# Opportunity Assessment Concept Card: [Opportunity, venture or course case]

> A concept card for screening a new venture or new-business opportunity before money and years are committed:
> market sizing as TAM, SAM and SOM computed both top-down and bottom-up, problem severity measured in the customer's
> money, competitive alternatives including "do nothing", unit economics, a weighted scoring model with sensitivity
> checks, and written kill criteria that say in advance what evidence would stop the project. It is for MBA students in
> the entrepreneurship and strategy courses, founders and corporate venture teams, and investors who screen deals.
> Worked example A assesses a fictional Corvane Robotics product line and is fully solved; Worked example B is half
> solved and Exercise C is left for you, with a separate answer key. Source basis: Blank (2005) on customer discovery,
> McGrath and MacMillan (1995) on discovery-driven planning, Ries (2011) on testing assumptions, and Gupta, Lehmann and
> Stuart (2004) on customer value, written up in original words. Replace every [bracketed] value and every row marked
> *Example*. This is a learning aid, not investment advice.

<!--
Guidance: How to use this template
- Read "The idea in plain language" first. The single most useful habit is to size a market twice, top-down and
  bottom-up, and to explain the gap.
- Rebuild Worked example A in a spreadsheet with the Excel formulas shown, then replace every assumption with your own
  opportunity and keep the same structure.
- Fill every [ ] in Worked example B before you look at the answer key. Do Exercise C on paper first.
- Write the kill criteria before you collect the evidence. Criteria written afterwards tend to fit the result.
- All companies and numbers are fictional and were computed in Python; the calculation script is kept with the library
  sources so a reviewer can re-run it.
- Related templates: mba/mba-entrepreneurship-guide, mba/mba-venture-finance-cap-table-card,
  mba/mba-customer-lifetime-value-card, mba/mba-stp-positioning-card, mba/mba-pricing-card,
  mba/mba-five-forces-card, mba/mba-competitive-advantage-card, mba/mba-market-sizing-drills,
  mba/mba-npv-irr-card, mba/mba-decision-trees-card, mba/mba-error-log, finance/unit-economics-analysis,
  startup/customer-discovery, startup/pitch-outline, startup/pivot-memo, business/lean-canvas,
  business/feasibility-study, business/market-research-report, product/competitive-analysis.
-->

## Document control

<!-- Guidance: Record the evidence base (interviews, data sources, dates) so that a reviewer can see how fresh and how strong it is. -->

| Field | Value |
|---|---|
| Opportunity | [Product, segment and geography in one line] |
| Assessed by | [Your name or team] |
| Evidence base | [Number of customer interviews, pilots, data sources, dates] |
| Decision this informs | [Go to pilot / build MVP / raise seed / stop] |
| Reviewed by | [Mentor, tutor or investment committee] |
| Version | [0.1] |
| Last updated | YYYY-MM-DD |

## Contents

- [When to use it](#when-to-use-it)
- [The idea in plain language](#the-idea-in-plain-language)
- [Formulas and notation](#formulas-and-notation)
- [Assumptions and limits](#assumptions-and-limits)
- [How managers and investors use it](#how-managers-and-investors-use-it)
- [Worked example A (fully solved): Corvane Frost cold-storage picking](#worked-example-a-fully-solved-corvane-frost-cold-storage-picking)
- [Kill criteria](#kill-criteria)
- [Worked example B (half solved): Northfield gear rental subscription](#worked-example-b-half-solved-northfield-gear-rental-subscription)
- [Exercise C (unsolved): Tallow Creek scheduling software](#exercise-c-unsolved-tallow-creek-scheduling-software)
- [Process diagram](#process-diagram)
- [Connections to other cards](#connections-to-other-cards)
- [Common mistakes and misconceptions](#common-mistakes-and-misconceptions)
- [Self-test questions](#self-test-questions)
- [My notes and next steps](#my-notes-and-next-steps)
- [Sources and further reading](#sources-and-further-reading)
- [Answer key](#answer-key)
- [Revision history](#revision-history)

## When to use it

<!-- Guidance: Use the card at the screening stage. Once the opportunity passes, move to a business plan and a financial model. -->

**One-sentence rule:** an opportunity is worth pursuing when a specific customer has a costly problem, will pay
enough to solve it with your offer rather than the alternatives, can be reached at an acquisition cost the customer's
lifetime value repays several times, and sits in a market large enough to matter; write down what would prove each of
these wrong and test the riskiest first.

