For my senior marketing capstone, four of us ran competing firms through eight periods of a market simulation. I was CEO of Team M — the team handed the smallest starting budget of the group — and over those eight periods we built it into the dominant position in the industry by a wide margin. This is a look back at how we did it, told entirely through the simulation's own data.
MarkStrat is the strategy simulation used in a lot of MBA and capstone marketing programs. Four student-run firms compete in the same simulated electronics market (the Sonites), and every period each team makes a stack of granular decisions — R&D feature targets, pricing, production volumes, ad budgets and messaging, sales-force staffing. Once everyone locks in, the market resolves all four teams' moves at once and publishes the results.
Our grade was tied to the Share Price Index, which is a relative measure — because it's finite, one team running away with it can mathematically keep everyone else from an A. I was CEO of Team M (Moose).
Sonite demand grew from 746,000 units in Period 0 to 2.4 million by Period 8. Then, about halfway through, a second product category — the Vodites — opened up from nothing and climbed to 900,000 units of its own.
Growth that fast doesn't leave much room to coast. Since SPI is relative, a team that only holds its ground is actually falling behind, so standing still was never really an option.
The starting budgets weren't equal, which the professor framed as a nod to real markets — not every company gets to begin with the same resources. Team M opened with $10.9M in earnings, the lowest of the four and 22% behind Scorpion, along with the smallest budget going into the next period.
In MarkStrat, your budget each period is tied to how you performed the period before. A weak start compounds — but so does a strong one, which meant the first couple of periods would basically decide who got to play offense for the rest of the game.
By Period 8 our marketing budget was $25.3M, the largest in the industry, while Rumble — who out-earned us at the start — was working with $9.0M.
Just about every chart that follows traces back to this one. Budget is the compounding variable, and it's what turns a few good periods into a lead nobody can catch.
MarkStrat sells market research the same way the real world does — piece by piece — and it's expensive. Consumer surveys, panel data, semantic scales, competitive ad estimates, market forecasts, plus advertising and commercial-team experiments: ten separate studies per market, every period.
I bought all of them, every period, starting from the very first turn — $6.1M in total, and most of it spent when we had the least to spare. Each bundle came out to around 40 pages of tables. From what I could tell, most teams never really read them. That was our edge.
The semantic-scale studies give you each segment's ideal product on a 1–7 scale, one attribute at a time. So if the Savers' ideal price came back at 2.0, we'd set our marketing to 1.9 — a small, deliberate undercut, betting the other teams would either sit right on the published number or ignore it.
The experiment studies were just as useful: they told us the marginal return on the next dollar of advertising, or the next salesperson, before we actually spent it. We built our own Excel calculators to turn those report values straight into R&D targets.
We plotted every brand on the two attributes the semantic scales said mattered most to buyers — price and processing power. Period 1 read cleanly: our budget product MOST was sitting nicely near the Savers' ideal, but MOVE was stranded out in the middle of nowhere, and Turtle already had the premium corner locked up.
So we made the call to R&D MOVE toward the Shoppers — a segment nobody was really serving, and one the forecasts said was about to grow.
R&D takes a full period to finish and another before it reaches shelves, so any repositioning move is really a bet on where the market is heading, not where it sits today. While MOVE's project was in the pipeline, we held our pricing steady on MOST and kept buying the studies that confirmed the Shoppers were still on their way.
The repositioning paid off: MOVE was sitting right on the Shopper ideal, and that proximity turned into a purchase-intent lead of over 70% in the segment. MOST, meanwhile, was in a scrap with Scorpion's SOLO and SOFT over the Savers — a fight we were losing on pure positioning but winning on awareness and budget.
One rule we never broke was to keep our own products away from each other — when two of your brands overlap, you're paying to compete with yourself. By the final period MOVE had the Shopper territory almost to itself, MOST held a defensible spot with the Savers, and none of our brands were stepping on another.
The competitive-advertising studies let us keep an eye on what every rival was putting up. As our budget compounded, our own ad spend climbed to $15.9M by Period 8 — the most in the industry, about 2.5× what Rumble could manage. Awareness followed spend almost mechanically: the more we put behind a brand, the more people recalled it.
