to be at the top rather than buried somewhere in the report.
Podcast timestamp: 22:30
3. Why should you look at volume growth as well as sales growth?
Because a strong sales number can hide a different story.
In one example, sales grew by 85% while volume grew by 188%. That means the business had to sell considerably more units to achieve its revenue growth.
Putting a complementary metric alongside your headline KPI helps you identify that relationship before you've even looked at a chart.
Podcast timestamp: 16:20
4. What should be at the top of a management dashboard?
Put the key numbers, such as actual, budget, variance and variance percentage, at the top.
Executives check these figures against what they expect before they start interpreting the charts. If the numbers don't reconcile, they're unlikely to trust the rest of the dashboard.
Podcast timestamp: 08:40
5. What can you use as a benchmark if there's no target?
You can use a statistical measure such as the mean, median or a selected percentile.
If you don't have a formal target, these measures can provide a useful reference point for comparing performance. As Roger puts it, "There's no excuse not to have a target."
Roger often uses reference lines in charts to show where the median or average sits and identify which items are performing above or below that level.
Podcast timestamps: 38:30
6. Why should you analyse trends over different time periods?
Because long-term growth can hide short-term decline.
Volume might be up over two years but falling quickly over the last six months.
Looking at several time horizons, such as monthly, quarterly and rolling six-month periods, helps you understand what's happening more recently rather than relying only on the overall trend.
Podcast timestamp: 19:40
7. What is a waterfall chart used for?
A waterfall chart shows how you got from one number to another, such as last year's sales to this year's sales, or budget to actual.
It highlights which categories contributed to the increase and which reduced the result.
As Roger explains, the big consulting firms regularly use waterfall charts to explain performance.
In the webinar, Roger demonstrates how a waterfall chart can reveal the movements behind a change in sales.
Podcast timestamp: 18:10
8. What is sales mix and why does it matter?
Sales mix is the composition of your sales, such as the percentage contributed by each product, region or category.
If you sold 60% hamburgers and 40% hot dogs last year, but 90% hot dogs and 10% hamburgers this year, the composition of your sales has changed significantly.
Depending on the profitability of those products, that change in mix could help explain movements in overall profit.
Podcast timestamp: 27:10
9. Should you start your analysis with summary or detailed data?
Start with the summary and drill down to the detail.
The human mind works from summary to detail. If your first view is a long transactional table, it's harder to identify what matters.
Summarise the information first, then allow users to drill down into the areas they want to investigate.
Podcast timestamp: 23:50
10. How can you show both the size and the quality of a result in one chart?
Use a chart where size represents quantity and colour represents quality, such as a treemap or bubble map.
For example, the size of each square can represent the monetary value of profit, while its colour represents the profit percentage.
This makes it easier to identify markets that generate significant profit in absolute terms but may be experiencing pressure on their profit margins.
Podcast timestamp: 30:20
Key Takeaways
- Start with the business objective. Understand who the analysis is for and what decision it needs to support.
- Make the important numbers visible. Put key KPIs at the top of a dashboard and compare actual performance against targets.
- Look beyond headline growth. Supporting measures such as volume can reveal a different performance story.
- Analyse multiple time periods. Long-term growth can hide short-term declines.
- Rank data by importance. Present the biggest contributors or problems first rather than automatically sorting alphabetically.
- Work from summary to detail. Give decision-makers the overall picture before allowing them to investigate individual items.
- Understand changes in mix. Shifts in products, regions or categories can help explain changes in profitability.
- Apply the 80/20 principle. Identify which items contribute most to the result and where management attention may have the greatest impact.
- Choose visualisations that support the analysis. Waterfall charts, treemaps, reference lines and other visualisations help communicate different types of insights.
- Use basic statistics. Measures such as the mean, median and percentiles provide useful benchmarks and help reveal patterns in data.
The key message: Effective data visualisation isn't simply about creating charts. It's about knowing what to look for, asking the right questions and presenting information that helps people understand performance and make better decisions.
Presented By
Roger Knocker
Roger Knocker is the CEO of the FP&Ai. With a background in accounting and 12 years in corporate, including as head of data analytics and strategy, he brings extensive experience in business improvement, performance management and data analytics.
Through KPI's training programmes, Roger has helped develop more than 100 data analysts using the same techniques he demonstrates in this episode.
About FP&Ai
FP&Ai helps finance teams transform planning, reporting, forecasting, consolidation, analytics, automation and decision-making through modern finance technology, AI and best-practice processes.