| Use this card when | Use another tool when |
|---|---|
| Choosing between several venture ideas or beachhead segments | Writing the full plan for investors (business/business-plan, startup/pitch-outline) |
| Deciding whether to fund a pilot or MVP | Valuing a business with a track record (mba/mba-dcf-valuation-card) |
| Screening a deal or a corporate venture proposal | Positioning an existing product (mba/mba-stp-positioning-card) |
| Preparing for a case on market entry or new business | Analysing industry profitability in depth (mba/mba-five-forces-card) |

Shane and Venkataraman (2000) defined the field of entrepreneurship around the discovery, evaluation and
exploitation of opportunities. This card is a structured version of the "evaluation" step.

## The idea in plain language

<!-- Guidance: Be able to say what each block of the assessment answers in one sentence. -->

The assessment has five blocks. Each answers one question and each has a typical way of fooling yourself.

- **Market size (TAM, SAM, SOM).** Total addressable market (TAM) is the annual revenue if every possible customer
  bought the category. Serviceable available market (SAM) is the part you can reach with your product, channels and
  geography today. Serviceable obtainable market (SOM) is what you can realistically win within a stated horizon. A
  **top-down** estimate starts from a published or estimated total and applies shares; a **bottom-up** estimate
  multiplies the number of customers by units per customer by price. Bottom-up is more credible to investors because
  every number can be tested in interviews. When the two differ by more than a factor of about two, one of them has
  a wrong assumption.
- **Problem severity.** How much does the problem cost the customer, how often, and how urgently? Measure it in the
  customer's own money and time. A problem worth solving costs more than your price by a clear margin.
- **Competitive alternatives.** The real competitor is often the current workaround: spreadsheets, manual labour,
  an incumbent supplier, or doing nothing. Price and differentiation are judged against these, not only against
  start-ups with similar products.
- **Unit economics.** For one customer: gross profit per period, how long they stay, what it costs to acquire them.
  Lifetime value (LTV) over customer acquisition cost (CAC) and the CAC payback period summarise it
  (Gupta, Lehmann and Stuart 2004; mba/mba-customer-lifetime-value-card; finance/unit-economics-analysis).
- **Scoring and kill criteria.** A weighted score makes the comparison of options explicit and debatable. Kill
  criteria, written in advance, turn the riskiest assumptions into tests with thresholds (McGrath and MacMillan 1995;
  Ries 2011).

## Formulas and notation

<!-- Guidance: Write each market-size formula with units on every term. Most sizing errors are unit errors (per site or per robot, per month or per year). -->

$$
TAM_{bottom\text{-}up} = N \times q \times p \qquad TAM_{top\text{-}down} = M \times s_1 \times s_2 \times \dots
$$

$$
SAM = TAM \times f_{reach} \qquad SOM_T = \sum_{t=1}^{T} \text{new customers}_t \times q \times p \quad \text{(no churn) or} \quad SAM \times \pi_T
$$

$$
LTV = m \cdot \frac{r}{1 + i - r} \qquad \text{LTV to CAC} = \frac{LTV}{CAC} \qquad \text{CAC payback (months)} = \frac{CAC}{m/12}
$$

$$
S_k = \sum_{c} w_c \, x_{kc} \qquad \sum_{c} w_c = 1
$$

| Symbol | Meaning | Unit |
|---|---|---|
| $N$ | Number of potential customers (sites, firms, households) | Count |
| $q$ | Units per customer (robots per site, seats per firm) | Count |
| $p$ | Price per unit per year | Currency per year |
| $M$ | Published or estimated total spend in a broader category | Currency per year |
| $s_1, s_2$ | Shares of that spend that fall in your category and segment | Ratio |
| $f_{reach}$ | Share of TAM your product, channels and geography can serve | Ratio |
| $\pi_T$ | Penetration of SAM achievable by year $T$ | Ratio |
| $m$ | Annual gross profit per customer | Currency per year |
| $r$ | Annual retention rate (1 minus churn) | Ratio |
| $i$ | Annual discount rate | Rate |
| $CAC$ | Fully loaded cost to acquire one customer | Currency |
| $w_c$, $x_{kc}$ | Weight of criterion $c$, score of option $k$ on criterion $c$ (1 to 5) | Ratio, points |
| $S_k$ | Weighted score of option $k$ | Points |

The LTV formula assumes constant margin and retention and that each year's margin is received at year end by the
customers retained that year (Gupta, Lehmann and Stuart 2004). Other conventions (margin at the start of the year,
monthly periods) give different numbers; state yours.

