Fractal Geometry and Market Analysis
This post on "What is Marketing" got me started thinking about markets and common mistakes by entrepreneur's.
In a software venture it is very common to talk about "the market". As in:
- No point entering such a well established market
- This is an early stage market
- We can have first-mover advantage in that market
- That market is not big enough for us
- This market is growing at 20% a year
- 40% of the customers in this market are actively looking for alternative solutions
- We are the clear leader in our market
- etc. etc. etc.
These statements are treated as absolute truths, when in reality their truth or falsehood is completely dependent on how you define the boundaries of "the market".
Often in these discussions my mind drifts to comparing fractal geometry and market analysis. They are pretty different disciplines, but they do have one important trait in common.
A definition of fractals including some gorgeous pictures can be found here. In brief, a fractal is a geometric shape. Such as this one:
One of defining characteristics of a fractal is self-similarity. Meaning that as you zoom-in and zoom-out you can see the same pattern repeated at different levels of granularity.
Fractals occur in nature as well. For example the coastline of a continent has various bays, points, deltas, inlets, etc. Then if you look at the coastline of one of those bays, you will see more bays, points, deltas, inlets, etc. just at a smaller scale. And so on.
Markets do not normally form fractal patterns. As you look at a product's market at different levels of granularity you may see very different patterns with each analysis.
However, markets and fractals are similar in one important way. In both cases doing your analysis once from one point of view is not enough! In thinking about the market for your software venture you must look at the market patterns from many different points of view. Until you do this you won't know what you have got.
For example, lets say you have a great idea for a massively multi-player (MMP) real-time strategy (RTS) game that is set in Europe in WWII. Well what is the market that you should analyze when making your business plan and when deciding who is your target user?
- market for European Theater WWII RTS MMPs
- market for WWII RTS MMPS
- market for RTS MMPs
- market for combat MMPs (including both RTS and first person shooter)
- market for all MMPs
- market for all computer games
It is easy to see that your business and usability analysis will be drastically different depending which of these target markets you chose as the basis for your analysis.
The above is an example of different ways to slice the market based on changing scale (as you would in looking at a fractal). However, there are plenty of other ways to slice and dice markets beyond fractal-style scale changes. The key lesson from fractals is not that scale rules all. Rather the lesson is that doing your analysis just once with one set of assumptions is not enough to answer your questions and tell you what you are looking at.
However, it is important that you don't get bogged down in analysis paralysis by trying to analyze every possible market definition. You should do more than one, but you should not do them all.
A good quick way to find what market definitions are most interesting is to do the following:
- Pick 3 to 5 comparable products
Make these products as diverse from one another as possible, while still having at least one key trait in common with your product. So in our example above you might pick: a non-MMP WWII RTS game; an MMP first-person combat game; and an MMP non-combat game as you comps.
- Find customers
Find a half-dozen to a dozen people who have recently bought examples of each product
- Identify alternatives
Ask each of those purchasers "when you were buying this product, what alternatives did you consider? In other words if you had not spent the money to buy XXX what would you have spent that money on instead?" In some cases you will get an answer like "I would have spent more money on my Mom's birthday present." which is not too useful. But in other cases you will get very useful answers such as "I gave up my Everquest subscription so I could spend the money on World of Warcraft" or "Well if I wasn't paying for World of Warcraft every month I would probably get the latest expansion pack for Age of Empires."
The point here is that what really makes the "interesting" market definitions is the customers' opinion about spending trade-offs. If the customer has a certain amount of money set aside for monthly MMP fees, then the market is for MMPs. If the customer has a certain amount to spend on all computer games, than that is the interesting market definition as they are making trade-offs among those choices. If the customer is really making an impulse purchase without considering trade-offs than that is also interesting data, at minimum it will impact your pricing , advertising and distribution schemes.
The above mechanism for picking a market definition is not the right one for all circumstances. So if you have a better approach use it. The key thing is to use some kind of analytical approach that will encourage you to be open-minded to new data and new conclusions. Too often market analysis is an exercise in reaffirming your existing opinions through careful selection of target market and relevant data. Try your best to avoid this sort of wish-based planning.
Previous posts talked about the dangers of picking the wrong market, what gets in the way of listening to the customer, and how to deal with an unpleasant truth when you discover it.
copyright 2007 Kerry Champion
