[MUSIC] Right, so the first thing we want to think about in terms of customers is what is it that drives price sensitivity, this fundamental thing that dictates how high or how low somebody's willingness to pay is for the product.? So I'm gonna go through some of the most common things that dictate price sensitivity. And I'll leave you as an exercise to think about a coupe of others. The first thing that really drives price sensitivity is ease of product comparison. So if I have one product, product A, that's say manufactured by Johnson & Johnson, and right next to it on the shelf, I have another product that's made by the retailer, a private brand, as Barbara talked about. If they're really easy to compare and I can turn the products over and see that they're essentially the same thing, then I might be willing to buy the unbranded or the private label product. So when comparison is made easy, consumer price sensitivity typically goes up. So as a seller or as a firm, as an entrepreneur, usually what you want to do is you want to make price comparison across you alternative and the competitor alternative somewhat difficult for the customer. You want to focus them on other things like the quality of your product, your service, and so on. The second thing that's very, very important in terms of driving price sensitivity is the amount of overall expenditure that's being done. So if I'm going out and I'm buying new tires for my car and the tires, $500 versus $600, I might be like man, I think I have to get those $500 tires. 100 bucks is a lot of money. But if I'm buying a car for $20,000 and the dealer Amy says to me you know what? You can have these fancy tires for 600 or like the rubbish tires for 500, I'm like well, it's only another 100 bucks on 20,000, it's nothing. So when we're thinking about things in percentage terms or as a small piece of an overall expenditure, our price sensitivity also goes down. So what's the implication for you, the seller? You want to try and get the buyer to think about what they're buying as really a small piece of the overall picture. The other thing, of course, that's related to this is if there's a separation between the buyer and the payer. So maybe I shouldn't say this on a public video, but I've already started down the path so let me do it anyway. So I try to save the school money when I fly from Philadelphia to San Francisco, usually buying a coach class ticket and then trying to upgrade into first class if I'm lucky. But sometimes I might actually buy a first-class ticket, but of course I'm being reimbursed for that. So I'm less price-sensitive because it's not coming directly from my pocket. But it's coming from the pocket of the institution. The other thing that's very, very interesting in price sensitivity is when there's a separation in time or method of payment. So in an earlier video, we spoke about the matching of supply and demand and the example I gave you was Uber, which is a car service where you can take your mobile phone, and you can order a car to come and pick you up. And then the driver, let's say Chris, takes you wherever you want to go. You get out of the car, no money changes hands. Simply what happens is you get a text message, or on the app, you get the bill. So in that case, I don't really feel the pain of payment. The pain of payment. If I had to take $20 out of my wallet every time I used Uber, I might think about walking a little bit more often or taking the subway. But because the payment is happening just purely in my phone and I'm not feeling it directly, I'm because less price sensitive. So that's another thing that makes price sensitivity lessened. A couple other things that are interesting here, when there's a price/quality inference, I've become less price sensitive. So if I needed to hire a lawyer, for example, if I were in some kind of trouble with the immigration service, do I want Amy's cheap lawyers $50 an hour, or do I want Chris's expertise lawyers $500 an hour? In that case, where there's an inference that the higher price leads to a higher quality, certainly for an important service, then my price sensitivity is again lessened. So hopefully, those four things give you a sense of how you can have the customer psychologically feel a little bit less price sensitive. Now, of course, this begs the question of how would one measure price sensitivity? Would you just do it intuitively, or would you try and do some research? What I'm gonna take you through now is there are really four ways to figure out price sensitivity. I'm gonna show on the screen a 2 by 2 matrix that explains these four different ways. You could measure people in their natural environment, buying things or filling in surveys. That's the first column. Or you could run an experiment, sort of an unnatural environment, but it's controlled, either in the field or the lab, or you could engage in something called trade-off analysis. Those are the two columns of this matrix. And then on the rows, you could either measure actual purchase behavior or you could measure their preferences and intentions. So here's an example of an experiment. This was actually done by a colleague of mine at the Wharton school, Steve Hoch, who's a professor here. And he wanted to try and understand for supermarket retailers whether or not they could raise prices or lower prices. So what they did is they cooperated with an institution in Chicago, those of you who are in Chicago, called Dominick's Finer Foods. It's a supermarket chain located in the Chicago area. And what they did was in some of the stores, all of the prices were systematically lowered by about 9% on a bunch of different products, like detergents, paper towels, canned tuna and so on. In another group of stores, the prices were just kept as is. And in the third group of stores, all of the prices on those same group of goods were increased by 9%. So, that's a classic experiment where we have a control group, we manipulate some things downwards, some things upwards, and then we look at what happened. And what they found was very, very interesting. They found that when prices were raised, demand went down by a little bit, as you can see in the light grey, there on the chart, but actually, the drop in demand wasn't that much. Customers hadn't really noticed these pretty small price