Chapter 26 — Week 8: Pricing, GTM, and unit economics

Welcome to Week 8. The beta report is signed off. Five of six centre owners (or whatever fraction your team produced) have committed to GA pricing in some form. The data flywheel has 200+ labelled inferences and is turning. By Friday you will have a finalised pricing structure backed by evidence from beta, a unit-economics model that demonstrates the business is plausibly viable, a market-sizing analysis that quantifies the upside, and a one-page go-to-market plan that the Week-9 pitch deck will draw from. The single most-common Week-8 mistake is to under-do the unit economics — to write a price down without the gross-margin, CAC, LTV, and payback arithmetic that demonstrates the price actually works as a business. This chapter is about doing the arithmetic seriously, in the context that distinguishes graduate-level startup work from undergraduate startup work.

Chapter overview

This chapter follows the same six-part structure. §26.1 (Concept) sets out pricing theory and practice for AI products, the unit-economics framework (LTV, CAC, payback, gross margin, the magic number), the TAM/SAM/SOM market-sizing methodology, go-to-market motion classification, the first-100-customers playbook with bowling-pin segmentation, pricing experiments at student-team scale (Van Westendorp, Gabor-Granger, “ask three prices”), and the path-to-break-even analysis. §26.2 (Method) is the day-by-day Week 8 sprint: pricing finalisation, unit-economics modelling, market sizing, GTM planning, integration and pitch preparation. §26.3 (Lessons from the cases) pulls eight specific pricing-and-GTM lessons from Parts I–III. §26.4 (Tools and templates) gives you the pricing-model decision rubric, the unit-economics spreadsheet structure, the TAM/SAM/SOM calculator, the GTM motion canvas, the sales-funnel template, the cohort-retention analysis pattern, and the deliverable bundle structure. §26.5 (Worked example) continues Team Aroma through their Week 8: pricing finalisation at RM 30/student/month with Year-1 introductory at RM 25; gross-margin calculation at 62% after foundation-model and infrastructure costs; bottom-up Malaysian SPM tutoring market sizing producing a TAM of RM 43.2M annual, an SAM of RM 14.1M (Klang Valley + Penang + JB), and an SOM of RM 1.65M (initial 110 centres reachable in Year 1); the founder-led B2B sales motion for the first 30 centres with explicit unit-economics calculations. §26.6 (Course exercises and deliverables) specifies the Week 8 submission with grading rubric.

How to read this chapter. Read §26.1 in full at Sunday-evening Week 7 or Monday-morning Week 8. The conceptual material is the densest of any Part V chapter; do not skip it. Read §26.2 with the team and assign per-day owners — pricing typically goes to the CEO/commercial lead, unit economics to the CFO-equivalent (often the team’s most-quantitative member), market sizing to the curriculum/research lead. Treat §26.3 as Wednesday-evening reading. Use §26.4 throughout the week. Read §26.5 before drafting your own deliverables on Thursday. Submit against §26.6 by Friday 23:59.

26.1 Concept

26.1.1 Pricing for AI products

Pricing theory has three classical approaches: cost-plus (the price is determined by the cost to produce plus a margin), value-based (the price is determined by the value the customer captures), and competitive (the price is determined by what competitors charge). Most successful B2B SaaS pricing in 2024–2026 is dominantly value-based, with cost-plus as a sanity check (the price must clear gross-margin requirements) and competitive as a constraint (the price must be defensible against alternatives). For AI products specifically, three additional considerations apply.

The capability-quality continuum complication. The same foundation-model API can produce $0.60 of value or $60.00 of value depending on the prompt engineering, the workflow integration, and the customer’s use case. Cost-plus pricing on the foundation-model API would price your product at the API cost — wildly under-pricing the integration value. Value-based pricing requires that you understand your customer’s willingness to pay, which the Week-7 pricing conversations have begun to establish.

The variable-cost-per-inference structure. Unlike classical SaaS, AI products have meaningful variable costs (typically USD 0.005–0.10 per inference depending on the model and task). The cost scales with usage, which means the pricing model must align with the cost structure. Per-seat pricing with unlimited usage on a freemium tier can produce negative unit economics if a customer uses the product heavily. The Klarna case (Chapter 8) is a recent reminder; their AI customer-service deployment is reported to have had per-conversation costs that were not fully internalised in the pricing.

