BU303 Simulation Notes: Marketing, Forecasting, and Production (Round 0–4)

Overview of the session

  • Instructor discusses the simulation in depth, focusing on marketing and forecasting, with a promise to expose all pieces of the puzzle so students can understand how they fit together.
  • Structure of the module: marketing decisions, forecasting, production planning, and how rounds progress in the simulation. Four decision rounds (rounds 1–4) must be completed; round 0 is the starting setup used for forecasting for round 1. Promised sample decision walkthroughs for two strategies: cost leader and differentiation.
  • Quiz reminder: there is a quiz on Thursday (no lecture that day). The instructor will post sample decisions to illustrate the two strategies to help students prepare and ask questions in class.

What the marketing department controls

  • Price, promo budget, and sales budget for every product.
  • Each product has a price range around $10; actual price can vary within that window.
  • Customers value price differently; price sensitivity varies by customer segment (e.g., car buyers: price matters more for budget cars than for luxury cars).
  • Promo budget = advertising spend that increases customer awareness. Awareness drives demand but has diminishing returns as spend increases.

Customer awareness and forgetting (story and model)

  • Awareness concept: what fraction of customers are aware a product exists.
  • Real-world analogue: Coca‑Cola is universally known; RC Cola is far less known. Advertising increases awareness but not everyone never forgets.
  • Forgetting over time: in the simulation, people forget after time (e.g., New Year’s hangover analogy).
    • If awareness at year end is $A$, after the forgetting event (one year later), awareness becomes roughly
      Aextnext=frac23A.A_{ ext{next}} = frac{2}{3} A.
      (Because about one third of customers forget each year; the example uses 77% → 52% after forgetting, since $77 imes (1 - frac{1}{3})
      eq 52$ exactly but the intended operationalization is that awareness is multiplied by 2/3 when not refreshed.)
  • Advertising effect over time: the slope of the awareness vs. advertising spend curve changes over time (an S-shaped progression of reach as spend increases).
    • Example trend described: spending $500k buys ~7% awareness; spending $1M buys ~22%; spending $1.5M buys ~34%; $2M buys ~43%; $2.5M buys ~47% awareness.
    • Practical takeaway: best practice is to spend roughly between $1.5M and $2.5M per product per round to refresh awareness without over-spending on the same eyeballs.
  • Accessibility vs. awareness (pharma analogy): awareness is how many know you exist; accessibility is how easily you reach customers (e.g., ease of buying or obtaining the product). Advertising drives awareness; accessibility drives ease of purchase or distribution.

Forecasting and production coupling (high-level workflow)

  • Forecasting vs. production: forecast informs how much to produce; inventory on hand buffers demand/production imprecision.
  • The Capstone Courier (the “newspaper” of the sensor industry) reports last year results and helps interpret market dynamics.
  • Key pages/metrics in the Capstone Courier:
    • Financial statistics: emergency loan status, profit, contribution margin.
    • Market share (overall for the entire segment, not per product).
    • Stock price and bonds on the books (initially three bonds; more can be issued later).
    • Financial statements: cash flow, balance sheet, income statement.
    • Inventory line (on the balance sheet) is crucial – must stay in a healthy range; it changes with rounds.

- Production analysis: capacity by factory and per product; five factories and five products in the base setup.

Factory capacity and production decisions

  • Five factories at game start: Able, Acre, Atom, Performance, and Agape.
    • Traditional Able: 1,800,000 units per round (3,600,000 if run 24 hours).
    • Low-end Acre: 1,400,000 per round (2,800,000 if 24 hours).
    • Atom: 900,000.
    • Performance: 600,000.
    • Agape: 600,000.
  • Capacity is the max you can manufacture in a round given the shift plan.
  • Overtime (running around the clock) costs time-and-a-half; may be a good strategic move but increases cost.
  • Plant improvement and equipment budget limit for round 1: up to 32,000,00032{,}000{,}000 (the cap is noted as a constraint).
  • Automation and cost leadership: higher automation reduces labor costs but costs money to implement (e.g., raising automation from 50% to 70% costs about 11,000,00011{,}000{,}000 in the example). Allocation of funds affects strategy feasibility (cost leadership vs differentiation).
  • Differentiation strategy considerations: target higher-value segments (e.g., high end, performance, size) with product improvements and higher price points; tradeoffs with available budget.