## Assumptions and limits

<!-- Guidance: List your assumptions in a table with their source and the test that would confirm them. -->

1. **Market sizes are hypotheses, not facts.** Published market reports use different definitions; cite the source and
   year, and check the definition before applying shares.
2. **SOM is a plan, not a forecast.** It is only as good as the sales capacity, cycle length and win rate behind it.
3. **Unit economics at small scale are noisy.** Early CAC is often too low (warm leads, founder selling) or too high
   (no brand, no referrals). Treat the first cohort with caution.
4. **Gross margin must match the business model.** In a robots-as-a-service model the hardware sits on your balance
   sheet: gross margin after depreciation can look healthy while cash is consumed by fleet capital expenditure.
5. **Scores hide judgement.** A weighted score is a structured opinion. Its value is in making the weights and scores
   explicit and in testing whether the ranking survives other reasonable weights.
6. **Severity is not willingness to pay.** Customers may describe a painful problem and still not buy. Paid pilots and
   letters of intent are stronger evidence than interview enthusiasm (Blank 2005).
7. **The assessment is a snapshot.** Re-run it when evidence arrives; that is the point of discovery-driven planning.

## How managers and investors use it

<!-- Guidance: Note which version of this card your audience expects: an investor wants the bottom-up market and unit economics; a corporate committee wants the strategic fit and the kill criteria. -->

| User | What they focus on | Typical threshold or question (illustrative, not a rule) |
|---|---|---|
| Founder choosing a beachhead | Severity, reachability, speed to first revenue | Can we name the first 20 customers? |
| Seed investor | Bottom-up market, evidence of pull, team fit | Could this be a large company if it works? |
| Growth investor | Unit economics by cohort, CAC payback, net revenue retention | Is growth efficient? |
| Corporate venture committee | Strategic fit, right to win, kill criteria and staging | What do we learn for how much money by when? |
| MBA case discussion | Logic and consistency of the numbers, the recommendation and its risks | Would you invest, and on what condition? |

Commonly quoted practitioner rules of thumb, such as an LTV to CAC above about 3 or a CAC payback under about 12 to 18
months for business software, are conventions, not laws. Use them as prompts for questions, and compare with your own
cost of capital and cash position.

## Worked example A (fully solved): Corvane Frost cold-storage picking

<!-- Guidance: Rebuild each table in a spreadsheet. Then change one assumption at a time and see which conclusion moves. -->

**Situation (fictional).** Corvane Robotics, a growth-stage warehouse-robotics company, considers three new product
lines: **Frost**, a picking robot for cold-storage (refrigerated and frozen) warehouses; a **retail micro-fulfilment**
system for grocery back rooms; and a **parcel sortation retrofit** for existing hubs. Frost would be offered as
robots-as-a-service (RaaS) at USD 42,000 per robot per year; a typical site uses 6 robots. All figures are fictional
assumptions for the exercise.

### A1. Market sizing, top-down and bottom-up

| *Example* item | Assumption | Result | Excel formula |
|---|---|---|---|
| Top-down TAM | Global warehouse automation spend USD 28.0bn a year, cold-storage share 13% | USD 3.640bn | `=28*13%` |
| Bottom-up TAM | 15,000 cold-storage sites worldwide x 6 robots x USD 42,000 | USD 3.780bn | `=15000*6*42000/1E9` |
| Gap, bottom-up vs top-down | | +3.8% | `=3.78/3.64-1` |
| Top-down SAM | TAM x North America and Western Europe 45% x mid-size sites 62% | USD 1.016bn | `=3.64*45%*62%` |
| Bottom-up SAM | 4,200 mid-size sites in those regions x 6 x USD 42,000 | USD 1.058bn | `=4200*6*42000/1E9` |
| SOM (penetration) | 3% of SAM sites by year 5 = 126 sites | USD 31.75m ARR | `=ROUND(4200*3%,0)*6*42000/1E6` |
| SOM (sales capacity) | Account executives 2, 4, 6, 8, 8; 4 new sites each a year; 112 sites by year 5 | USD 28.22m ARR | `=SUM(2,4,6,8,8)*4*6*42000/1E6` |

The two TAM estimates agree within 4% and the two SAM estimates within 5%, which gives some confidence in the order of
magnitude. The sales-capacity SOM (112 sites, 2.67% of SAM sites) is lower than the penetration SOM and is the one to
plan on, because it is tied to hiring. Annual recurring revenue at each year end would be USD 2.02m, 6.05m, 12.10m,
20.16m and 28.22m (cumulative sites 8, 24, 48, 80 and 112).

