Practical framework / 5 MIN READ
Show related trends without manufacturing visual agreement
Use aligned panels or a disclosed index instead of tuning two axes to tell a story.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-06
Two lines with different units can be made to overlap or diverge by changing their separate scales. A clear comparison preserves units and avoids suggesting a relationship that comes from chart settings rather than evidence.
What you will leave with: Redesign a two-measure trend chart with explicit units, comparable timing and a bounded interpretation.
Use this when
- A spend-and-sales chart is used to claim advertising caused growth.
- A chart designer adjusts axis limits until two lines align.
This lesson concerns visual comparison, not estimating correlation, attribution or causal effects.
State what the pairing should reveal
Write why the two measures belong together: whether spending and orders moved at similar times, whether service demand exceeded capacity, or whether a cost increase preceded a margin change. Identify units and expected timing before plotting.
Check that the periods and evidence maturity align. A daily spend series and weekly completed-sales series need careful aggregation, not merely a shared horizontal axis. Record relevant contextual changes such as pricing or availability so the chart does not imply that one plotted variable is the only explanation for the other.
Remove the arbitrary scale relationship
The ONS advises avoiding dual-axis charts because they can mislead, suggesting separate charts or annotations instead. In a marketing report, two aligned panels preserve each measure’s unit while making common dates easy to compare.
Keep labels close to their series and use a consistent time axis. Do not adjust separate vertical limits until peaks appear synchronized. The visual crossing of dollars and order counts has no natural commercial meaning. If a stakeholder asks what the crossing means, the design has probably created a question the data cannot answer.
Use an index only for a stated relative question
A common-base index can show relative movement when absolute magnitudes are secondary. Choose and disclose the base period, calculate each series consistently and retain the actual values nearby. An index of 120 means 20% above that series’ own base, not equality between the two underlying measures.
Test whether the base is unusually low, incomplete or zero. A convenient launch date can exaggerate subsequent growth. Consider showing several defensible baselines or using absolute panels if the story changes materially with the chosen starting point. Do not use indexing to hide the scale of spend.
Separate the observation from the claim
Describe what the chart establishes: the timing and size of observed changes under the chosen display. A shared upward trend can reflect seasonality, growth or other common factors. It does not by itself establish that one measure caused the other.
If the decision requires a causal answer, identify the additional design or evidence needed. Keep the chart as descriptive context. Ask a reader to explain both units and the conclusion they draw; revise the title if it invites an unsupported causal statement even when the axes themselves are technically correct.
Make the reasoning concrete.
A hypothetical report plots spend rising from $1,000 to $1,200 and orders rising from 100 to 105. Separate axes can make the two lines look almost identical. An index with the same initial base shows 120 for spend and 105 for orders.
Tune the axes
Opening: Align both series’ starting and ending heights.
Hypothesis: The graphic looks visually coordinated.
Risk: It conceals the different proportional changes.
Use aligned panels or a disclosed index
Opening: Show the actual amounts and their separate changes.
Hypothesis: The reader sees spend increased 20% while orders increased 5%.
Risk: This descriptive contrast still does not estimate incremental return.
Present the measured movements without claiming either caused the other. Use the appropriate causal evidence before making a budget-effectiveness conclusion.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Purpose
What comparison should the two measures support?
Units
Are dates, maturity and measurement units explicit?
Design
Would aligned panels answer the question?
Index
If used, is the base valid and disclosed?
Claim
What conclusion is supported beyond visual similarity?
Watch for the wrong conclusion.
- Tuning scales to make lines overlap.
- Giving cross-unit line intersections a business meaning.
- Hiding an unusual index baseline.
- Treating parallel trends as causal evidence.
Your next action
Replace one dual-axis performance chart with aligned panels and rewrite its takeaway as a supported descriptive observation.
Sources & scope
Frameworks, worksheets and hypothetical examples are original editorial synthesis. Linked primary sources support only the stated concepts. Source checking is not an independent audit of effectiveness. No named human author or reviewer is claimed.
- ONS: Axes and gridlines, dual axis charts →Primary statistical-agency visualisation guidance · Supports: Advises avoiding misleading dual axes and using separate charts or annotations. The campaign examples and alternative-selection worksheet are original illustrations.