Colquhoun.AI
← All guides

Practical website research

Choose the Right Time Scale Before Explaining a Traffic Spike

· Colquhoun.shop editorial

A daily chart looks more detailed than a monthly chart. That does not mean it contains more measured information. Colquhoun.shop lets you switch the imported-analytics chart between monthly, weekly and daily views, but the underlying imported records are monthly totals. The daily and weekly shapes are illustrative allocations, not observed daily visits.

That distinction matters when someone asks whether a campaign, product launch or website edit caused a spike. Here is how to use the chart without giving it a job the data cannot support.

Start with monthly values

Use the monthly view to discuss an imported month’s total. Check the dataset’s source and selected time range before interpreting a change. Demo data is fictional and should stay out of real business conclusions.

The dashboard’s three-, six- and twelve-month selections use the latest available monthly records. If a dataset has fewer records, the view cannot create missing measurements. Confirm the displayed month labels rather than assuming that every named range represents a complete calendar period.

For example, suppose your source supplies 30,000 visits in September and 36,000 in October. The monthly data supports a statement that the supplied October total is 20% higher. It does not, by itself, show which October day changed or which campaign drove the increase. If either month is incomplete or the measurement method changed, qualify even that monthly comparison.

Understand what the daily view adds

For imported analytics, the daily view distributes each monthly total across the days in that month using a deterministic pattern. The allocated values add back to that monthly total, but their individual peaks are not observations from your analytics platform.

This can help you explore how the interface presents a time series. It cannot identify your best weekday, measure a one-day promotion or establish that a website change improved traffic on the following morning. A visually prominent peak is not evidence of an event just because it appears beside a date.

When sharing the chart, keep the illustrative-data warning visible. A screenshot cropped to the line alone can turn a clearly labeled visualization into a misleading claim.

Weekly buckets are not calendar-week measurements

The weekly view groups the illustrative daily values into consecutive seven-day buckets starting at the beginning of the selected period. These are not necessarily Monday-to-Sunday reporting weeks, and the final bucket can contain fewer than seven days.

Do not compare one of those buckets directly with an ISO-week export from another analytics service. First align the date boundaries, time zone and measurement definition. For a real week-by-week business decision, return to your original analytics service and obtain measured daily or weekly records there.

A traffic-channel filter does not add missing history

In the imported-analytics workflow, channel shares are supplied as dataset-level attributes. Selecting a channel applies its share to the monthly totals; it does not reveal a separately measured channel history for each month.

If the dataset says Organic Search accounts for 40% and the month has 30,000 visits, the chart’s allocation is 12,000. That arithmetic is useful to understand the display, but it is not proof that exactly 12,000 organic visits occurred that month. Actual channel mix can change over time. AI-referral shares require the same care.

Label such a view as an allocation based on the supplied share, not a new source of historical attribution.

Match the evidence to the decision

Use imported monthly totals to discuss the monthly figures you supplied. Use the daily and weekly views as illustrative displays. Use your original analytics service when the question requires observed daily events, campaign attribution or changing channel shares.

For a campaign review, record the question before looking at the graph: “Did measured organic visits rise during this campaign?” Then collect the matching measured period and channel from the source. Avoid choosing a story first and using an illustrative line to support it.

Colquhoun.shop helps organize and present the information, but changing a chart interval cannot manufacture a more granular measurement. For the input side of the workflow, read Bring Your Own Analytics Without Losing Their Meaning. Before forwarding a spreadsheet, use the traffic CSV handoff guide to retain the dataset’s context.

This guide describes the built-in Web Reader and Keyword Density behavior verified for this publication. Available data depends on the page response; these tools do not measure visits or search rankings.