### A2. Problem severity, in the customer's money

Interviews with cold-storage operators (fictional) point to labour availability in sub-zero conditions, high staff
turnover and injury risk. Per site and year:

| *Example* item | Result | Excel formula |
|---|---|---|
| Labour saving: 6 full-time equivalents at USD 58,000 loaded cost | USD 348,000 | `=6*58000` |
| Turnover and injury cost avoided | USD 45,000 | Estimate from interviews |
| Total customer benefit | USD 393,000 | `=348000+45000` |
| Frost fee: 6 robots x USD 42,000 | USD 252,000 | `=6*42000` |
| Net saving to the customer | USD 141,000 | `=393000-252000` |
| Benefit to fee ratio | 1.56 | `=393000/252000` |
| Customer payback on a USD 60,000 integration cost | 5.1 months | `=60000/(141000/12)` |

A benefit of 1.56 times the fee is attractive but not overwhelming; the case depends on the robots really replacing
one full-time equivalent each across shifts, which is the assumption to test in pilots.

### A3. Competitive alternatives

| *Example* alternative | What the customer does today | Frost's advantage | Frost's weakness |
|---|---|---|---|
| Do nothing (manual picking) | Hire, train and replace staff, pay cold-store premiums | Reliability in staff shortages, safety | Upfront change, integration effort |
| Fixed automation (automated storage and retrieval) | Large capital project | No capital outlay, faster to deploy, scalable | Lower throughput at very high volumes |
| General-purpose picking robots | Robots not rated for sub-zero work | Built for cold, sealed components | Smaller installed base, less proven |
| Temporary labour agencies | Flexible staff at a premium | Predictable cost | Agencies scale down more easily |

### A4. Unit economics per site

| *Example* item | Result | Excel formula |
|---|---|---|
| Annual revenue per site | USD 252,000 | `=6*42000` |
| Gross margin (after robot depreciation, field service, cloud) | 55% | Assumption |
| Annual gross profit $m$ | USD 138,600 | `=252000*55%` |
| CAC per site (sales, pilot, integration support) | USD 180,000 | Assumption |
| Retention $r$ / discount rate $i$ | 95% / 12% | Assumptions |
| LTV | USD 774,529 | `=138600*95%/(1+12%-95%)` |
| LTV to CAC | 4.30 | `=774529/180000` |
| CAC payback | 15.58 months | `=180000/(138600/12)` |
| Undiscounted LTV (margin / churn), for comparison | USD 2,772,000 | `=138600/(1-95%)` |
| Fleet capital per new site (6 robots at USD 55,000) | USD 330,000 | `=6*55000` |

The discounted LTV to CAC of 4.30 clears a 3x convention; the undiscounted figure is 3.6 times larger, which shows
how much a lazy LTV formula flatters a long-lived customer. The fleet capital of USD 330,000 per site is not in CAC or
gross margin but must be financed before revenue arrives: Frost is capital-intensive, a point for
mba/mba-venture-finance-cap-table-card and for any lease or debt financing of the fleet.

### A5. Weighted scoring of the three options

Scores run from 1 (poor) to 5 (strong); risk is scored so that 5 means low risk.

| *Example* criterion | Weight | Frost | Micro-fulfilment | Sortation retrofit |
|---|---|---|---|---|
| Market size and growth | 20% | 4 | 5 | 3 |
| Problem severity | 20% | 5 | 3 | 3 |
| Differentiation against alternatives | 15% | 4 | 2 | 3 |
| Unit economics | 20% | 4 | 3 | 4 |
| Right to win (team, technology, channel) | 15% | 4 | 2 | 3 |
| Execution and regulatory risk (5 = low) | 10% | 3 | 3 | 4 |
| **Weighted score** | 100% | **4.10** | **3.10** | **3.30** |

Excel: with weights in B2:B7 and Frost's scores in C2:C7, `=SUMPRODUCT($B$2:$B$7, C2:C7)` gives 4.10.