Free-tier strategy. A free tier that allows unlimited usage is dangerous for AI products because variable cost scales with user activity. Three patterns work: - Hard caps on free tier usage (e.g., “5 inferences per day on free tier; upgrade for unlimited”). Easy to implement; clear customer expectation. - Reduced-capability free tier (e.g., “free tier uses smaller/cheaper model; paid tier uses frontier model”). Aligns cost with revenue tier. - No free tier (B2B-typical). The pricing conversation begins immediately at the paid tier.

Pricing models for B2B AI products in 2026.

Model	Mechanism	Strengths	Limitations	Examples
Per-seat	Price scales with named users	Simple to forecast and bill	Misaligned with usage variation	Cursor, Notion
Per-usage	Price scales with inferences/transactions	Aligned with cost structure	Less predictable revenue; harder for customer to budget	OpenAI API, Anthropic API
Hybrid (per-seat + usage cap)	Per-seat with included quota; overage billed	Predictable + aligned	More complex to communicate	Most enterprise SaaS
Tiered (good/better/best)	Discrete tiers with feature differentiation	Captures different willingness-to-pay segments	Tier design is hard	Anthropic Pro/Team/Enterprise; Slack Business/Plus/Enterprise+
Freemium	Free with usage caps; paid tier removes caps	Drives adoption	Can produce negative unit economics if not careful	ChatGPT Free vs Plus
Per-outcome	Price scales with successful outcomes	Aligned with value	Hard to define “outcome” cleanly; legal complexity	Some AI sales tools, recruiting AI

For most student-team B2B MVPs in 2026, the recommended default is per-seat with a usage component (e.g., “RM X per active user per month, including up to N inferences; overage at RM Y per inference”). The structure is forecastable for the customer and aligned with cost on the team’s side.

26.1.2 Unit economics primer

Unit economics is the per-customer (or per-segment) financial analysis that tests whether a business can be profitable at scale. Five quantities define unit economics; mastering them is the single highest-leverage analytical skill for graduate-level startup work.

Gross margin (GM). The fraction of revenue that remains after the direct cost of serving the customer (foundation-model inference, hosting, payment processing).

\[ \mathrm{GM} = \frac{R - C_{\text{direct}}}{R} \]

where $R$ is revenue and $C_{\text{direct}}$ is the direct variable cost. Classical SaaS achieves 70–85% gross margins. AI products in 2024–2026 typically achieve 50–75%; the foundation-model inference cost reduces gross margin relative to classical SaaS. A 60–70% GM is a reasonable Year-1 target; below 50% is concerning; above 80% suggests the model isn’t being heavily used (which is its own concern).

Customer acquisition cost (CAC). The cost of acquiring one new customer, including all sales and marketing spend allocated to that acquisition. For founder-led sales in Week 8, CAC is dominated by the team’s time; even at zero out-of-pocket spend, the team’s hourly opportunity cost is the implicit CAC.

\[ \mathrm{CAC} = \frac{\text{total sales \& marketing spend}}{\text{number of new customers acquired}} \]

For a student team’s Week-8 founder-led sales: estimate 4–8 hours of team time per centre acquired (cold outreach, calls, demos, follow-ups, contract negotiation) at an opportunity cost of, say, AUD/MYR 30/hour. 6 hours × MYR 30 = MYR 180/centre as a lower bound; 10 hours × MYR 50 = MYR 500/centre as an upper bound. The CAC is your time-investment estimate, not your zero-out-of-pocket spend.

Lifetime value (LTV). The total revenue (or contribution margin, depending on convention) from a customer over their full relationship with the product, discounted appropriately.

For a subscription product: \[ \mathrm{LTV} = \frac{\mathrm{ARPU} \cdot \mathrm{GM}}{\text{monthly churn rate}} \]

where ARPU is average revenue per user (per month) and the monthly churn rate is the fraction of customers who cancel each month. Equivalently, $\mathrm{LTV} = \mathrm{ARPU} \cdot \mathrm{GM} / r$ where $r$ is the constant churn hazard, giving expected lifetime $1/r$ months. Real customer cohorts have non-constant churn (typically front-loaded), but the simple form is a useful first approximation.

