COVID19 Conduct your courses online, or Blend onsite and distance learning with our web-based simulations! We are here to assist you.
By Robert Duke, Senior Teaching Fellow, Leeds University Business School, UK
Markstrat and I go way back. My first taste of the marketing simulation was as a participant rather than an instructor, as part of my MBA studies in 1986. It made a deep impression on me, one that endured until I became a lecturer at Leeds University in 1987 in what was then the Department of Management Studies.
I persuaded my professor to buy a site licence for Markstrat, then called “Markstrat D”. The software came on two, five-and-a-quarter inch floppy diskettes, less than 720 kilobytes of data in total. Amazing how such a powerful program could be so small.
Every year since then, I have used Markstrat on undergraduate and MBA courses. In a lazy moment, I performed a rough calculation of how many students I have guided through the Markstrat world and how many simulations I have run. At the latest count, it is about 50 simulations and about 1,500 students.
How to allocate firms in Markstrat: a bidding process
Over the years, I have added my own enhancements to Markstrat. One I would like to share with the community is my bidding process. Instead of randomly allocating Markstrat firms to teams, I allow teams to bid for and win the Markstrat firms they go on to manage.
This bidding process takes advantage of the different company start positions that are characteristic of most scenarios. This characteristic presents the possibility that some Period Zero positions are slightly better than are others. What makes this even more interesting is the subjectivity regarding what makes one Period Zero position better than another. Is it better to win a firm that has both of its offers in market segments that are profitable in the short run, or to take over a firm with both of its offers in market segments that promise large sales volumes in the long run? The third option is to place highest value on a firm that has one of each type.
To facilitate this bidding process, I give all teams full sets of Period Zero data – market, industry, and company reports on all firms. I then tell them to analyse it, rank the firms’ start positions from best to worst, form an opinion on how much a good start position is worth in Markstrat Dollars over a less good one, and then send me their bids.
In this “sealed bid” process, each team must submit a dollar bid for every Period Zero firm, which means that in, for example, a five firm scenario I receive 25 bids. To allocate firms, I identify the single highest bid, and use that to allocate the corresponding firm to the bidding team. I then remove that firm and team from the process, and repeat using the remaining highest bid. In a simulation with five firms, I repeat this five times, at the end of which every team has a firm. I have created a spreadsheet to help with this process, using conditional formatting to highlight the highest value in the table of bids. As I delete the team and firm for each successful bid, the next successful bid shows up.
The time needed
Of course, this is a big challenge for participants, but fortunately, I use Markstrat as the centrepiece of a university course that stretches over two semesters. This allows me to give student teams two weeks to make every Markstrat decision, including formulating their bids. With two weeks for every decision, students have the opportunity to wring every drop of learning from the PAKs that Leeds University Business School buys for them.
How to best incorporate a bidding process
I incorporate the bid process into the simulation in a way that makes it meaningful but does not adversely affect the learning experience. Simple and effective, I merely subtract each team’s successful bid from its cumulative net contribution at the end of the simulation to create a second criterion for winning : bid-adjusted cumulative net contribution. I keep this outside the simulation rather than making bid repayment an extraordinary cost within it.
This is not totally realistic of course. This adjustment does not affect share price index (SPI, the first criterion for winning) and it does require a peculiar interpretation of the bid: the successful bid becomes money the team must repay at the end of the simulation, including interest and charges. However, this approach means that the bid process remains external to the simulation, and so a team cannot spoil the core of the Markstrat learning experience by being stuck with repaying an overlarge bid.
However, this does create two criteria for winning – bid-adjusted cumulative net contribution and SPI. Therefore, you might ask, what happens if one team wins on one criterion and another team wins on the other? In such cases, I decide the winner based upon the degree to which the two teams beat each other on each criterion. For example, if team one has bid-adjusted cumulative net contribution 5% higher than team two, but team two has SPI 10% higher than team one, team two is the winner.
A learning curve for students
The bid process is a big challenge, and students often submit bids that they reflect on later with astonishment (“what were we thinking?”). This gives another opportunity to highlight how much they learned since they first encountered Markstrat. Sometimes, teams will bid high for what they thought at the time was a good start position, and later conclude that they paid a fortune for the worst firm.
A daunting task faces the teams. On the one hand, the team has a huge amount of Period Zero data; on the other, they have to figure out the relative values of the four, five or six Period Zero start positions in Markstrat Dollars. To do this, I recommend a powerful tool.
A powerful tool to compliment the bidding process
I utilise the Strategic Triangle as described by Kenichi Ohmae in his book Mind of the Strategist. The three parts of the triangle are customers, company and competitors, although during this bidding process, the last two become one: company and competitors are the four, five or six start positions, and the bidding process outcome decides which one of them becomes the team’s company and which three, four or five are to be their competitors.
Teams are usually predisposed to start their analysis with company / competitors, but I strongly recommend that they begin with market analysis. I focus this analysis on four major criteria for segment attractiveness: size, growth, profitability and competition, and suggest a systematic approach to market analysis that I have dubbed Semi-Quantitative Segment Analysis – SQSA. I call it ‘semi quantitative’ because it involves numerical analysis, but determining the numerical values for some criteria (profitability and competition) involves judgement more than calculation. The analysis focuses on filling in this table:
Teams can decide what types of numbers to insert into the spaces. The simplest SQSA approach is to rank segments against the criteria, but more advanced is to score them, and even more advanced (and more subjective) is to determine scores and multiply scores by weights, where weight indicates the relative importance of the criterion. The approach used in the General Electric Matrix inspired my SQSA structure.
The totals then indicate which sonite segments the team judges to be most attractive. However, given that much of the analysis that leads to these totals is subjective, we can expect teams to generate a variety of different outcomes.
Once the teams have decided which segments are most attractive, the next task is to determine which firms have the best offers in the most attractive segments in Period Zero. Teams decide this via exhaustive Four Ps analysis of existing offers.
The big reveal
It is a pleasure to see the anticipation on students’ faces as, in a lecture, I announce the outcome of the bid process, and the astonishment visible when I reveal the size of the highest bid (“they bid how much?”). It is inspiring to see how engaged they are with Markstrat even at this early stage.
Enriching the Markstrat Experience
There are several reasons why I like to use this bidding process to allocate Markstrat firms. First, I can say that the team’s firm allocation resulted from their own analysis and bids, so there is more justice than would have come from random allocation. More importantly though, it gives participants a relatively gentle introduction to Markstrat. Teams have the challenge of analysing a bewildering amount of data (more than usual since they have to look at all four, five or six company reports) but the decision they need to make is relatively simple compared to a full Markstrat decision – just four, five or six dollar bids. Furthermore, if a team bids in a way they later regret, it is something they can put behind them: the bidding is separate from the simulation, only resurfacing after it has finished in the form of bid-adjusted cumulative net contribution.