changes, order of magnitude 9%. So profits went up quite a bit. So this experiment would indicate that those multi-product retailers could probably increase their price a little bit, and you can probably think of some psychological reasons why that works in a grocery environment. Like it's not really efficient for me to pay attention to every single price and try to remember it. Now of course, this study was done in the 1990s. It might be a different thing if they did it today in 2013, because again, I could take out my friendly iPhone, and have my entire grocery list on here. Or I could use an app like the SaveOn app or the SnipSnap coupon app, and from those apps I would be able to remember the prices, or at least my device would do that for me. Second way that we can measure price, and this is a method that was really developed by one of our former colleagues here at the Wharton school, Professor Paul Green. Paul was a very, very influential fellow in the area of marketing, and he was one of the founders of what's been known as conjoint analysis or trade-off analysis. So in a conjoint analysis, you present people with different kinds of stimuli. And what you do is you manipulate certain things. I'll give a personal example on this. When the four students, the former students who founded the company WarbyParker.com selling glasses and eyewear on the Internet, we did a project here together at the Wharton school to try to understand gee, what the heck should we charge for these glasses? What should be the price? Now, we knew, of course, we didn't want to price purely from the cost and the cost plus pricing, we wanted to figure out from the top down the customer willingness to pay the ceiling. So what we did is we presented different groups of people glasses, and we manipulated just one feature of the glasses. So Amy's group would see a pair of nice blue Warby Parker glasses, and the price would be $75. And then we'd ask Amy and the other people in the survey how willing they would be to buy that product. And then the second group with say, Chris' group, we would raise the price to $85, but show the same pair of glasses and see what the response was. And we did this for four prices, 75, 85, 95, and 105, and through the conjoint analysis we noticed the following. Demand was highest at $75. It dropped at $85. $85 to $95 was about the same, and then once we went to $105 it was another drop. So from this pretty rigorous statistical analysis it was clear to us that among those four prices, $95 would be the right price. This is something for those of you out there who want to do conjoint analysis. There are many very good commercial providers and consulting firms that can help you do this, and a lot of good publicly available information. I'd encourage you to learn a little bit about conjoint analysis. We don't have time to go into all the details here, because its also a method that's very useful for trying to value your brand as well as just set prices. Okay, the other two methods that sometimes get used are just direct surveys. Now, when you ask people how much they are willing to pay, you have to be very careful if you ask them directly, because they'll lowball you and give you a really low price. So the better way to ask that question through a survey is indirectly. So again I could say to my friend Amy, here's a pair of Warby Parker glasses for $95. On a scale of 1 to 7, 1 meaning very unlikely to buy, 7 meaning very likely to buy, how likely would you be to buy this product? And maybe the group exposed to that condition gives an average score of 6. And then I take the same survey, and I take it to Chris and a bunch of other people. And I say how likely would you be to buy these pair of glasses were they $105? And I find that the average score from that group on a 7 point scale is only 5. So see the way I'm getting at the price? I'm getting it indirectly, rather than directly. That's the best way to do it through a survey. And, then finally, this is not a direct topic for our course together, but something I do a lot of in my own research, is you could run what's called a regression analysis. So you could take some real sales data, and you could look at various prices and other things that are affecting sales, and you could compute that thing that we've all heard about in economics, the price elasticity of demand, by doing some statistical and quantitative analysis. So those are the four ways that you can really measure price sensitivity. You can run an experiment, you can run some statistical regression analysis, you could do surveys, or you can finally do a conjoint analysis. All four of those methods would allow you to get there. So now I'd like to spend a little bit of time just on psychological factors. Of course, there are a huge number of books written on consumer psychology and so forth. So I just want to touch on the main ideas as they relate to pricing. So in certain countries or cultures, digits have particular meanings. So in Western countries, nines usually indicate a discount or sale or something of that magnitude. And so that's why a lot of products are priced at 3.99, 2.99, 1,999. Somehow, it feels better than just $2,000. So 9's are a special kind of number, at least in Western cultures. The second thing that's important to notice, from a psychological point of view, is the demand curve is not always smooth. It doesn't always go down, if this is price on the x-axis, demand doesn't always fall as prices are increased, and it doesn't always fall evenly. So let me give you an example of that simple experiment that was done in the UK in some supermarkets over there where the regular price of a product of margarine was about 83 cents. And at 83 cents the supermarket was selling a couple thousand units every week. When they discounted the product to 63 cents, the price went down and demand went up, increased by quite a lot, almost 200 percent. So dropping the price 20 cents led to a 200 percent increase in demand. However, when the experimenters dropped the price just a little bit more from 63 to 59, then the increase relative to the 83 cent price was 406. So that's a classic example of a threshold or a nonlinear response to a price reduction. Just by taking off another four cents or a small percentage, you almost double the lift that you got. [MUSIC]