For Team Aroma’s Pulse with ARPU MYR 30/student/month, GM 62%, and an assumed monthly student-churn rate of 4% (95% retention/month — typical for B2B SaaS with sticky workflows): LTV = 30 × 0.62 / 0.04 ≈ MYR 465/student. (For centres rather than per-student, the math scales by the average student count.)

LTV:CAC ratio. The standard health metric. Industry benchmarks: - 3:1 or higher: healthy unit economics; the business can be invested in - 2:1 to 3:1: viable but constrained; aggressive growth may erode further - Below 2:1: unsustainable at the current pricing/CAC combination; either pricing must rise or CAC must fall

For Team Aroma, at LTV MYR 465/student and CAC at MYR 200/centre with 50 students per centre on average: LTV per centre = 465 × 50 = MYR 23,250; CAC per centre = MYR 200; ratio = 116:1. This is suspiciously high; it likely reflects (a) the team’s CAC estimate being too low (unaccounted-for time; underestimated churn) and (b) the LTV calculation using an optimistic 4% monthly churn. A more conservative Year-1 estimate would be CAC at MYR 600/centre and 7% monthly churn (early-stage churn is higher), giving LTV MYR 265/student × 50 = MYR 13,250, ratio = 22:1 — still healthy but more realistic.

CAC payback period. The number of months required for a customer’s gross-margin contribution to recover the CAC.

\[ \mathrm{Payback} = \frac{\mathrm{CAC}}{\mathrm{ARPU} \cdot \mathrm{GM}} \]

Industry benchmarks: <12 months is healthy for mid-market SaaS; <18 months is acceptable; >24 months is concerning. For Team Aroma at the conservative case (CAC MYR 600/centre, monthly contribution per centre = 50 × 30 × 0.62 = MYR 930), payback is 0.65 months — exceptional, because B2B selling to centres at 50+ students each is high revenue per acquisition relative to the founder-led-sales effort.

The unit-economics math is the discipline that distinguishes credible business cases from aspirational ones. A team that can articulate GM, CAC, LTV, and payback with their evidence is doing graduate-level startup work; a team that has only revenue and no margin/CAC analysis is doing undergraduate-level startup work.

26.1.3 The TAM / SAM / SOM framework

Market sizing is the analysis that tells investors and the team how large the upside could be. The standard framework has three nested sizes:

Total Addressable Market (TAM). The total revenue opportunity if your product were used by every plausible customer at full pricing. TAM is the universe.

Serviceable Available Market (SAM). The fraction of TAM you could realistically serve, given your distribution channels, regulatory permissions, and product capabilities. SAM is what you can reach now.

Serviceable Obtainable Market (SOM). The fraction of SAM you can realistically capture in your initial target window (typically 3–5 years). SOM is what you will plausibly capture.

Two methodological approaches:

Top-down sizing. Start with a large number from a published source (“the global EdTech market is USD 400B”), then drill down by applying realistic capture percentages. The approach is fast but produces inflated numbers because each step’s percentage is itself a guess and errors compound multiplicatively. Top-down TAMs of “USD 100B” with no defensible derivation are the canonical “TAM trap” of investor-presentation theatre.

Bottom-up sizing. Build up from the unit of analysis (per-customer or per-account revenue), multiplied by the number of plausible customers. The approach is slower but produces defensible numbers because each input is checkable.

