Sam reads an 89 percent share of the standard plan as what customers want, and Maya reverses the order on the page to find out whether it is a preference or a position
Eighty-Nine Percent Chose Nothing — Volume I, Understanding Humans Before Products · Decision Making · Two paths · Foundations. Sam reads an 89 percent share of the standard plan as what customers want, and Maya reverses the order on the page to find out whether it is a preference or a position
01 · Opening
Sam reads an 89 percent share of the standard plan as what customers want, and Maya reverses the order on the page to find out whether it is a preference or a position
02 · The setup
Sam has one slide and it is a good one. Eighty-nine percent of new accounts are on the standard plan.
Sam: The market has told us what it wants. We should build for standard and stop hedging.
03 · The setup
New accounts by plan, last quarter — Standard: 89%; Compact: 7%; Extended: 4%; Accounts: 2,310
04 · The evidence
Maya: Which one is ticked when the page loads?
Sam: Standard. It has been for two years.
05 · The evidence
Two readings of one number. Either customers picked standard, or standard was already picked and nobody moved.
Maya thinks: The table cannot tell me which. It never recorded whether anyone touched the control.
06 · The evidence
Maya: Can we run the same page with compact pre-selected instead?
Dev: Half the traffic for a fortnight. Nothing else changes.
07 · The evidence
Two weeks. Same page, same copy, same prices. The only difference is which row starts ticked.
Plan chosen, by which plan was pre-selected — Standard pre-selected — chose standard: 89%; Compact pre-selected — chose compact: 71%; Compact pre-selected — chose standard: 22%; Accounts in test: 1,940
08 · The evidence
Sam: So the eighty-nine percent was us.
Maya: Mostly. About a fifth of them go and find standard anyway. Those are the ones who actually chose it.
09 · The evidence
Most people do not choose. They accept, and acceptance looks identical to preference in every table you will ever build, because the table records what was selected and never records whether anyone touched the control.
This is not a trick being played on users. A page has to start somewhere, and refusing to pre-select anything is also a decision — a worse one, because it hands the choice to whoever happens to be tired that afternoon and nobody owns the outcome.
The useful question is not whether to have a default. It is whether the current one was chosen by a person who understood what it would do, or inherited from a page nobody has looked at in two years.
10 · The evidence
Priya: So which one should be ticked?
Maya: The one that is right for someone who is not paying attention. That is who the default is for.
11 · You make the call
The story does not tell you first.
A. A genuine preference for the standard plan
B. Mostly the position of the option, plus a real minority who chose it
C. That the pricing page is badly designed
12 · What happened
The default stays on standard, and this time somebody decided it on purpose and wrote down why.
Maya thinks: Same tick box. It just has an owner now.
13 · Complete
Acceptance and preference look the same in every table, and only a reversal tells them apart. Next: what happens when you stop deciding for people and give them seven options instead.