I just saw the image below on dataviz and was astounded.
The image clearly and dramatically shows the non-linear relationship between the quality of a wine (as defined by Gary Vaynerchuk’s rating out of 100) and its price. A couple of observations:
- The graph clearly demonstrates a common refrain I hear people say about wine, “I can generally tell the difference between cheap swill and good wine, but once the price gets over $20-30, it’s really hard to know why one wine is better than another.” Check out the point of diminishing returns (where the line starts to flatten out) which is right around $20!
- Personally, I’d rather spend $15-20 and like my wine than spend $6 and have to fight my way through it and I admit that, in absence of deeper knowledge about specific labels and vintages, and in the rush to make a purchase decision and move on to the next task in my busy life, I use price as an indicator of quality. On average, a $15-20 wine is going to be objectively better than a $5-10 bottle, and these data bear that out.
- At the same time, I know people (and I’m related to some) for whom bragging to their friends that “this wine was only $6!” is key, if not the key to their enjoyment of it. This graph shows that even in that price range ($5-10) there is a huge range in ratings, suggesting that the savvy buyer can find excellent deals. In fact, I would love it if my local wine store a) curated their selection based on this graph (selecting only wines that are “above the line”) and b) placed each wine on the graph at the point of purchase. I would have greater confidence that I was getting good value and I’d be very loyal to that store.
- Based on on this analysis, if you are buying wine at prices over say, $50, you either really better know what you’re talking about or just admit that you’re buying at a price point or a label mainly to impress people. Cue, Thorstein Veblen’s “conspicuous consumption.”
- It would be very interesting to super-impose “volume sold by price point” over this analysis. Almost certainly, it would be an asymptotic curve starting from the top left and ending at the bottom right – nothing earth shattering there; basic economic theory – price goes down, volume goes up. But what would be interesting, if you are a wine maker, would be to understand, within a more specific price band, how much the rating affects volume. Objectively speaking, are there increasing returns to investments in quality or does price point dominate?
On a broader level than just wine, I’ve been thinking that someone should do this analysis in other categories. For example, my company sells skin care products at prestigious retailers such as Neiman Marcus at Bergdorf Goodman. These stores have small bottles of creams that cost $100, 200, $500 even $1,000. Sephora has creams that sell in the $30-$200 range. Walgreen’s has products that sell between $5-50. Which are better? Is a cream from Saks worth 10x a cream from CVS?
It would be interesting to create a graph like the one above for skin care, using customer ratings as the (somewhat) objective quality rating. In fact, it would be interesting to see if consumers would buy that way…maybe this is the future of ecommerce…