Segment structure and market growth (example numbers)

  • There are five products, five segments: traditional, low end, high end, performance, and size.
  • Market data (example figures discussed):
    • Traditional segment last year demand: 7,387,0007{,}387{,}000 units.
    • Low-end segment last year demand: 8,9608{,}960 units (this may reflect a per-year or per-product figure; the transcript is informal here).
    • High-end segment last year demand: around the order of a few thousand units (text shows a number like 1,915; interpreted as a segment size in thousands in the demo).
    • Performance segment last year demand: 1,915,0001{,}915{,}000 (interpreted from the transcript context).
    • Size segment last year demand: (not explicitly given with a clean value in the transcript).
  • Growth rates by segment (illustrative):
    • Traditional growth: 9.1 ext{%} per year.
    • Low end growth: 11.3 ext{%} per year.
    • High end growth: around 16.2 ext{%} per year.
    • Performance growth: 19.2 ext{%} per year.
    • Size growth: not explicitly given in a single clear figure in the transcript.
  • Next-year demand is computed by applying growth rates to last year’s segment demand:
    • For a segment with last-year demand $Ds$ and growth rate $gs$, next-year demand is
      D<em>s,extnext=D</em>simes(1+gs).D<em>{s, ext{next}} = D</em>s imes (1 + g_s).
  • After computing per-segment next-year demand, total market demand is the sum across segments:
    D<em>extmarket,next=extsum</em>sDs,extnext.D<em>{ ext{market, next}} = ext{sum}</em>{s} D_{s, ext{next}}.
  • Example numbers provide a sense of scale (the exact figures in the transcript are noisy, but the method is clear).

Market share per product and forecasting approach

  • At game start, with six competitors in the market, each company starts with a 16.7% market share (because 100/6 = 16.7 ext{%}).
  • Forecasting approach described (a practical, heuristic method):
    • Start with a modest, disciplined improvement in market share year over year (e.g., +5% absolute share) rather than assuming large leaps.
    • Acknowledge that you cannot be the leader in all segments simultaneously; strategies benefit different segments depending on price, age, and technology.
    • For a given product and segment, forecast next-year demand as:
      extForecastSales<em>p,t+1=MS</em>p,t+1imesD<em>s,t+1,ext{ForecastSales}<em>{p,t+1} = MS</em>{p,t+1} imes D<em>{s,t+1}, where $MS{p,t+1}$ is the forecasted market share for product $p$ in that segment and $D_{s,t+1}$ is the segment’s next-year demand.
  • Example strategy discussion (cost leadership vs differentiation):
    • Cost leadership tends to focus on low-end and traditional customers, with price cuts and cost reductions (e.g., automation) to preserve margins while selling more volume.
    • Differentiation targets higher-end segments with improved features and technology, potentially higher prices, but requires more marketing and R&D investment.
    • Budget constraints: you don’t get a fixed, predefined budget like “one million dollars for everything.” You must allocate across products/segments and still ensure the finance tab shows you can pay for everything (or raise funds via bonds or equity).
    • It’s hard to pursue both a high degree of cost leadership and differentiation simultaneously due to limited resources.

How price, age, and buying criteria interact with strategy

  • Buying criteria by segment (illustrative):
    • Low-end customers: price is most important; want price in the low teens to mid-twenties; age preference around two years.
    • Traditional customers: price still matters but technology age around ~2 years (somewhat older than low-end) and they value reliability and cost efficiency.
    • The transcript notes that low-end customers want age around seven years in the game’s model, while traditional customers want around two years; these age targets influence decisions about product refresh cycles.
  • Age (technology freshness) mechanics:
    • Perceived age of a product is a function of time; to age a product (to be older in the buyers’ eyes), you wait a fixed amount of time (e.g., two years) in the game’s model.
    • Example calculation: if your product is 4.6 years old, waiting two more years would bring it to about 6.6 years (in-game approximation) and in the model you approximate this with fractional years (e.g., 40% of a year ≈ several months).
  • Pricing example (to illustrate cost leadership):
    • If competitors price at $21 and you want to appeal to price-sensitive buyers, you might set your price at $20.50 (50 cents cheaper) to gain a share lift while not leaving too much margin on the table.
  • The rhetorical intent: to connect consumer behavior (price sensitivity, age preferences) with strategic choices (where to allocate budget, which products to invest in, and how to position each product in its segment).

