Missing Days in a Dashboard Line Chart: A Gap Is Not a Measured Zero
A line chart can make a business trend easy to follow. It can also make missing observations look like a continuous, confidently measured history. Before commissioning a reporting screen, decide what happens when an expected day or month has no usable value. Connecting across it, drawing a zero and leaving a gap communicate different things.
This guide concerns the continuity of an observed time series. It is not about keeping a browser dashboard refreshed or calculating a running total. The question is what the chart may truthfully say about the periods between the observations it actually has.
Define the measurement schedule first
The Office for National Statistics guidance on line charts recommends them for displaying trends over time. It distinguishes irregular observations from missing values in an otherwise regular series. Your report needs the same distinction before a developer can choose the right rendering.
Consider a hypothetical service business recording a daily count of completed jobs. If the measurement covers every calendar day, a date with confirmed no completed jobs may legitimately have a count of zero. A date whose records have not arrived is not the same observation. Nor is a day outside the agreed operating schedule necessarily missing.
Write down the expected calendar, the reporting time zone, the event that assigns a job to a date and the point at which the period becomes complete. These are proposed business requirements, not rules supplied by the ONS page. They establish what the app should look for before it describes a value as absent.
If the source records only occasional measurements, do not silently pretend there was supposed to be one every day. A series of inspection readings taken when an inspection occurs has a different observation schedule from a daily closing count.
Leave an honest break in a regular series
For otherwise regular data with missing observations, ONS says to leave gaps in the line so that continuity is represented accurately. The visible interruption matters. It prevents a smooth connection from suggesting that the missing period has been measured.
My recommendation is to distinguish the following states in both the chart explanation and its accompanying values:
- A usable measured value, including a genuine zero where the metric allows it.
- An expected observation that is missing or unusable.
- A period not included in the measurement schedule.
- A failed report request, where the app cannot yet establish the series.
Agree how each state appears. An unavailable period should not become a zero merely because the plotting component requires a number. A request failure should not be presented as a complete empty history.
Do not ask a developer to fill missing periods automatically just to make the line attractive. If the business genuinely requires estimated values, commission the estimation method and its visible distinction separately. This article does not establish that a particular interpolation method is appropriate, or that an estimate is interchangeable with a recorded count.
Irregular readings need visible observation points
ONS recommends data markers when observations occur at irregular intervals so that readers can understand their frequency. For an operational app, pair that with a clear specification of the time axis.
Ask that positions reflect the actual observation dates or times rather than placing every received record at an equal distance merely because it is the next row. Otherwise a long interval and a short interval can occupy the same horizontal space. This spacing requirement is commissioning guidance; it is not a claim that every chart library already behaves that way.
Markers identify where readings exist. They do not turn the line between readings into a continuous measurement. Make that limit clear when users might rely on an apparent intermediate value.
Also decide what happens when multiple observations have the same timestamp. They may require a defined aggregation or a different view. Arbitrarily dropping all but one is not a reasonable default to hide from the business owner.
Some values stay fixed until a deliberate change
Not every time-based value represents an observation that should be joined by a sloping line. ONS recommends a stepped line for variables that change at distinct points and remain constant between them, such as interest rates or price caps.
A hypothetical app might show a business's configured service threshold over time. If the threshold is explicitly effective until replaced, a diagonal slope between two changes could suggest gradual values that were never configured. A stepped representation may fit that contract better.
Do not use the same reasoning to carry every missing measurement forward. A setting known to remain in effect and a daily measurement that failed to arrive have different meanings. Establish the data contract before selecting the line shape.
Filters and exports must preserve the missingness
A user may compare branches or reporting periods. Decide whether all series use the same measurement schedule and completeness rules. A group with unreceived records must not look better merely because its difficult periods disappear from the visible history.
For the proposed report, retain enough context in an export to identify the period, unit, schedule and unavailable observations. A screenshot detached from its dashboard should not accidentally turn a gap into an apparent absence of business activity.
The readable values accompanying the chart should use the same selection as the plot. If a correction later supplies a missing observation, the updated chart and table should agree. This is a proposed acceptance requirement, not a data-reconciliation feature verified in an existing application.
Test absence deliberately
Put continuity cases in the acceptance criteria for your web app, not only in the designer's notes. Use harmless synthetic records with a known expected schedule.
Ask to see a confirmed zero, a missing expected observation, a date outside the schedule and an unavailable report request. Inspect the line, labels and readable data for each. The missing expected value should produce the agreed break rather than a substitute measurement.
Then test irregularly timed observations, repeated timestamps and an explicitly persistent setting with a deliberate change date. Confirm that the app follows the correct contract for each series instead of applying one generic line option to everything. Include a narrow screen and an export in the proposed acceptance work.
Bring the observation schedule and the business decision to a reporting-app discussion. The useful brief is not simply “draw a trend.” It explains when you have evidence of a value, when you do not, and what a connecting line is allowed to imply.
Source and limits
The ONS line-chart page was read on 26 October 2026. It supports the distinction between irregular and missing data, markers for irregular readings, gaps in otherwise regular series and stepped lines for values that remain constant between changes. Reporting schedules, time-zone rules, missing-state contracts and acceptance cases above are product guidance. No reporting app, chart library or real business dataset was executed or assessed for this article.