What does up and to the right mean?

It is deck shorthand for a chart whose line rises left to right, taken as evidence that a business is growing. It describes the shape of the picture, not a property of the business, which is why it is worth asking what produced the shape.

Why does a cumulative chart always go up?

Because each point adds the period to everything before it, and you cannot add a negative number of customers signed up. A cumulative line falls only if the underlying measure can go negative. On counts and revenue it cannot, so the rising line carries no information about whether this month beat last month.

How can I tell if a rising chart is real growth?

Ask four things: does the axis start at zero, is the window the full history or a chosen slice, is the measure cumulative or per period, and is the series adjusted for anything that grew on its own such as prices or the size of the market.

What should I show instead of a cumulative chart?

Show the per period figure as bars and, if the trend is noisy, a moving average line over the top. That chart can fall, which is the whole point: it can tell you something a cumulative line cannot.

← All Kits · Excel Kit

The Up and to the Right Chart: What It Promises and What It Hides

Michael Nocito · Updated August 2026 · Every number on this page was worked before it was published

"We need it to go up and to the right." Everybody who has sat in a review has heard it. It is a request about the shape of a picture, and the awkward part is that the shape can be arranged four different ways without touching the business at all.

What you do: before you accept or draw a rising line, ask what makes it rise. There are only four common answers, and three of them are properties of the chart rather than of the company.

The short version. A cumulative chart cannot go down. If your growth story rests on one, it rests on arithmetic.

The four ways to make a line climb

Cause	What is happening	Tell
The business grew	Each period is bigger than the one before	The per-period chart also rises
The measure is cumulative	Every point adds the period to the running total	The line never falls, not once, in years of data
The axis is truncated	A small real change fills the plot area	The y axis starts at an odd number like 1,600
The window was chosen	The chart starts at a low point	The series has more history than the chart shows

Cumulative is the one to check first

Take twelve months of signups: 420, 390, 405, 370, 355, 380, 340, 360, 330, 345, 310, 325. That series is drifting down. The last three months average 327 against the first three at 405, a fall of about 19 percent.

Now plot total signups to date. It runs 420, 810, 1,215, 1,585, 1,940, 2,320, 2,660, 3,020, 3,350, 3,695, 4,005, 4,330. A clean rising line, every month higher than the last, ending at 4,330. Both charts are drawn from the same twelve numbers. One of them is incapable of falling, because signups cannot be negative.

The test for a cumulative chart is one question. Could this line ever go down? If the answer is no, the line is not evidence of anything except that time passed and the number was not negative.

The chosen window

Pick the start date after seeing the data and any series will rise. The honest version shows the whole history, or states the window in the subtitle with the reason it was chosen: since the product launched, since the pricing change, since the last full year. A window with a reason is analysis. A window without one is a decision you made and did not report.

The axis

Two bars, 1,750 and 3,150. On a zero-based axis, one is 1.8 times the other and it looks like it. Start the axis at 1,600 and the visible bars become 150 and 1,550, a ratio of over ten to one. Nothing in the data moved. The full worked version of that trick is in how charts mislead, and the rule for when a non-zero axis is legitimate is in should the axis start at zero.

Growth that was not yours

Revenue in cash terms rises while prices rise. Users rise while the market grows. Neither is fraud and both need a second line: the same series in real terms, or as a share of the market. If the share is flat, the honest sentence is that you grew with the market, and somebody in the room already knows that.

What to draw instead

Bars per period, not a cumulative line. A chart that can fall is a chart that can tell you something.
A moving average over the bars when the periods are noisy. Three or four periods, chosen from the cycle rather than from how smooth it looks. See moving averages.
The full history, with any chosen window marked on it rather than cut out of it.
A second panel for the reference: market size, price index, headcount. Whatever the growth should be judged against.

Cumulative charts are not banned. They are the right answer to a genuinely cumulative question: how far through the annual target are we, how many licences have we issued in total, how much of the backlog is cleared. The failure is using one to answer "are we growing".

Reading someone else's rising line

Four questions, in this order, because each is cheaper than the next:

Ask	Look at	Bad answer
Is this cumulative?	The axis title, the last value	"Total to date"
Where does the y axis start?	The bottom tick	Anything not zero, unlabelled
Why does the chart start there?	The first x value	"That is when the data starts", when it is not
Compared to what?	Any second series at all	There is no second series

How to apply this to your own work

Open your most-used deck and find every rising line. Mark which ones are cumulative.
Redraw one cumulative chart as per-period bars and look at what it says. If the two disagree, that disagreement is your next finding.
Check the first x value of each chart against the first date in the source. Note any gap and either close it or explain it in the subtitle.
Add one reference series to your headline chart. Last year, plan, or market. A single line is an assertion, two lines are a comparison.
Write the finding in the title so nobody has to infer it from the slope. That method is in chart titles that say the finding.

The one habit to keep

When someone asks for up and to the right, ask which measure they want to see going up. That turns a request about the picture into a decision about the metric, and the metric is a thing you can be right or wrong about.

Is the headline chart in your deck capable of showing a bad month?

Every number here was worked before it was published. The twelve monthly signups sum to 4,330, and the cumulative series is that running sum. First three average 405, last three average 326.7, a fall of 19.3 percent.

A line that only goes up is usually a line that cannot go down.

Charts and Visualization is the chart-choosing book: what each shape can carry, what it quietly distorts, and how to label it so the reader reaches your finding without being told.

Charts and Visualization, $19 →

Learn to read the shape before you draw one.

How charts mislead works eight distortions on one table, and choosing the right chart picks the shape in one question. The Excel Kit has the build steps.

Read How Charts Mislead →