Forecasting workflow in practice (step-by-step flavor from the walkthrough)

  • Round 0 setup: you forecast for round 1; you’re shown how to compute per-segment demand for the next year using growth rates.
  • Per-segment demand forecasting:
    • Given last year demand $Ds$ and growth rate $gs$, compute next-year demand as
      D<em>s,t+1=D</em>s,times(1+gs).D<em>{s,t+1} = D</em>{s,t} imes (1 + g_s).
  • Total market forecast for next year:
    D<em>extmarket,t+1=(D</em>exttraditional,t+1+D<em>extlowend,t+1+D</em>exthighend,t+1+D<em>extperformance,t+1+D</em>extsize,t+1).D<em>{ ext{market},t+1} = \bigl(D</em>{ ext{traditional},t+1} + D<em>{ ext{low end},t+1} + D</em>{ ext{high end},t+1} + D<em>{ ext{performance},t+1} + D</em>{ ext{size},t+1}\bigr).
  • Market share forecasting approach (example): start from the previous year’s shares (initially 16.7% each if evenly distributed) and adjust by a modest amount (e.g., +5% in a given year) for the product/segment you’re targeting.
  • Forecast to production: use forecasted demand to set production targets, but also take into account inventory on hand to avoid overproduction or stockouts:
    • Production after adjustment + Inventory on Hand should be >= Forecast.
  • Example process for round-by-round forecasting in Excel (practical tips):
    • Forecast for each segment with a formula like
      D<em>s,t+1=D</em>s,times(1+gs).D<em>{s,t+1} = D</em>{s,t} imes (1 + g_s).
    • Copy-paste across segments while keeping the growth rate reference fixed using absolute references in Excel (e.g., lock the growth rate cell with $ signs so dragging the formula doesn’t shift the rate).
    • Label totals (e.g., insert a row labeled “Total” and merge/center for readability).
  • Round progression:
    • You save decisions after you’ve adjusted marketing, R&D, production, and finance.
    • You can advance to the next round once decisions are saved.
    • You must complete four rounds (rounds 1–4) of decisions; round 0 is the baseline/forecasting setup.
    • If results look off, you can undo decisions (e.g., due to an emergency loan) and adjust.

Practical tips and classroom workflow

  • Excel/Sheets practice: the instructor emphasizes using small Excel sheets to do straightforward calculations and to get comfortable with data manipulation. He plans to share templates but encourages students to build their own to learn the skill.
  • Office hours and interaction: the instructor offers office hours on Thursday for questions.
  • The mental model: the instructor acknowledges the cognitive load of the content and frames it as puzzle pieces that will fit together over time, like assembling a Mona Lisa from individual facial features.

Ethical, philosophical, and real-world implications

  • Complexity and realism: the simulation abstracts real-world market dynamics with simplified, rule-based mechanics (e.g., fixed forgetting rate, fixed growth rates by segment, simplified age as a proxy for technology freshness).
  • Decision-making under constraints: students learn to balance marketing spend, product aging, and production capacity under finite resources, highlighting tradeoffs that mirror real firms’ budgeting processes.
  • Labor vs automation: automation reduces labor costs but requires capital investment. Students confront the ethical and economic implications of automation decisions (job impact vs efficiency gains).
  • Advertising vs consumer autonomy: constant advertising can shape perceptions and awareness; model’s forgetting mechanism mirrors how consumer memory fades without ongoing reinforcement, raising questions about the role of marketing in perpetually maintaining demand.
  • Strategic realism vs. classroom simplification: the “start with 16.7% market share” baseline is a simplified proxy for competition; in the real world, market shares are influenced by brand equity, distribution, product fit, and macro factors beyond price and advertising.

