Logarithmic Scale

What a logarithmic scale does

A logarithmic scale compresses wide numerical ranges by mapping equal visual spacing to equal multiplicative change. That makes it useful when a quantity spans several orders of magnitude, because the reader can compare both tiny and huge values on the same axis without the largest numbers overwhelming the chart.

The key idea is proportional change. On a log axis, moving from 1 to 10 and from 10 to 100 takes the same space, so the chart emphasises ratios rather than absolute gaps. That is why logarithmic scales are common in latency graphs, signal measurements, sound, astronomy, finance, and other domains where values grow nonlinearly.

Why logarithmic scaling changes interpretation

A log scale changes what “distance” means. Two points that look equally far apart may differ by very different absolute amounts, but by the same factor. This is helpful when a chart would otherwise be dominated by outliers, yet it can also mislead readers who expect a linear axis and read the spacing as simple addition.

Because the scale is multiplicative, a small upward move at the low end can represent a much larger relative change than the same visual move at the high end. In practice, that makes log plots especially good for spotting growth rates, decay, and dispersion across orders of magnitude, while still requiring careful labelling so the axis is not mistaken for a linear one.

Where logarithmic scales are used

Logarithmic scales appear anywhere a quantity ranges from extremely small to extremely large values. In cybersecurity and systems work, that often includes response times, throughput, error rates, packet loss, and signal-related measurements. The value of the scale is not just presentation, but analysis: it can reveal patterns that would be hidden if everything were squeezed into a linear display.

They are also useful for comparing distributions. If one service response is 10 times slower than another, a log axis shows that relationship cleanly, whereas a linear axis may make the faster values look indistinguishable. This is why analysts often use log plots for performance troubleshooting, capacity planning, and anomaly exploration.

Common pitfalls and how to read them correctly

Logarithmic scales are easy to misread when the axis labels are sparse or the viewer assumes constant difference instead of constant ratio. Zero and negative values cannot be shown directly on a standard log axis, so data often needs transformation, offsetting, or a different scale choice altogether.

The most common mistake is treating equal vertical or horizontal spacing as equal absolute change. Another is using a log scale when the audience needs intuitive magnitude comparisons rather than ratio comparisons. A logarithmic scale is most effective when the underlying question is about multiplicative behaviour, not simple addition.

Practitioner Guidance

Why practitioners should care: Use a logarithmic scale when the main analytical goal is to compare relative change across a large range, not to preserve additive distance. It is especially valuable when outliers would otherwise flatten the rest of the data.

What to watch for: Label the axis clearly and make sure readers know whether the chart is base-10, base-2, or another logarithmic form. If the data includes zeros, negatives, or values that must be read as absolute differences, a log scale may be the wrong display choice.

Related resources from NHI Mgmt Group

• What is the first step in managing non-human identities at scale?
• How should security teams govern non-human identities at scale?
• How do organisations operationalise NHI ownership at scale?
• What are the main reasons AI agents struggle to achieve enterprise-scale deployment?