Dealer AI HQ

Evidence review

The Engagement Science of Asking One Question at a Time

Question-asking research from Harvard, choice-overload studies, and cognitive-load theory — why conversations that ask less convert more, and how it applies to dealership BDC.

· 8 min read · Dealer AI

Key takeaways

  • In research published in the Journal of Personality and Social Psychology (Huang, Yeomans, Brooks, Minson & Gino, 2017), people who asked more questions — especially follow-up questions — were consistently better liked by conversation partners.
  • Follow-up questions signal listening: they can only be asked by someone who processed the previous answer, which is why they outperform scripted question lists.
  • Choice-overload research (Iyengar & Lepper, 2000) found that presenting fewer options dramatically increased purchase rates — a caution against replies that stack three questions and four vehicle links.
  • The practical synthesis for dealer conversations: one question per message, built on what the shopper just said, with options narrowed rather than multiplied.

Why 'engagement' has an actual literature

Most conversation advice in sales is folklore. But two well-replicated research streams bear directly on how a dealership should text a lead: question-asking research from behavioral scientists at Harvard, and the choice-overload literature from consumer psychology. Both converge on the same design principle — ask less per message, and build each message on the last one.

The question-asking studies: follow-ups win

Huang, Yeomans, Brooks, Minson, and Gino (2017) — "It Doesn't Hurt to Ask: Question-Asking Increases Liking," published in the Journal of Personality and Social Psychology — analyzed live conversations, including speed-dating and online chats, and found that people who asked more questions were better liked, and that follow-up questions carried the effect.

The mechanism matters for conversation design: a follow-up question is unfakeable evidence of listening, because it can only exist if you processed the previous answer. "Are you still in the market?" is a question. "You mentioned the Silverado needs to tow your camper — what's it weigh?" is a follow-up question. Only the second one builds the relationship.

Choice overload: why stacking questions backfires

Iyengar and Lepper's field experiment (2000) is the classic: a display of 24 jams attracted more browsers, but a display of 6 jams produced roughly ten times the purchase rate. The broader literature is nuanced about when choice overload occurs, but the direction is consistent for low-attention contexts — and a text message glanced at during dinner is the definition of a low-attention context.

A reply that asks about budget AND timeline AND trade-in AND preferred color gives the shopper four decisions to make before they can respond. Most resolve the overload the easy way: they don't respond. One question per message keeps the reply cost near zero, and the answer to each question makes the next question smarter.

Application: conversation design for BDC (human or AI)

The synthesis of both literatures is a compact rule set that applies equally to human BDC reps and AI agents. It is also measurable — a conversation log can be audited against every rule below.

  • One question per message. If a draft contains two question marks, cut one.
  • Prefer follow-up questions built on the shopper's last answer over new-topic questions.
  • Answer their question before asking yours — reciprocity precedes discovery.
  • Narrow options rather than multiply them: two appointment slots beat 'when works for you?'; one strong vehicle match beats four links.
  • Match the shopper's register — length, formality, energy. Mirroring reduces the friction of replying.

Frequently asked questions

Why should a BDC only ask one question at a time?
Behavioral research shows follow-up questions increase liking (Huang et al., 2017, JPSP) while multiplying simultaneous choices suppresses action (Iyengar & Lepper, 2000). One question keeps the shopper's reply cost near zero; stacked questions get skipped.
What is engagement science?
Engagement science applies replicated behavioral research — question-asking studies, choice-overload experiments, cognitive-load theory — to conversation design, so outreach decisions are grounded in evidence about how people actually respond rather than sales folklore.

Sources

  1. Huang, K., Yeomans, M., Brooks, A. W., Minson, J., & Gino, F. (2017). "It Doesn't Hurt to Ask: Question-Asking Increases Liking." Journal of Personality and Social Psychology, 113(3).
  2. Iyengar, S. S., & Lepper, M. R. (2000). "When Choice is Demotivating: Can One Desire Too Much of a Good Thing?" Journal of Personality and Social Psychology, 79(6).