Ok, so you want to give a discount. After deciding how much you can afford to mark off, the next question may be whether to frame it in terms of a dollar amount off or a percentage off.  But which is better?  Is there any reason to think one is more effective than the other?

As it happens, yes, there are some differences. Some of the issues at play here include first of all, which is more effective at getting people to buy during the promotion; secondly, how does it affect whether people will buy after the promotion is over? Does the depth of the discount make any difference in all this? And how might the discount effect people’s expectations about the price once the promotion is over with? 

DelVecchio, Krishnan, and Smith (2007) ran a few studies to look at these questions, and found that, in a deep-discount situation, a percentage-off, being ever-so-slightly more demanding to calculate, leads people to expect higher prices when the discounted item isn’t on sale. This is good, because you don’t want people to find your regular prices too high, and you want to minimize any negative price-quality inferences they might make based on sale price.  And it seems that during the sale itself, people respond equally well to (meaning they’ll buy) both percentages off and cents-off.  So there are a couple reasons why percentages off are better, and no strikes against.

The same basically holds true for the period after the sale ends. There’s a bit of a difference between deep discounts (about 40% off in this study)and lower-depth discounts (15% off), in that the percentage-off deep discounts enjoyed greater sales than the cents-off deep discounts.  Not much difference was found between the two frames for lower-depth discounts.

Take-home message: if you’re going to offer deep discounts (and recall that that’s not necessarily a good long-term strategy), express them in terms of a percentage off, preferably not a super-easy-to-calculate percentage off. The reason behind the advantages of percentage-off sales seems to be the increased difficulty of making the calculation, so if you hand customers an easy problem like “50% off”, it kind of ruins the effect.


DelVecchio, D., H. Krishnan, and D. Smith. 2007.  Cents or Percent? The effects of promotion framing on price expectations and choice. Journal of Marketing 71: 158-170.


January 27, 2009

An earlier article I had read suggested that the value of coupons and premiums should stay below 20% of the product value, or you run the risk of negative brand perceptions.  Now I’ve had to revise that conclusion due to some new research* suggesting that promotions that include free stuff may actually be better than previously thought, because the value of free is way, way more than just zero.

The researchers did a series of experiments where they offered chocolates for sale– in one experiment, they offered Hershey’s kisses at a negligible price and Fererro Rocher at slightly higher price (2¢ and 27¢ in one condition, and in the second condition, 1¢ and 26¢).  They then reduced the price of each product in by 1¢, to 1¢ and 26¢ and to 0¢ and 25¢, respectively.  Here’s the cool part: even though the monetary value of the discount was the same across the board at 1¢, the demand for the chocolate offered for free went waaaayyy up (yes, that’s the technical term– ‘waaaayyyy up’).  Much, much more than the demand increase for the chocolate with the price reduced to 1¢. 

Why on earth would this be? It turns out that zero has a special status in a number of ways, especially when the zero in question is a monetary value.  For example, there are some interesting social norms that come into play when something comes for free– if you let people buy candy for 1¢ apiece, they’ll take an average of four pieces. If you offer it for free, more people will take it, but they tend to take only one each. And instituting a penalty for bad behavior (tardiness in picking up children at day care) can actually increase tardiness.  People doing things for free often put out more effort than those giving their services for some small amount. Free is different.

The researchers did another series of experiments trying to get at what drives the increase in demand they had found.  It doesn’t seem to be the social norms, but rather affect– this may not come as a surprise to you, but apparently free stuff makes people feel good. A lot better than its monetary value might indicate, in fact.  It’s a purely an emotional response– once you make people think a little about which product they actually like better (like they did in Experiment 6), it can make the zero-price effect nearly disappear.

The researchers suggest that if you’re willing to offer an item on discount, maybe it would be better to offer it (or a different, less costly item) completely free.  So maybe the value of a Buy One Get One Free offer is much more than 50% off, after all.  And free shipping! We already knew that was good because it doesn’t add cost onto a transaction, which can discourage new buyers– but apparently it may get you a lot of extra good feelings, too.   And, if you can manage it without sounding preposterous, perhaps it would be better to describe some items as coming ‘free with purchase’ rather than ‘included in price’.  I’m thinking about chains or cords for pendants, here.

Now, we should note that they didn’t do experiments to find out whether people make any assumptions about the quality of free goods (and about the brand in general).  They did ask the study participants some questions afterward, and none of them happened to mention anything about brand quality, but that’s suggestive at best.  Nor has there been any work done (that I’ve found yet, anyway) to see how well free items attract new customers and repeat business. Maybe you can get people to try your product, but will they come back?   Do the good feelings created by offering free stuff carry over as a positive attitude to the brand as a whole?   