Awareness across our brands remained steady across periods; it was a necessary condition, not a sufficient one.. Turtle's Vodite brand was the proof — it held an awareness lead over ours the entire time we were taking its market out from under it.
Purchase intent — the share of buyers who name your brand first — comes down to how closely your product and your messaging match a segment's ideal. That's exactly what all the semantic-scale discipline was buying us: by the final period, MOVE's intent among Shoppers hit 70%.
From there, the ad experiments told us how hard to push — we scaled spend wherever the measured marginal return was highest.
Our planned production kept landing short — we'd order 1,200K units and get back 960K. And in the fast-growth periods we sold everything we made: MOVE finished Periods 3 through 6 with zero inventory, which looks efficient on paper but is really a hidden tally of sales we never got to make.
Selling out sounds like a good thing, but every stockout was a sale we handed straight to a competitor.
Holding cost in MarkStrat is only 8% of transfer cost — basically pennies once a product has healthy margins. Once we actually ran that math, the whole policy flipped: deliberately overshoot production and treat the leftover inventory as cheap insurance against selling out.
Even our worst case backed this up. When a MOST refresh in Period 8 made 357K older units obsolete, we ate a $3.2M disposal loss — not optimal, but nowhere near what repeated MOVE stockouts would have cost us at those margins.
We tracked net contribution per brand in real time, and when a brand's contribution started to flatten, we moved resources toward it — but only after the experiment studies confirmed another dollar would actually move the needle.
MOVE's line is basically the whole strategy in one curve: $3.9M up to $87.5M over the course of the game.
A commercial-team experiment in Period 3 flagged a lot of unmet revenue sitting in MOVE's channels. We expanded the sales force to match, and MOVE's contribution jumped 4.5× in a single period.
And keep an eye on the pink line that shows up at Period 6: MEME went from nothing to $128.8M in contribution within three periods — more than half the company's profit by the end.
The Vodite market launched with three segments — Innovators, Adopters, and Followers — and a forecast I trusted: the whole market was going to shift hard toward the Followers, who ended up growing from 7K to 677K units by the end.
Entering meant two full periods of cost — R&D, then launch — and money we simply didn't have yet. So we waited, saved, and kept buying the Vodite reports while the other teams paid the price of going in first.
Rumble pioneered the market with REF in Period 3, but positioned it poorly and got knocked down to a 29% share the moment Turtle's TEPETE landed near the Followers' ideal in Period 4. By Period 5, TEPETE was holding 100% purchase intent — for a moment, it was the only real option in the market.
Letting two rivals spend their own money mapping the market for us was the best research we never had to pay for.
We didn't want to launch a half-right product and fix it later — on a tight budget, an extra R&D cycle after launch would have cost us two more periods of income. So MEME came out needing no correction, positioned right between the Adopter and Follower ideals just as the majority was tipping from one to the other.
First period on the market: 72% unit share. Our budget doubled the period after.
TEPETE's purchase intent fell from 100% to 22% the same period MEME arrived — even though TEPETE was still beating MEME on brand awareness. People knew Turtle's product; they just wanted ours instead.
A follow-up R&D project then slid MEME right onto the Followers' ideal as that segment became the whole market, and we finished at 84% unit share. First-mover advantage matters a lot less than showing up positioned correctly.
Team M's earnings before taxes grew 20.8× over the game, against 1.7× for Rumble, 2.7× for Scorpion, and 3.4× for Turtle. From the smallest starting position in the industry, we finished with more profit than the other three teams put together — twice over.
Plot our EBT on a log scale and it comes out almost perfectly straight. An exponential fit gives an R² of 0.97 — about 47% compound growth every period, held up across eight periods of competitors actively pushing back. For comparison, Rumble's fit is an R² of 0.06.
A Pearson correlation coefficient that high isn't luck. Steady exponential growth under real competition is what a strategy looks like when it's actually working — scattered results are what guessing looks like.
Three times Turtle, four times Scorpion, six times Rumble. On the one metric the whole course was graded on — a metric that's zero-sum by design — we didn't just clear the bar for an A. We set the curve everyone else was measured against.
Looking back across the eight periods, the playbook splits pretty cleanly into two halves: a handful of rules we never broke, and a set of habits we re-tuned every single period.