**Sensitivity.** With equal weights the scores are 4.00, 3.00 and 3.33; with a market-heavy set (market 35%, problem
10%, differentiation 15%, unit economics 20%, right to win 10%, risk 10%) they are 4.00, 3.45 and 3.30. Frost ranks
first under all three; micro-fulfilment overtakes the sortation retrofit only when market size dominates, which says
the micro-fulfilment case rests on market size alone.

```vega-lite
{
  "$schema": "https://vega.github.io/schema/vega-lite/v5.json",
  "description": "Weighted score contributions by criterion for three Corvane options (fictional)",
  "width": 420,
  "height": 180,
  "data": {"values": [
    {"option": "Frost", "criterion": "Market", "points": 0.8},
    {"option": "Frost", "criterion": "Problem", "points": 1.0},
    {"option": "Frost", "criterion": "Differentiation", "points": 0.6},
    {"option": "Frost", "criterion": "Unit economics", "points": 0.8},
    {"option": "Frost", "criterion": "Right to win", "points": 0.6},
    {"option": "Frost", "criterion": "Risk", "points": 0.3},
    {"option": "Micro-fulfilment", "criterion": "Market", "points": 1.0},
    {"option": "Micro-fulfilment", "criterion": "Problem", "points": 0.6},
    {"option": "Micro-fulfilment", "criterion": "Differentiation", "points": 0.3},
    {"option": "Micro-fulfilment", "criterion": "Unit economics", "points": 0.6},
    {"option": "Micro-fulfilment", "criterion": "Right to win", "points": 0.3},
    {"option": "Micro-fulfilment", "criterion": "Risk", "points": 0.3},
    {"option": "Sortation retrofit", "criterion": "Market", "points": 0.6},
    {"option": "Sortation retrofit", "criterion": "Problem", "points": 0.6},
    {"option": "Sortation retrofit", "criterion": "Differentiation", "points": 0.45},
    {"option": "Sortation retrofit", "criterion": "Unit economics", "points": 0.8},
    {"option": "Sortation retrofit", "criterion": "Right to win", "points": 0.45},
    {"option": "Sortation retrofit", "criterion": "Risk", "points": 0.4}
  ]},
  "mark": "bar",
  "encoding": {
    "y": {"field": "option", "type": "nominal", "title": null},
    "x": {"field": "points", "type": "quantitative", "aggregate": "sum", "title": "Weighted score (max 5)"},
    "color": {"field": "criterion", "type": "nominal", "title": "Criterion"}
  }
}
```

**Conclusion.** Frost scores highest (4.10) and the ranking is robust to the weights. The recommendation is not "build
Frost" but "fund a staged test of Frost against the kill criteria below", because the scores for severity and unit
economics rest on interviews and assumptions, not on paid deployments.

## Kill criteria

<!-- Guidance: Write each criterion as a measurable test with a threshold, a date and an owner. Decide now what you will do if it fails. -->

| *Example* ID | Riskiest assumption | Test | Kill threshold | By | Owner |
|---|---|---|---|---|---|
| KC-01 | Cold-storage operators rank labour as a top-3 problem | 20 structured interviews (startup/customer-discovery) | Fewer than 10 of 20 rank it top-3 | Month 2 | [@owner] |
| KC-02 | Customers will pay for a pilot | Paid pilot offers to 8 qualified operators | Fewer than 2 paid pilots signed | Month 4 | [@owner] |
| KC-03 | One robot replaces about one full-time equivalent across shifts | Two pilot sites, 8 weeks | Below 0.8 FTE per robot after tuning | Month 7 | [@owner] |
| KC-04 | Unit economics hold at pilot prices | Pilot cost and price data | LTV to CAC below 3.0 on pilot data | Month 8 | [@owner] |
| KC-05 | The market is large enough to matter | Bottom-up count of mid-size cold-storage sites verified with two data sources | SOM by year 5 below USD 20m ARR | Month 3 | [@owner] |

A failed criterion leads to one of three decisions agreed in advance: **stop**, **pivot** (change segment, product or
model; startup/pivot-memo) or **persevere** with a documented reason. Ries (2011) calls the underlying beliefs
"leap-of-faith assumptions"; McGrath and MacMillan (1995) recommend a reverse income statement and an assumptions
checklist updated at each milestone.