For Team Aroma’s Pulse, the bottom-up sizing:

TAM (Malaysia SPM-focused tutoring centres, all states):
  Number of SPM-focused centres in Malaysia: ~3,200
  Average students per centre: ~50
  ARPU per student per month: RM 30
  Months per year: 12
  TAM = 3,200 × 50 × 30 × 12 = RM 57.6M annual

SAM (Klang Valley + Penang + Johor Bahru, the urban-Malaysia core):
  Number of centres in target geography: ~1,300
  Same per-centre assumptions
  SAM = 1,300 × 50 × 30 × 12 = RM 23.4M annual

SOM (5-year capture target):
  Aspirational 5-year capture: 15% of SAM
  SOM = 0.15 × 23.4M = RM 3.5M annual ARR by Year 5
  Year-1 SOM: ~30 centres = RM 540K annual ARR

The numbers above are illustrative and depend on inputs the team must source (the centre count from MyStartup directories, MOE data, or sector-association reports; the ARPU from beta pricing commitments; the geography distribution from population data). The discipline matters more than any specific number: every input is checkable, every multiplication is explicit, and the resulting TAM is at the right order of magnitude rather than off by a factor of 100.

26.1.4 Go-to-market motion classification

The GTM motion is how customers find and buy the product. There are three primary motions; most successful products have a dominant primary plus secondaries.

Product-led growth (PLG). Customers find the product themselves (typically through search, social, or word-of-mouth), try it self-serve, and convert via a freemium-to-paid or trial-to-paid funnel. Examples: Slack, Notion, Calendly, Cursor, Linear. PLG works when the product can demonstrate value within 5–10 minutes of first use without sales involvement.

Sales-led growth (SLG). A sales team identifies prospects, conducts outbound outreach, demonstrates the product, and closes contracts. Examples: Salesforce, ServiceNow, most enterprise software. SLG works when (a) the buying decision involves multiple stakeholders, (b) the contract value is high enough to support sales-team economics (typically AUD 30K+ ACV), and (c) the product requires customisation or integration.

Community-led growth (CLG). Customers find the product through a community (a Discord, a sub-Reddit, a meetup, an open-source repository). Examples: HashiCorp, GitHub, Hugging Face. CLG works when the product is technically deep and the community itself is part of the value proposition.

For most student-team B2B MVPs in 2026, the dominant motion is founder-led sales — a hybrid of SLG and PLG where the founders are doing the selling personally because they cannot yet afford a sales team. Founder-led sales has characteristic properties: high-touch (founders do every demo); slow per-customer (4–8 hours per centre acquired); but the conversion rate is high because founders carry the most credibility a startup has at this stage. The motion is appropriate for the first 30–100 customers; it does not scale to 1,000.

For B2C and prosumer products, the dominant Week-8 motion is typically PLG with a marketing-supported funnel: organic content, social media, influencer / community partnerships, paid acquisition (in moderation). The founder-led-sales pattern does not apply when the unit economics do not support the per-customer time investment.

26.1.5 The first 100 customers — bowling-pin segmentation

Geoffrey Moore’s Crossing the Chasm (1991) and Inside the Tornado (1995) developed the bowling-pin metaphor for early-stage segmentation: you target a specific segment first (the head pin), use early success there to credibly target an adjacent segment (the second pin), and so on. The segments compound — each conquered segment provides reference customers, case studies, and channel partnerships that make the next segment easier.

For Team Aroma’s Pulse, the bowling-pin sequence:

Pin 1 (Klang Valley SPM-focused centres, 50–200 students)
  → 30 centres in Year 1
Pin 2 (Penang + Johor Bahru, same segment characteristics)
  → 25 additional centres, Year 1.5–2
Pin 3 (Smaller centres, <50 students; lower ARPU but volume)
  → 50 additional centres, Year 2
Pin 4 (Adjacent subjects: BM and Sejarah, then Sciences)
  → cross-sell to existing centres, Year 2–3
Pin 5 (IGCSE / A-Level tutoring centres, similar workflow)
  → 15 centres in adjacent qualification, Year 3
Pin 6 (Singapore / Indonesia regional expansion)
  → 30 centres, Year 4

The sequence is not a forecast (the 30/25/50/etc. numbers are assumptions to be tested). The discipline is the ordering: each pin is conquered before the next, and the order is chosen so each pin’s conquest enables the next. Resist the temptation to attack multiple pins simultaneously; Week-8 sales effort should concentrate on Pin 1.

26.1.6 Pricing experiments at student-team scale

Three structured methods produce useful pricing signal at small scale.

