SCIP Oregon

The Pricing Strategy Simulation

Posted by scottswigart - 02/02/09 at 05:02:53 pm

By Mark Chussil, Advanced Competitive Strategies, Inc.

Participants at the January 27 SCIP Oregon meeting faced a challenge. Three, actually. What pricing strategy should they implement in a fictitious ailing industry, mature industry, and fast-growth industry? And who, among the strategists, would win?

We gave each participant information about the industries in which their businesses would compete, and we asked each of them to design pricing strategies for their businesses. They could choose from strategies such as match the market average, cut, tit for tat, do whatever the most-profitable did, and many more, and they could select different strategies at different points in time. All told, they could choose among 14,739 combinations, in each industry, to find the strategies just right for their businesses.

We also asked each participant how much they cared about profitability and how much they cared about market share. After all, we should measure the success of a strategy by whether it provides what the strategist wants. If you want profits, a good strategy gets you profits. If you want profits and you get market share, your strategy failed. Not that market share is bad. But let’s put it this way: if you want to go to Paris and you end up in New York, your navigation failed even though New York is a nice place to be.

Then we fed their strategies into a simulator. The simulator ran each participant’s strategies against every possible combination of two other participants. Since our database of pricing-simulation participants now includes 156 people, that means we ran 12,090 simulations for each participant in each industry. That’s a total of 1,886,040 simulations per industry, or 5,658,120 in total. The simulation software ran them in about 2 minutes. No, it’s not an Excel spreadsheet.

Participants would get a high score if their strategies 1) usually got a lot of whatever they wanted (profits, share, or a combination), and 2) got it regardless of what competitors did. In other words, we want strategies that are successful and robust, not just lucky. That’s why we ran all those simulations of each person’s strategies against everyone else’s strategies.

In theory, scores range from 0 (you never get what you want) to 100 (you always get what you want). In practice, the highest score ever recorded in this simulation is 90 and the lowest ever recorded is 20. It’s very hard to score higher because competitors sometimes do silly or nasty things, and it’s very hard to score lower because competitors sometimes do silly or generous things. (Whether the silliness, nastiness, or generosity is intentional is another important subject.)

The winners!
Interesting. One person won in the Ailing industry; a second person won in the Fast Growth industry; and a third person won in the Mature industry. Our congratulations, respectively, to Roger, Mike, and Anonymous. (Someone provided strategies and chose not to provide a name.) Across the three industries, Roger had the highest average score.

Interesting too that two of the three champions also had an Achilles heel. As a result, a fourth person, Scott, did very well overall even though he wasn’t number one in any single industry. It’s as though he’s a decathlete competing against the strongest in individual sports.

None of the participants developing strategies on January 28 were among the best ever and none were among the worst ever.

Analyzing the results
The participants will receive a full set of charts describing the results of their strategies. Here we tease and tantalize you with a couple of extra-provocative charts. Okay, okay, that’s a bit overheated. Not so much, though, if you really like simulations, which I do. I’ve never seen any other business analysis produce charts like these.

The chart below analyzes success in the Ailing industry for one of the January-28 participants.

Each dot represents the average performance of the 12,090 simulations run for a strategist. The green triangle for “hold” is a baseline: it’s what would happen by following a strategy that never changes its price. The red square is the performance for this participant. This person averaged 36.3% market share and -13.9% profitability (return on sales, or profit divided by sales) in this difficult industry against the 155 tough competitors who also developed strategies.

Notice that there are no dots — that is, no strategies — to the northeast of this person’s strategy. That means that there is no other strategy (among the 155 competitors) that would improve this person’s performance on both market share and profitability. There are many that would improve share or profitability, but a different strategy would require that this person sacrifice some performance on one measure in order to improve on the other. Of course, that requires a value judgment that only this person can make.

It is possible to see if any of the other strategy combinations would provide this person with results that improve both share and profitability. (There are roughly 14,600 other combinations to test: 14,739 minus the one chosen by this person and minus those chosen by the other 155 competitors, with a few duplicates taken into account.) We won’t go into that here; there are already quite enough numbers buzzing around. We will note that such a simulation would take about 3 hours. Definitely doable.

There’s room for one more chart, though, which appears below. Here we’re looking at the robustness of a different participant’s pricing strategy in the Mature industry. The horizontal axis is the strategy’s performance on profitability (ROS, in blue) and market share (red), measured as the percentile score achieved by the strategy. The vertical axis is the number of simulations, out of the 12,090 run on this strategy, that performed at those percentiles.

We see that this person’s strategy has a strong peak between the 60-69th percentiles for market share, with a smattering at other levels. That means this strategy is fairly robust: it has a high likelihood of producing a certain level of performance, across a wide variety of competitor strategies. (Note that a robust strategy is not necessarily a good one. It is possible to be robustly lousy.) In other words, the market-share performance of this strategy is largely up to the strategy itself, as opposed to being vulnerable to competitors’ moves.

Contrast that robust performance on share with the much less robust performance on profitability. The profitability results are all over the chart, which means that the profit impact of this strategy greatly depends on what competitors do.

And in conclusion
And in conclusion
There’s much, much more to learn from the simulation, of course. Those who provide strategies receive not only their scores but also a wonderfully fascinating white paper full of strategy insights. Sorry, we don’t provide the white paper to those who have not entered strategies. That’s because the white paper will contaminate your thinking (and therefore our research) if you decide to supply your strategies later on.

One last thing we learn from the simulation is the simulation itself. This form of analysis, involving millions of simulations, is easy to run but hard to write, and it has only become practicable in the age of fast computers. The simulations that took 2 minutes on my computer would have taken hours or days not many years ago, and of course would be impossible without computers. As computing power has grown we have gained calculators, the electronic spreadsheet, conjoint and discrete-choice analysis, Monte Carlo simulation, data mining, and now Decision Tournaments. Just as the Internet has revolutionized data gathering, intensive strategy simulation can revolutionize strategy analysis.

Our thanks to all the intrepid strategists who participated in the simulation, and our congratulations to Mike, Roger, and Scott! And Anonymous.

You can participate in the pricing simulation. It’s easy and it’s free. Write to info@whatifyourstrategy.com and we’ll send you an entry form.

This process is called a Decision Tournament™. The simulator was designed and built by Mark Chussil of Advanced Competitive Strategies, Inc., designer of the award-winning ValueWar® strategy simulator and the patent-pending DXMA™ crisis simulator. It can be customized for real-life business situations.

Search