I once attended a sales pitch by some very cute guys from Microsoft. They launched into their PowerPoint presentation, which was co-branded on every page with Microsoft logo and the logo of my company, the prospective buyer. About three-quarters into their sales patter a slide appeared with lots of numbers – and co-branded Microsoft with the logo of one of our biggest competitors. The slide was projected on the screen just long enough for some of our guys to take note of some of the key figures, at which point the presenter set the screen to blank and apologised for the ‘mistake’. But it wasn’t a mistake at all. It was the perfect example of a very deliberate and clever tactic of ‘anchoring’. By silently suggesting that our competitors were paying £X for the same service, they knew that we would be satisfied if we managed to negotiate anything better than £X-1. They suspected that if we came back to our management with a deal at £X-3, armed with this ‘secret knowledge’, we would be heroes.
Another great example of anchoring is what it has done to taxi-tipping in the credit card age. Credit card systems automatically suggest a 30%, 25%, or 20% tip. This causes passengers to think of 20% as a low tip. In fact, data has shown that - pre-credit card days- the previous average taxi tip was only around 8-10%. Anchoring through credit-card systems has seen the average tip for a taxi rising to 22%.
Finally, someone in the restaurant trade told me that most diners tend to order the second least-expensive bottle of wine in an attempt to avoid looking cheap. Consequently, restaurants tend to put their highest mark-up on that very bottle.
The Force
The anchoring and adjustment phenomena was first studied by Amos Tversjy and Daniel Kahneman in 1995. In one of their studies, participants were asked to compute, within 5 seconds and without an electronic calculator, the product of the numbers one to eight (i.e. 1x2x3x4x5x6x7x8=). Another group had to calculate the product of eight to 1 (that is, 8x7x6x5x4x3x2x1=). {\displaystyle 8\times 7\times 6\times 5\times 4\times 3\times 2\times 1}Because those being tested did not have enough time to calculate the answer, they had to make a guess. The first group gave a range of small-ish answers – possibly because the sequence started with small numbers. Their median estimate was 512. The second group – which calculated the product starting at 8 made much higher guesses - their median estimate was 2,250. (The correct answer in both cases is actually 40,320.)
In another test Tversky and Kahneman had testers observe a roulette wheel that was rigged so that the ball would stop on either 10 or 65. Participants were then asked to guess what percentage of members of the United Nations are African nations. Participants whose roulette wheel had stopped on 10 guessed lower values (25% on average) than participants whose wheel had stopped at 65 (45% on average).
In a 2006 study by Dan Ariely conducted at MIT - an audience was first asked to write down the last two digits of their social security number and consider whether they would pay this number of dollars for certain items whose value they did not know, such as wine, chocolate and computer equipment. Participants were then asked to bid for these items. The result was that audience members with higher two-digit social security numbers submitted bids that were between 60% and 120% higher than those with the lower social security numbers. Amazingly, the mere recall of the participants’ social security numbers significantly impacted their bidding behaviour – even though there was absolutely no logical link between that recall and the value of the items concerned.
The Force is Strong
The anchoring phenomenon is very strong. It’s so strong in fact that in one study students were given anchors that were highly questionable – and they still veered towards them. One group of students were asked whether Mahatma Gandhi died before or after the age of 9. Another group were asked if Gandhi had died before or after the age of 140. Clearly neither of these anchors could be correct. However, the two groups were subsequently asked to suggest what age they thought Gandhi actually was when he died. The first group guessed an average age of 50, whereas the second group guessed an average age of 67. The only explanation for this variation is the anchoring effect of the first question they were asked.
Other studies have confirmed the strength of an anchor number even when people are explicitly told that the anchor is “wrong” and that they should ignore it. Research shows that people even persist in veering towards an anchor despite being offered tangible incentives not to do so.
Why do people submit to anchoring?
The reason people submit to anchoring is probably not one thing - it’s a combination of different factors.
Tversky and Kahneman hypothesised that once an anchor is set, people adjust away from it to get to their final answer; however, they adjust insufficiently, resulting in their final guess being closer to the anchor than it would be otherwise be. This presupposed that there is some rational conscious decision-making going on.
More recent research suggests that Tversky and Kahneman were incorrect about the adjustment theory. The anchoring effect holds even when the anchor is subliminal – as in the case with the Microsoft sales pitch that I sat through.
Comentários