The Van Westendorp price-sensitivity meter. Four questions per respondent:

At what price would the product be too expensive to consider buying?
At what price would it be expensive but you might consider it?
At what price would it be a bargain?
At what price would it be so cheap you would question its quality?

The four price points form distributions; their intersections identify the price range of acceptance. The optimal price is typically the intersection of “too expensive” and “bargain” (also called the “indifference price point”). The technique requires only 20–30 respondents to produce useful directional signal — well within student-team reach.

Gabor-Granger. A single question per respondent, varied across the sample: “Would you buy at price $P$?” with $P$ varied across the sample. The fraction of “yes” responses at each price gives the demand curve. Useful for confirming a specific price point rather than discovering the range.

The “ask three prices” technique. In the pricing conversation (per Chapter 25’s script), ask the user explicitly: “if our price were RM X, would you buy? RM Y? RM Z?” — typically three prices spanning 50–200% of the team’s planned target. The technique elicits the customer’s price sensitivity directly and is fast to run during the Week-7 pricing conversations or Week-8 follow-ups.

For Week 8, the realistic experiment is to use the Week-7 commitments (which used a single anchor price) plus 5–10 follow-up “ask three prices” conversations to triangulate the optimal price. Conjoint analysis (the gold standard for pricing experiments) requires more sophisticated execution and is typically deferred to post-launch when the cohort is larger.

26.1.7 The path to break-even

Most early-stage startups burn cash before reaching profitability — the J-curve of startup financials. Three quantities define the path:

Burn rate. The monthly cash outflow when revenue is below costs. For a student team’s beta and early-GA, burn is dominated by infrastructure (foundation-model API, hosting, tooling) plus founder time (which, in formal accounting, becomes a real cost once the team is full-time post-graduation). Pre-graduation: monthly burn might be USD/MYR 100–500. Post-graduation: USD/MYR 5,000–20,000 if founders take below-market salaries.

Runway. The cash available divided by the monthly burn. For a venture-backed startup, runway is the months of operation available before raising the next round. For a student team, runway is set by the team’s commitment to continuing (semester-aligned for most teams, pre-graduation continuation for some).

Time to break-even. The month at which monthly revenue equals monthly costs. Beyond break-even, the business is self-sustaining; before break-even, it requires capital.

For Team Aroma’s Pulse, an illustrative path:

Year 1, Month 1-3: Beta + early GA
  Revenue: ~RM 0-3,000/month (alpha free; beta partly free; first 2-3 paying centres)
  Costs: ~RM 2,000/month (infrastructure + tools; team time uncosted)
  Net: -RM 500 to +RM 1,000

Year 1, Month 4-6: Early scale
  Revenue: ~RM 5,000-12,000/month (10-15 paying centres)
  Costs: ~RM 4,000/month
  Net: +RM 1,000 to +RM 8,000

Year 1, Month 7-12: Pin 1 capture
  Revenue: ~RM 15,000-30,000/month (25-30 paying centres)
  Costs: ~RM 8,000-12,000/month
  Net: +RM 7,000 to +RM 18,000

Break-even reached: roughly Month 4-5 if founders unsalaried
                    roughly Month 14-18 if founders take MYR 5K/month each

The arithmetic produces a forecastable path that the team can defend in pitching. The quality of the forecast depends on the input estimates (centre acquisition rate, churn, ARPU); each input should be defensible from beta evidence or industry benchmarks.

26.2 Method — the Week 8 sprint

26.2.1 Day 1 (Monday): pricing finalisation

By Monday end-of-day the team has a finalised pricing structure with explicit defensibility:

The headline price (per-seat, per-usage, hybrid, or tiered structure with specific numbers)
Year-1 introductory pricing (typically 15–30% below the long-term price) to drive early adoption
The free / freemium / trial structure (or none)
Discount policies (volume, multi-year, non-profit)
The Week-7 commitments mapped to the new pricing — which commitments still hold, which need renegotiation

The pricing decision is made with input from beta evidence: the price points where Week-7 conversations produced “yes,” “yes with conditions,” and “no” form the empirical floor and ceiling. Pricing significantly below the floor is leaving money on the table; pricing above the ceiling is missing the obtainable market. The optimal point is typically near the upper end of the “yes” range.

