Based on the comments to my last post, I need to apologize. Having worked at Microsoft as long as I have, I’ve acquired some verbal habits that have meaning inside these halls, yet fail to adequately convey the full meaning in discussions with people who don’t participate in the same internal discussions on a regular basis. Sorry for the lack of clarity, and, as penitence, I’ll see if I can’t clarify my remarks.
In my last post, I used the term “value proposition” in the singular form. That’s because, here, we almost always use it in the singular form. However, when we do, it’s tacitly understood that we’re really talking about an aggregated concept; that there are really many value propositions. Indeed, given any two users, their respective value propositions are rarely, if ever, going to be precisely identical.
There are both significant and insignificant differences in value propositions, and, when we start talking about “market segments,” we’re really talking about groups of users for whom the differences in their respective value propositions are not significant. And, for the sake of precision, I’ll say that an “insignificant” difference between two value propositions occurs when both value propositions can be satisfied by the same feature implementation.
Moreover, we’re talking about an aggregated quantity, not an objective quantity. There is no such thing as intrinsic value. A number of economists have tried to come up with some notion of “intrinsic” value, but all of these attempts have been based on assumptions that are logically equivalent to giving primacy to one user (or customer)-centered value proposition to the detriment of others. In other words, any attempt to come up with some notion of intrinsic value merely begs the question.
While we’re discussing terminology, it’s worth spending some time on the meaning of the word aggregate in this context. It’s not really a sum in the pure mathematical sense, because we’re not talking about quantities that have equivalent units. It’s rather like adding apples and oranges. The sum is in terms of fruit, and not in terms of any of the individual addends. We use the word aggregate to underscore the idea that we are not simply adding together quantities of identical types.
With this clarification in mind, let’s revisit the issue of market success. One of the comments to my last post suggested that I was equating market success with user satisfaction. That’s not entirely accurate. I claim that market success is presumptive evidence that a product maximizes user satisfaction. That presumption, however, leaves the question of user satisfaction open to contrary evidence. What I hope to do with this discussion of value propositions is to limit the scope of arguments, and the nature of the evidence, that one can invoke to counter the presumption of user satisfaction.
In particular, no argument that involves some form of preferential treatment of one value proposition over other value propositions qualifies. Nonetheless, nearly every argument you’ll see involving product comparisons resolves to some form of preferential treatment of one value proposition over other value propositions.
For example, one of the comments to my last post suggested that Word’s success has a lot to do with “compatibility” both with particular operating system software and with itself in Word’s various incarnations. Economists refer to these kinds of factors as “network effects.” There is, however, no reason to exclude network effects from any particular value proposition. One might be able to design the perfect screw, for example, but if the design calls for a head that’s incompatible with any known screwdriver, then nobody’s going to buy the screw. Such a screw will not satisfy anyone’s value proposition.
The phrase “total cost of ownership” has worked its way into the software industry’s lexicon. Why? Because total cost of ownership has a direct bearing on any given value proposition. TCO includes a number of factors that don’t directly relate to the price, and can well account for an observed fact that the most expensive piece of software, in terms of purchase price, is the most popular in its category. Presuming that the highest-priced piece of software in any given category can’t be the most popular piece of software in that category implicitly excludes a number of other factors that form significant portions of many different value propositions.
Any argument says that product X is better than product Y because product X has a feature that product Y doesn’t have also carries an implicit value preference. The reason we refer to the “vi vs. emacs” debate as a “religious debate” is precisely due to the fact that the entire basis for the debate is conflicting, yet subjective, value propositions.
Now, if all I wanted to do is limit the scope of various product comparison debates, none of this would be all that worth reading. If, however, you’re in the business of creating software for general consumption, and presumably you want to achieve at least a modicum of market success, then you absolutely must understand that your potential customers have a variety of value propositions. It really is worth the effort to segment your market based on those value propositions, because that’s the most effective way to ensure that you’re maximizing aggregate user satisfaction.
Lastly, at the end of my previous post, I’d suggested that there is another topic that relates to value propositions, and it’s particularly relevant to mature products. I hope to make this topic the subject of my next post.