Key equations and formulas (LaTeX)

  • Segment next-year demand (growth applied):
    D<em>s,t+1=D</em>s,times(1+gs)D<em>{s,t+1} = D</em>{s,t} imes (1 + g_s)
  • Total market demand next year:
    D<em>extmarket,t+1=(extTraditional</em>t+1+extLowEnd<em>t+1+extHighEnd</em>t+1+extPerformance<em>t+1+extSize</em>t+1)D<em>{ ext{market},t+1} = \bigl( ext{Traditional}</em>{t+1} + ext{LowEnd}<em>{t+1} + ext{HighEnd}</em>{t+1} + ext{Performance}<em>{t+1} + ext{Size}</em>{t+1}\bigr)
  • Forecasted product sales from forecasted market share:
    extForecastSales<em>p,t+1=MS</em>p,t+1imesDs,t+1ext{ForecastSales}<em>{p,t+1} = MS</em>{p,t+1} imes D_{s,t+1}
  • Awareness decay (no promo):
    A<em>t+1=frac23A</em>tA<em>{t+1} = frac{2}{3} A</em>t
  • Awareness vs. promo (illustrative impact; not a single closed form in the transcript): awareness increases with promo spend but with diminishing returns; example curve points noted:
  • ext{Awareness} ext{(at }$500k$)
    ightarrow ext{7%}
  • ext{Awareness} ext{(at }$1{,}000{,}000$)
    ightarrow ext{22%}
  • ext{Awareness} ext{(at }$1.500{,}000$)
    ightarrow ext{~34%}
  • ext{Awareness} ext{(at }$2{,}000{,}000$)
    ightarrow ext{~43%}
  • ext{Awareness} ext{(at }$2{,}500{,}000$)
    ightarrow ext{~47%}
  • Production capacity (per factory, per shift): constants listed for each factory (e.g., Able 1,800,000 per round; with 24h shift, 3,600,000).
  • Maximum cap to plant improvements for round 1: 32,000,00032{,}000{,}000
  • Automation levels and cost example: increasing automation from 50% to 70% costs about 11,000,00011{,}000{,}000 (illustrative).
  • Market share baseline: starting point per product in a six-competitor market is 16.7 ext{%} (i.e., 100/6 = 16.7 ext{%}).

Quick recap of the core takeaways

  • Marketing decisions are tightly coupled to forecasting: price, promo, and forecast determine production and inventory needs.
  • Awareness decays over time; continuous advertising is needed to refresh demand, with diminishing returns as spend grows.
  • Production capacity and capital spending (automation, plant improvements) are central to strategy; cost leadership relies on cost reductions to enable price competition, while differentiation relies on product improvements to attract higher willingness-to-pay.
  • Forecasting is a structured process: compute segment demand growth, aggregate to market demand, assign market shares, convert to product forecasts, and align with production/inventory constraints.
  • Four rounds of decisions must be completed; the process explicitly encourages deliberate, well-documented decisions rather than quick, impulsive moves.
  • The exercise blends practical spreadsheet skills (Excel/Sheets) with strategic thinking to simulate real-world business planning under resource constraints.

Quick glossary of terms used in this module

  • Forecast: Estimated future demand for a product/segment in the next period.
  • Market share: Portion of the total market demand captured by a product or company.
  • Awareness: Propensity of potential customers to know that a product exists.
  • Accessibility: How easily customers can obtain or purchase the product.
  • Attrition/forgetting: The loss of customer awareness over time without reinforcement.
  • Capacity: The maximum production output a plant/factory can deliver given the shift setup.
  • Overtime: Running a factory beyond normal hours to increase output, at a higher labor cost.
  • Automation: The degree to which production is automated, affecting labor costs.
  • Capex: Capital expenditure used to purchase plant/equipment or automation upgrades.
  • Inventory on hand: Stock kept to buffer forecasting errors and demand variability.
  • Emergency loan/bonds: Financing tools to fund operations when cash flow is tight.