When Should a Dashboard Show a Range Instead of One Number?
From the earlier reference collection. Read the current software blog.
A range can show meaningful uncertainty hidden by estimates, sensor variation, and aggregation—provided the dashboard explains what the endpoints mean.
Use a range when the spread matters
A single value is easy to scan, but it may conceal differences that change how a person interprets the data. If a model estimates demand between 80 and 110 units, displaying “95” can suggest a certainty the estimate does not have. If several sensors report different readings, a range may reveal variation that a mean suppresses.
Ranges are useful when the underlying uncertainty affects a decision: whether conditions are within a safe operating band, whether a forecast overlaps a threshold, or whether locations differ substantially. If the variation is small relative to the decision, a single rounded value may be sufficient, with detail available on request.
Name what the range represents
A range is not self-explanatory. It could be the minimum and maximum observed values, a confidence interval, a forecast interval, or a set of plausible scenarios. Label it accordingly. “Observed across reporting stations: 11–16°C” differs from “model estimate: 11–16°C.” Avoid language such as “likely range” unless the method supports that interpretation.
For example, suppose four sensors report 11.2, 11.7, 14.8, and 15.1°C. The observed range, 11.2–15.1°C, communicates that the sites differ. Their average alone would hide the split. A location map or per-site table can help explain whether the spread reflects geography, sensor conditions, or a data problem.
Keep endpoints honest
The displayed range should match the calculation and data coverage. If one sensor is missing, say how many contributed. If the endpoints are rounded, choose a precision that does not imply unsupported accuracy. Do not present a minimum and maximum as if they capture all uncertainty: an unobserved error source may still exist.
When comparing ranges, keep units, time windows, and definitions consistent. A daily minimum-to-maximum range cannot be fairly compared with a weekly interval without clear labeling. If a range crosses a meaningful threshold, call attention to that fact rather than emphasizing only the midpoint.
Make the number and its limits work together
Sometimes readers still need a concise summary. A dashboard can show a rounded central estimate alongside a clearly labeled range, for instance “about 95 units; estimate 80–110.” The central figure helps scanning, while the interval preserves the uncertainty relevant to decisions.
A range is useful when it clarifies what a point value would hide. Explain its source, coverage, and meaning, and readers can judge how much confidence to place in both endpoints.