"Nine, that's a magic number" or so De La Soul might have sang if they were marketers rather than a New York hip-hop trio. An increasing body of marketing evidence shows that consumers, rather than being rational decision makers, are prey to a number of biases. One of the most interesting biases revolves around the positive impact of prices ending in nine, known as charm prices. For retailers this should be a reason to be cheerful as it means sales can be encouraged with less need for margin destroying price cuts.

At ZenithOptimedia we have run experiments amongst 650 consumers across 6 products, from TVs to bread. For each product we asked about value perceptions. The twist was that we had discrete cells of consumers. Some consumers saw prices ending in 99p, the other groups saw the same good as just one or two pence more expensive. Despite the minor variations in price consumers were 9% more likely to think a brand was good value when the price ended in 9p – this occurred even though the difference in prices was c1%.

This process seems to be exacerbated in real-world shopping situations when consumers are rushed and harassed. A French study found that when pizzas prices dropped from €8.00 to €7.99 they boosted sales by 15%. In 2013 Gumroad, a site similar to ebay, also released figures showing that goods that ended in 99p sold significantly better than those with round prices.

What makes prices seem so much better value when they drop by just one pence? One hypothesis is known as the "left hand digit effect". This theory suggests that consumers give undue prominence to the first digit. In a quest for simplicity they encode a price such as £6.21 as just £6.

However, a study by MIT and the University of Chicago showed there might be other forces at work. They partnered with a mail order firm to test the impact of charm prices on a particular item. At $34 dollars the item sold 16 units at $39 it sold 21 units and at $44 it sold 17 units. The left hand digit effect explains why the $39 pricing sold better than $44 but there must be more going on to explain why $39 sold better than $34. This leads us to the second theory…

Consumers have been exposed to thousands of sales through their lifetime. Repeated exposure to sale prices ending in 99p has led to a strong cultural association between nine and a deal. Like Pavlov's dogs our deal instinct begins to bark when we see a "charm price".

So why's this of interest to retailers? Well it seems that they may be moving away from these lessons. When we looked at 330 prices tracked by BrandView across the five main supermarkets, the goods were more than three times more likely to end in a zero as they were in a nine. An explanation for this may, paradoxically, be an over-reliance on consumer claims. If you ask consumers directly they'll say that a penny's difference is immaterial. We asked 312 consumers if their purchases would be swayed by that token amount – and the 7% who said they'd be a lot more likely to buy were off-set by the 7% who said they were a lot less likely to buy the goods. However, consumers' claims are notoriously unreliable. As David Ogilvy said, "Consumers don't think what they feel, don't say what they think and don't do what they say". Far better to rely on behavioural data which shows the significant influence of charm prices.

The lesson for retailers, who are currently investing hundreds of millions in price cuts, is that consumer decisions are made on perceptions of value rather than absolute value alone. The evidence from sources as diverse as ZenithOptimedia and Gumroad suggest that the use of charm pricing is an under-utilised strategy for boosting value perceptions. If these experiments don't persuade you then perhaps we should turn to the sage advice of Roger Sterling, from Mad Men, who said "I'll tell you what brilliance in advertising is: 99c".

And the price of De La Soul's album on Sainsbury's: £6.99… perhaps they were onto something after all…