The higher price lost. Did it make more money?
A price test where conversion fell 15.5% and the test says the fall is real. Revenue per visitor went up, by an amount this tool cannot test at all, and knowing which of those two facts it measured is the whole template.
Operations Starter Statistics free
After you install, this is the model to open.
Two Prices, One Answer
- In your spreadsheet, click the Sortia icon in the strip of icons down the right-hand edge. No strip? Click the arrow at the bottom-right to open it. You can also use Extensions, then Sortia, then Open Sortia.
- Click Start from a template and put that name in the search box.
- Pick the card with that name and click Load this template. It arrives on a new tab with real numbers already in it.
The answer
- The conversion fall is real
- p = 0.021 9.07% to 7.66% at the higher price, z = -2.31
- Revenue per visitor
- +1.8% $3.60 against $3.54, a change this test never measured
- Margin per visitor
- +46% $1.46 against $1.00 once the $19 unit margin is in
- The winner
- Price B it lost the test it was given and made more money
Two prices on the same product for two weeks: $39 converted 372 of 4,100 visitors and $47 converted 318 of 4,150. Click Run: the rates are 9.073% and 7.663%, an absolute fall of 1.411 percentage points and a relative fall of 15.5%, with z of -2.31 and a two-tail p-value of 0.0207. The higher price genuinely cost you conversion, and the 95% interval says the true cost is between 0.22 and 2.61 percentage points.
Now read the sheet, because the test has answered a question that is not the one you asked. Revenue per visitor went the other way, $3.5385 at $39 against $3.6014 at $47, a gain of 1.78%. The version that lost the test made more money. And once margin is in it, at $11 a unit against $19, the picture is not close: 99.8 cents of margin per visitor at $39 against $1.456 at $47, a gain of 46%.
The tool did not test any of that, and it cannot. A two-proportion test compares the share of visitors who converted, so it can tell you whether a rate moved and it has nothing to say about how much each conversion was worth. If the money question is the one that matters, and it usually is, the test to run is on revenue per visitor itself: record one row per visitor with the revenue they generated, most of them zero, and run t-Test: Unequal Variances on the two columns.
That test is slower to reach significance because revenue per visitor is a spikier quantity than a conversion flag, which is exactly why so many price tests get decided on the conversion number instead. What neither test can see is what the higher price does to returns, to repeat purchase, or to the customer who bought at $39 last month. A two-week window measures the first order and nothing after it. To use your own test, overwrite the four counts in the panel and the two prices and two margins in the sheet.
The model
It arrives on a tab called Template: Two Prices, One Answer, carrying these columns:
- Price test on the mid-range kettle
with the model computed beside the data:
| Change in revenue per visitor | 0.01778 |
| Change in conversion | -0.1555 |
| Margin per visitor, A ($) | 0.998 |
| Margin per visitor, B ($) | 1.456 |
Once it is in your sheet
- The model arrives with real numbers in it and runs as it stands, so you can press the button first and understand it second.
- Change the numbers to yours. The sheet marks which cells are inputs and which hold formulas, and most labels carry a note explaining the row.
- Press the run button at the bottom of the panel. It is labeled for the tool you are in, and the result lands on its own tab, with a written reading of it beside the figures.
Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.
Next question
- What should you reserve against a recall?What Would a Recall Cost?
- Book the same room again or take the bigger one?Should We Run It Again Next Year?
- Is a second supplier worth paying for?What If the Single Source Fails?
- Is your worst supplier really worse than the rest?Are These Three Suppliers Really Different?
- Does this delivery note match a supplier on file?Which Rows in Two Vendor Lists Are the Same Supplier?
- What does an unreliable supplier cost you in stock?Safety Stock for a Plant That Cannot Stop
Every model like this one, and the method behind them: Statistics in Google Sheets.