*K. Shampanier, Mazar N., Ariely D. 2007. Zero as a special price: the true value of free products.  Marketing Science 26:6, 742-757.

I don’t tend to run promotions, specials, or give out coupons, but a lot of artists put a lot of time and effort into their promotional activities, so I’ve become curious about how well this really works– not just in the short term, but over the long term.  Do the new customers you get through promotions become repeat customers? Do constant sales undermine your ‘normal’ prices? What does it do to price-quality inferences?

I managed to scare up a meta-analysis* relevant to these questions. They looked at 51 past studies and did some formal comparisons and stats to get an idea of the bigger research picture of how promotions affect people’s brand preferences over the long term.  A caveat: much of the research reviewed concerned consumable goods (grocery store items); some info was relevant to durable goods, but who knows how applicable the findings would be to luxury goods (like art).  Anyway, given that, here are some highlights:

Promotions aren’t unambiguously good or bad. It depends on your product, and the kind of promotion you use. Dropping your prices without warning or explanation is about the worst thing you can do– people will wonder what happened, assume your quality has gone in the toilet, and will not come back unless they are related to you by blood (actually, I made that last bit up– there’s actually no research to indicate that you even your mother will buy your work again after an unannounced price cut.)(<– joke).

So, temporary sales are best.  Coupons or premiums (like a Buy One Get One) are the most effective types of promotions.  Coupons are especially good, since it seems that the very act of printing/clipping the coupon can make the consumer like the product more (at work here is a kind of after-the-fact rationalization that if they put some effort into obtaining the product, they must like it more than other products). But for coupons and premiums, keep the value of the discount under 20% of the product’s price– more than that, and it will start to affect your brand in a negative way. That means your BOGOs should be more in the range of Buy Two, Get One Half Price (16% off) rather than Buy One, Get One Half Off (25% discount) or Buy One Get One Free (50% off).

Moreover, the negative effect of promotions was worse when a) the brand was unfamiliar or new to the buyers, or b) the promotion was for a durable good or service. If I’m reading the stats right, the combined negative effects of promotions for unfamiliar, durable products are greater than the positive effects of using a coupon promotion.  That’s a red flag for most artists trying to attract new customers through promotion. Maybe save your 10% off coupons for folks already on your mailing list, who have already purchased your product and are familiar with its high quality. 

That’s not to say all sort of promotions can’t bring sales in the short term. But it looks like you need to consider your promotional choices very carefully if you care about the long-term health of your brand.

Questions for later: What about giveaways? How about sales donating a prtion of the price to charity? Good or bad in the long term? Stay tuned.


*DelVecchio, Henard, and Freling. 2006. The effect of sales promotion on post-promotion brand preference: A meta-analysis. Journal of Retailing 82: 3, 203-213.

In a recent issue of Marketing Science, there’s an interesting article on brand building*.  Now, the study concerned consumables, not durable goods, but insofar as jewelry attracts repeat purchase, I think the results are at least suggestive for artists.  They compared a number of marketing strategies over the long term (52 weeks) and found that distribution breadth had the greatest long-term effect on growth.  Discounting had a large short-term effect, but over the long run actually had a negative effect.  The effect of advertising was about 1/3 the size of the distribution effect.  Other strategies that had a positive effect on growth were display/presentation (eye-catching presentation that caused buyers to notice the brand in the store), distribution depth (how many different items of a brand a store carries), and line length (how many different items a brand offers in a store in a given week).  Most interestingly, wide distribution enhanced the effects of all of the other strategies– so for example, if you put x dollars into advertising, the advertising got better results when there was wide distribution as opposed to limited distribution.

So if you have a limited marketing budget (and who doesn’t?), you’ll do best to spend the better part of it on getting as wide a range of your work as possible into as many venues as possible. Advertising is all well and good, but don’t spend to much: the researchers report that it’s only profitable if its costs are “less than .8% of marginal retail value” (as opposed to distribution, which is profitable at 23%).  Don’t ask me to explain “marginal retail value”– I tried to get the house economist to explain it in a nutshell, but it was still tricky. Let’s just say that it involves factoring in the value of repeat sales.  The gist is clear: biggest bang for your buck with money spent on increased distribution.

Etsy sellers, take note: if your online shop is the only place for buyers to find your work, an emphasis on sales, discounts, and advertising may not pay off as much as you’d like.  

I came across this article after I had already made the decision to shift my energy from my Etsy shop to reaching out to more retailers this year, but I’m more convinced than ever (despite the limitations of the study and of my layman’s interpretation of its findings) that this will turn out to be the right decision for me.

*Ataman, Mele, Heerde. 2008. “Building Brands”. Marketing Science 27:6, 1036-1054.