How many invitations does sixty replies take?
Sixty random draws from a list of three hundred come back with fifty-four different customers. Drawing seventy-five to end up with sixty is the fix, and the seed is what makes the sample something you can hand to a client.
Words on this sheet
- Seed: The starting number for the random draws. The same seed gives the same draws every time.
Marketing Starter Sampling free
After you install, this is the model to open.
Pull the People to Survey
- 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
- Draw 60
- 54 distinct customers; six draws repeat people already picked
- Draw 75
- 63 distinct at the same seed, clear of the 60 needed
- Invitations
- 172 for 60 completes at a 35% response rate
- The seed
- 314 makes the sample reproducible for a client
A list of 300 customers, identified by a numeric ID because the sampling tool reads numbers rather than names, which is the standard way to sample a list of people. Click Run with the method set to Random, 60 samples and a seed of 314: the tool returns 60 draws beginning 5098, 5144, 5275, 5224. Count the distinct values and there are 54, not 60, because six of the draws landed on customers already picked.
Random sampling here draws with replacement, so every draw is independent and somebody already picked can be picked again. This draw is exactly typical: the arithmetic beside the sampling plan says 60 draws from 300 return about 54 different customers on average, and on a list of 100 the same 60 draws would return nearer 45. The fix is arithmetic rather than a setting: draw more than you need and remove the duplicates.
Run it again with 75 samples at the same seed and you get 63 distinct customers, clear of the 60 you wanted. The second block on the sheet carries the larger correction, and it has nothing to do with sampling at all. You need 60 completed responses, not 60 invitations, and at a 35 percent response rate that means inviting 172 people. Sizing the draw to the number of completes you need is the most common mistake in survey fieldwork and it is not one the tool can warn you about.
Two more things to be careful with. The Email valid column says 26 of the 300 have no usable address; filter those out of the list before you sample rather than after, because removing them afterwards biases the sample towards whatever the unreachable ones have in common. And the Segment column is deliberately uneven, 135 retail against 33 wholesale, so a simple random draw returns the segments roughly in the proportions they occur.
That is right if you want a picture of the base and wrong if you want to compare segments, and for a comparison you draw separately from each segment by running the tool once per segment with the input range narrowed to it. Always set a seed on a sample a client will see. The seed is what turns "we picked 60 customers at random" into a sample anyone can reproduce, which is the difference between a defensible method and an assertion. To use your own list, put a numeric ID in the first column and point the input range at it.
The model
It arrives on a tab called Template: Pull the People to Survey, carrying these columns:
- Customer ID
- Signed up
- Segment
- Last order (months ago)
- Email valid
with the model computed beside the data:
| Distinct customers a draw this size returns on average | 54 |
| Invitations that implies (count) | 172 |
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
- How much is riding on your one big channel?What If the Ad Platform Changes the Rules?
- Did the campaign really move the brand tracker?Two Big Samples, One Real Difference?
- Which media split buys you the most orders?Three Media Plans, Compared
- Which client should you turn down?Which Retainers Fit the Team We Have?
- Would three extra days move your critical path?What Sets the Go-Live Date?
- How many retainers can you afford to lose?What If Two Clients Leave at Once?
Every model like this one, and the method behind them: Statistics in Google Sheets.