One of the first learnings I had in the on-line business was the trade off between offering the customer a lot of different options, and actually having the customer choose an option in the first place. The more options you throw at the average customer, the greater the likelihood of the customer continuing without an up-sell/accessory, or worst case abandoning the purchase process all together.
That said, there is some granularity to it. For some customers, e.g. the tech savvy recurring advanced customer, the absence of an option can mean the loss of a sale.
So how do you optimize the option range vs. up-sell conversion trade off?
A friend of mine who used to head up a movie theater operation explained to me that no matter how they priced popcorn, 80% of visitors chose the mid sized option (out of small, medium and large) and 75% always chose some kind of bundled offer with soda (note this is in Europe, in the U.S. I doubt anyone choses anything smaller than XXL) .
What was even more interesting was that margins in the movie distribution business are squeezed by the immense bargaining powers of Hollywood. So most of his margin $ were generated not by George Clooney but by fried corn, soda and candy. The situation is similar in the computer industry where the systems themselves for the most part generate little margin.
So how does this relate to accessories and up-selling in an on-line store, configurator or basket? For starters the Pareto rule applies here. My experience tells me that two to three choices out of five or more published options will capture 60-95% of the users that chose an option. These options are all you need (see below an example from Amazon default up-sells for a Garmin Nüvi). Adding more options than this will in most cases hurt your conversion of this option all together even though may sell more types of up-sells.
My advice is to on a monthly basis keep track of the number of up-sell options per assessory/upsell and your total up-sell options so to keep check out process at a minimum with conversion and margin dollar at a maximum. This will ultimately be a test of how good you are at saying no to the various product managers in your e-business.
2 responses so far ↓
Brian // March 12, 2008 at 9:57 pm |
Hey Philip — totally agree with you here. But, given what we both believe, why would you let the configurators in your region have so many options? We can control that.
Does your customer really need the choices? « Philip Hallenborg’s Online Business Blog // April 10, 2008 at 12:37 pm |
[...] is where the challenge starts. I have previously written about the pareto type distribution of up-sells on an option with say five choices. Two options will be chosen by the bulk of customers, [...]