## Worked example B (half solved): Northfield gear rental subscription

<!-- Guidance: Fill every [ ] before looking at the answer key. Write the Excel formula next to each number. -->

**Situation (fictional).** Northfield Outdoor Co., a mid-size outdoor-gear retailer with stores and an online shop,
considers a gear rental subscription at USD 29 a month in the three metro areas where it has stores.

| *Example* item | Assumption | Result | Excel formula |
|---|---|---|---|
| Households in the three metro areas | 2,400,000 | | |
| Target households (camp or hike 3+ times a year) | 18% | 432,000 | `=2400000*18%` |
| SAM households (within 30 minutes of a store) | 40% | [ ] | `=432000*40%` |
| SOM subscribers (2.5% adoption by year 3) | | [ ] | `=B4*2.5%` |
| Annual fee | USD 29 x 12 | USD 348 | `=29*12` |
| SOM revenue | | [ ] | `=B5*348` |
| Gross margin / annual churn / CAC / discount rate | 42% / 35% / USD 95 / 10% | | |
| Annual gross profit per subscriber | | USD 146.16 | `=348*42%` |
| LTV | | [ ] | `=146.16*65%/(1+10%-65%)` |
| LTV to CAC | | [ ] | `=B10/95` |
| CAC payback | | 7.80 months | `=95/(146.16/12)` |

Questions: (1) Fill the gaps. (2) Score the option with the weights of Worked example A, using scores of 3, 2, 3, 2,
5, 4; what is the weighted score? (3) Which single assumption would you test first, and with what kill threshold?
(4) Why does a fast CAC payback not rescue a low LTV to CAC here?

## Exercise C (unsolved): Tallow Creek scheduling software

<!-- Guidance: Do the whole assessment, including kill criteria, before checking the numbers in the answer key. -->

**Situation (fictional).** Tallow Creek Medical Center, a regional hospital, built an operating-theatre scheduling
tool that cut idle theatre time. A spin-out would sell it to other hospitals as software at USD 85,000 a year per
hospital. Assume 6,100 hospitals in the target regions, of which 55% are regional hospitals of 100 to 400 beds (the
segment the product fits). Sales capacity: 1, 2, 3, 4 and 4 account executives in years 1 to 5, each winning 5
hospitals a year, no churn within the five years. Gross margin 78%, annual retention 92%, CAC USD 120,000, discount
rate 12%.

1. Compute TAM and SAM (in hospitals and in USD) and the year-5 SOM in ARR.
2. What share of SAM hospitals does the year-5 SOM represent?
3. Compute annual gross profit per hospital, LTV, LTV to CAC and CAC payback in months.
4. Score the opportunity with the weights of Worked example A, using scores of 3, 4, 3, 4, 3, 2.
5. Write three kill criteria. At least one must address the long hospital buying cycle and one the risk that the
   tool's results depend on Tallow Creek's own processes.

## Process diagram

<!-- Guidance: Follow the flow in order. The loop back from the kill gate is where most of the learning happens. -->

The flow matches the sections of Worked example A: frame the customer and problem, size the market twice, check
alternatives and unit economics, score, then test against the kill criteria before scaling.

```mermaid
flowchart TD
    A["Define customer, problem and offer in one line"] --> B["Size market top-down and bottom-up"]
    B --> C{"Estimates within about 2x?"}
    C -->|No| B2["Find the wrong assumption"] --> B
    C -->|Yes| D["Quantify problem severity in customer money"]
    D --> E["Map competitive alternatives incl. do nothing"]
    E --> F["Unit economics: LTV, CAC, payback, capital needs"]
    F --> G["Weighted score and sensitivity to weights"]
    G --> H["Write kill criteria for riskiest assumptions"]
    H --> I["Run cheapest test first"]
    I --> J{"Kill criterion met?"}
    J -->|Fails| K{"Stop, pivot or persevere?"}
    K -->|Pivot| A
    K -->|Stop| L["Record learning and close"]
    J -->|Passes| M["Fund next stage and update the card"]
    M --> I
```

## Connections to other cards

<!-- Guidance: Each link answers a question this card leaves open. -->

- **Customer value in depth:** mba/mba-customer-lifetime-value-card (cohorts, retention curves, margin definitions)
  and finance/unit-economics-analysis (the working document).
- **Segment and positioning:** mba/mba-stp-positioning-card; price-setting: mba/mba-pricing-card.
- **Industry attractiveness and advantage:** mba/mba-five-forces-card and mba/mba-competitive-advantage-card.
- **Funding the opportunity:** mba/mba-venture-finance-cap-table-card (how much to raise, on what terms).
- **Staging and flexibility:** mba/mba-decision-trees-card and mba/mba-options-real-options-card value the option to
  stop after a pilot.
- **Market sizing practice:** mba/mba-market-sizing-drills.
- **Working templates:** startup/customer-discovery for interviews, business/lean-canvas for a one-page model,
  business/feasibility-study and business/market-research-report for a fuller assessment, product/competitive-analysis
  for the alternatives.

