Level 3 · Daily spend around Black Friday as a line chart

A line chart is the chart for time. The x-axis is a sequence of dates in order, and the line connecting the points lets the eye read rises, drops and peaks at once. Use it when the question starts with “how did it change”: daily sales, weekly signups, monthly spend.

Long data, wide chart

The export in A1:C61 is long: one row per day and channel, 20 days × 3 channels = 60 rows. A line chart with one line per channel needs the data wide: one row per day, one column per channel. The days are given in E2:E21 and the channels in F1:H1; the grid in between is yours.

One SUMIFS for the whole grid

Each cell needs the amount where the date matches its row and the channel matches its column:

F2   =SUMIFS($C$2:$C$61,$A$2:$A$61,$E2,$B$2:$B$61,F$1)

The references do the work when you fill:

Reference	Locked	Why
$C$2:$C$61, $A$2:$A$61, $B$2:$B$61	both	the data never moves
$E2	the column	every column reads the day in E
F$1	the row	every row reads the channel in row 1

Check one cell by hand. F15 should be the paid spend on Friday, November 28: 3,081.46, the value in the export row for that date and paid.

Making it in Excel

Select E1:H21, then Insert → Charts → Line. Excel puts the dates on the x-axis and makes one line per column, named after the headers. If Excel draws the dates as a fourth line, the date cells were stored as text: fix them first (chapter 4). In Chart Design → Add Chart Element, add an axis title for the value axis (“Spend, $”) and keep the legend, because with three lines it is what tells them apart.

What the chart shows

Paid spend sits between 974.81 and 1,336.44 a day from November 15 to 20, then jumps on Friday, November 21 to 2,234.09 and stays high: 3,064.30 on Saturday the 22nd, 3,081.46 on Black Friday (November 28) and a peak of 3,180.40 on Cyber Monday (December 1). The next day it falls back to 1,077.28. Social follows the same shape at about a third of the level, peaking at 1,185.48 on November 27. The campaign ran for eleven days, not for one weekend.

Email is the hard line to read. It never goes above 381.99, so on an axis built for 3,000 it hugs the floor and its own pattern disappears: it drops on weekends (52.20 on Sunday, November 16) and rises on weekdays. When one series is ten times smaller than the others, give it a secondary axis (a combo chart) or its own chart. The page draws every line on one axis, so here you see the problem rather than the fix; in Excel, right-click the email line → Format Data Series → Secondary Axis.

Common mistakes

• Relative references in a grid. =SUMIFS(C2:C61,A2:A61,E2,B2:B61,F1) breaks the moment you fill it right or down.
• A column or pie chart for a time series. Twenty days as columns works, but the trend reads faster as a line; a pie of days makes no sense at all.
• Missing days. A line chart joins the points it has. If a day is missing from the table, the line skips over it without a sign. List every day, even when the value is zero.