Watch for anchoring
Recognize when an arbitrary first number is silently dragging your estimate.
Why it works
When estimating, System 1 latches onto any available number — even an irrelevant one — and adjusts insufficiently from it. The first figure mentioned (a price, a quota, a guess) becomes a gravitational anchor, so awareness plus a deliberate independent estimate is needed to break its pull.
How to do it
- Before hearing any number, generate your own independent estimate and write it down.
- When negotiating, notice the opening offer is an anchor and re-anchor deliberately.
- Ask what the number would be if the anchor were doubled or halved — to feel its grip.
Evidence
Anchoring is among the most robust effects in judgment research, replicating across domains, including with anchors the subject knows are random. Jacowitz and Kahneman (1995) quantified the pull with a calibrated measure, showing that final estimates land systematically and predictably between a person’s own baseline and whatever high or low anchor they were first shown. (rct)
Awareness alone provides only partial protection; anchoring persists even when people are warned about it.
Sources
- Tversky & Kahneman (1974), "Judgment under Uncertainty: Heuristics and Biases", Science
- Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131.
- Jacowitz, K. E., & Kahneman, D. (1995). Measures of anchoring in estimation tasks. Personality and Social Psychology Bulletin, 21(11), 1161–1166.
Common mistake
Thinking you’re immune because you know about anchoring. The effect operates below awareness — knowing the name does not switch it off.
Practice this with IX Coach
7 days free, then $40/month (~$1.30/day).
More practices for Thinking, Fast and Slow, Made Usable
- Notice which system is driving
Before deciding, ask whether this is a System 1 (snap) or System 2 (effortful) problem.
- Slow down for high-stakes, irreversible decisions
Match deliberation to the cost of being wrong: spend System 2 where reversal is hard.
- Consider the opposite
Actively argue the case against your first conclusion before accepting it.
- Take the outside view (base rates)
Estimate using the track record of similar cases, not the vivid details of your own.
- Distrust easy fluency
Treat "this feels obviously right" as a flag to check, not a signal to proceed.