## Common mistakes and misconceptions

<!-- Guidance: Read the list before you present an assessment. Tick the ones you have made and log them in mba/mba-error-log. -->

1. **"If we get 1% of a huge market."** A top-down share of a big number is not a plan. Build SOM from customers,
   capacity and win rates.
2. **Calling TAM the whole economic category.** TAM must be the spend your category of solution could capture, not the
   customer's entire budget.
3. **Mixing units.** Per robot and per site, per month and per year: put units on every column.
4. **Measuring severity by enthusiasm.** Count paid pilots, letters of intent and budget owners, not compliments.
5. **Ignoring "do nothing".** The status quo is free of switching costs and is the most common winner.
6. **Using undiscounted LTV.** Margin divided by churn ignores the time value of money and flatters long-lived customers.
7. **Leaving capital out of unit economics.** Hardware, inventory or fleet investment can make a high-margin model
   cash-hungry.
8. **Fixing weights after scoring.** Weights chosen once the scores are known can produce any ranking. Fix them first
   and run a sensitivity check.
9. **Kill criteria without thresholds or dates.** "We will see how pilots go" cannot fail, so it cannot inform a
   decision.
10. **Treating the assessment as a one-off.** The card should be re-run after each test.

## Self-test questions

<!-- Guidance: Answer in writing, from memory, before scrolling to the answer key. Repeat after two days and after a week. -->

1. Define TAM, SAM and SOM in one sentence each.
2. Why do investors prefer a bottom-up market estimate? What does a large gap between top-down and bottom-up suggest?
3. A software product sells at USD 20,000 a year to 3,000 target firms; you can serve 40% of them today. Compute TAM
   and SAM.
4. Annual gross profit per customer USD 10,000, retention 80%, discount rate 10%. Compute LTV with the card's formula.
5. CAC is USD 15,000 and annual gross profit per customer is USD 10,000. What is the CAC payback in months?
6. Give two reasons a high LTV to CAC ratio can still describe a poor opportunity.
7. What is a kill criterion, and why should it be written before the test?
8. In Worked example A, why is the sales-capacity SOM the better planning number?
9. A weighted score ranks option X first only when one criterion carries 40% weight. What does that tell you?
10. Interleaving: which card or tool do you use? (a) Splitting ownership in a seed round. (b) Deciding whether to fund
    a pilot that can be stopped after six months. (c) Explaining why an industry earns low returns.

## My notes and next steps

<!-- Guidance: This section is yours. Rate your confidence honestly and plan the next review. -->

| Item | My entry |
|---|---|
| Confidence sizing a market two ways (1 = lost, 5 = could teach it) | [ ] |
| My riskiest assumption in a real or case opportunity | [ ] |
| First test and kill threshold | [ ] |
| Mistakes logged in mba/mba-error-log | [Date and short description] |
| Next review date (spacing: +2 days, +1 week, +1 month) | YYYY-MM-DD |
| Next card to study | [mba/mba-venture-finance-cap-table-card / mba/mba-customer-lifetime-value-card] |

## Sources and further reading

<!-- Guidance: Primary works are listed for attribution. Textbooks are named only. -->

- Blank, S. G. (2005). *The Four Steps to the Epiphany*. Self-published (CafePress).
- Ries, E. (2011). *The Lean Startup*. Crown Business.
- McGrath, R. G., and MacMillan, I. C. (1995). Discovery-driven planning. *Harvard Business Review*, 73(4), 44-54.
- Shane, S., and Venkataraman, S. (2000). The promise of entrepreneurship as a field of research. *Academy of
  Management Review*, 25(1), 217-226.
- Gupta, S., Lehmann, D. R., and Stuart, J. A. (2004). Valuing customers. *Journal of Marketing Research*, 41(1), 7-18.
- Osterwalder, A., and Pigneur, Y. (2010). *Business Model Generation*. Wiley.
- Textbooks (by name only): Barringer, B. R., and Ireland, R. D., *Entrepreneurship: Successfully Launching New
  Ventures*; Aulet, B., *Disciplined Entrepreneurship*; Christensen, C. M., Hall, T., Dillon, K., and Duncan, D. S.,
  *Competing Against Luck*.
- Open resource: OpenStax, *Entrepreneurship* (CC BY-NC-SA 4.0), chapters on opportunity recognition and feasibility.

## Answer key

<!-- Guidance: Check only after you have written your own answers. For each miss, write down why your first answer was wrong. -->

**Worked example B.** (1) SAM 172,800 households; SOM 4,320 subscribers; SOM revenue USD 1.503m a year; LTV USD 211.12
per subscriber; LTV to CAC 2.22. (2) Weighted score 3.00 (0.60 + 0.40 + 0.45 + 0.40 + 0.75 + 0.40). (3) Churn: at 35% a
year the subscriber stays under three years on average; test retention with a three-month paid trial cohort, kill
threshold for example monthly churn above 3% (about 31% a year). Even at 31% the LTV to CAC is only about 2.6, so
price, margin or CAC must also move for the case to clear 3x. Adoption is the second test. (4) Payback measures
speed of recovery, not total value; with high churn and a modest margin each subscriber returns only about 2.2 times
CAC in present value, below a 3x convention, and the SOM revenue of about USD 1.5m is small for Northfield.

**Exercise C.** (1) TAM 6,100 hospitals, USD 518.5m; SAM 3,355 hospitals, USD 285.2m; year-5 cumulative hospitals 70
(5 x (1+2+3+4+4)), SOM ARR USD 5.95m. (2) 2.09% of SAM hospitals. (3) Gross profit USD 66,300; LTV USD 304,980
(`=66300*92%/(1+12%-92%)`); LTV to CAC 2.54; CAC payback 21.72 months. (4) Weighted score 3.30. (5) Examples: KC-01
three of the first ten qualified hospitals sign within nine months, else stop; KC-02 idle theatre time falls by at least
[x]% at two hospitals that are not Tallow Creek, else pivot to a services model; KC-03 implementation under 12 weeks per
hospital, else rework the product before scaling sales.

**Self-test.**

1. TAM: annual revenue if every possible customer bought the category; SAM: the part reachable with your product,
   channels and geography; SOM: what you can realistically win in a stated horizon.
2. Each input (customers, units, price) can be checked in interviews and data. A large gap means at least one share,
   count or price is wrong, or the two estimates define the market differently.
3. TAM USD 60m (`=3000*20000`); SAM USD 24m.
4. $LTV = 10{,}000 \times 0.8/(1.1 - 0.8) = 26{,}667$, about USD 26,667.
5. 15,000 / (10,000 / 12) = 18 months.
6. Examples: the market is too small to build a meaningful business; large capital needs outside CAC; the ratio rests
   on an early, unrepresentative cohort; CAC excludes costs such as founder selling time.
7. A measurable test of a critical assumption with a threshold and date that triggers stop or pivot. Writing it first
   prevents fitting the criterion to the result.
8. It is tied to hiring, ramp and win rates the company controls and can track; the penetration figure is a top-down
   guess.
9. The ranking is fragile and depends on one judgement; discuss that weight explicitly and gather evidence on the
   criterion that drives it.
10. (a) mba/mba-venture-finance-cap-table-card; (b) a decision tree or real-options analysis
    (mba/mba-decision-trees-card); (c) mba/mba-five-forces-card.

## Revision history

<!-- Guidance: Add a row each time you replace assumptions with evidence. -->

| Version | Date | Author | Change summary |
|---|---|---|---|
| 0.1 | YYYY-MM-DD | [Your name] | First pass through the card |
| 0.2 | YYYY-MM-DD | [Your name] | *Example:* Replaced Frost assumptions with interview and pilot data, revised KC-03 |

## Standard and licence note

Based on the opportunity evaluation and customer development literature listed above (Blank 2005; McGrath and
MacMillan 1995; Ries 2011; Shane and Venkataraman 2000; Gupta, Lehmann and Stuart 2004). The text, examples, numbers
and questions are original; no textbook text is copied or paraphrased, and Corvane Robotics, Northfield Outdoor Co. and
Tallow Creek Medical Center are fictional. The template is independent and not affiliated with or endorsed by any
business school, publisher or investor. Market data in your own assessment must carry its source and date.
Template licence: MIT.