For Team Aroma’s pricing Monday: the Week-7 conversations established that RM 30/student/month was acceptable to most centres, with RM 25 strongly preferred by smaller centres. The team commits to:

Headline price: RM 30/student/month for centres ≥50 students; RM 35/student/month for centres <50 students (smaller centres have higher per-student cost-to-serve due to fixed overhead; the price reflects this).
Year-1 introductory: RM 25/student/month for the first 12 months from contract, applicable to centres signing in Year 1.
Volume discount: 10% for centres ≥150 students; 15% for centres ≥250 students.
Free / freemium: No free tier; 30-day trial available with a designed 5-day “aha moment” path.
Conditional commitments: T1’s customisation requirement is added to Week 8 must-build; T4’s Mandarin requirement is queued for Year-1 but not Year-1-month-1.

26.2.2 Day 2 (Tuesday): unit-economics modelling

By Tuesday end-of-day the team has a populated unit-economics model in Google Sheets / Excel / Airtable. The model should be a single-page summary view with detail tabs.

The summary view:

PULSE UNIT ECONOMICS — [DATE]
Per centre, monthly basis

REVENUE
  Pricing per student/month (Year 1 intro):       MYR 25
  Pricing per student/month (post Year 1):         MYR 30
  Average students per centre:                     50
  Monthly revenue per centre, Year 1:              MYR 1,250
  Monthly revenue per centre, post Year 1:          MYR 1,500

DIRECT VARIABLE COSTS
  Foundation-model inference cost:                 MYR ~150 (50 students × 5 inferences × MYR 0.20 × 30 days)
                                                   [actual is closer to MYR 90 based on beta data]
  Infrastructure (hosting, DB, auth, analytics):   MYR 30
  Payment processing (3% × revenue):                MYR 38–45
  Customer support (2% of revenue allowed):        MYR 25–30

  Total direct variable cost:                      MYR ~250–300

GROSS MARGIN
  Year-1 GM = (1,250 - 280) / 1,250 =              78%   (with conservative 280 cost)
  Post-Year-1 GM = (1,500 - 280) / 1,500 =         81%

CUSTOMER ACQUISITION COST (CAC)
  Founder-led sales hours per centre (estimate):   8 hours
  Opportunity cost per hour (Year 1 ramp):         MYR 50
  CAC per centre:                                  MYR 400

LIFETIME VALUE (LTV)
  Average customer lifetime (assumption):          24 months
  Monthly contribution margin per centre:          MYR 970–1,220 (rev × GM)
  LTV per centre (24-month base case):              MYR 23,000–29,000

UNIT ECONOMICS HEALTH
  LTV : CAC =                                      57x – 73x
  Payback period =                                 0.4–0.5 months
  Status:                                          HEALTHY

The numbers above are Team Aroma’s Tuesday calculation. They should be sanity-checked against industry benchmarks. The exceptionally high LTV:CAC reflects (a) B2B selling to centres where each acquisition produces 50 students of revenue, and (b) the founder-led-sales CAC being low because of personal networks. As the team scales beyond Pin 1, both numbers will deteriorate; the team’s Year-2 forecast should assume LTV:CAC drops to 5–15x as referrals are exhausted and the team needs to spend on lead generation.

A specific note on cost discipline. The MYR 90 actual foundation-model cost (vs the conservative MYR 150 in the model) is a real gap: the team’s beta data shows roughly 5 inferences per student per day on active days, with ~15 active days per month per student, at MYR 0.04 per inference = 50 × 5 × 15 × 0.04 = MYR 150/month. Hmm. Actually, let me redo. If actively 5 inferences per active student per active day and 15 active days per student per month: 50 students × 5 × 15 × 0.04 = MYR 150. So the conservative figure is right, and the MYR 90 reflects only ~3 active days per student per month. The team’s modelling should use the right number for the customer profile they expect; if the modal centre has students using the system 3 days a week (12–15 active days/month), MYR 150 is closer to reality.

26.2.3 Day 3 (Wednesday): market sizing

By Wednesday end-of-day the team has a defensible bottom-up TAM / SAM / SOM analysis with sources for every input.

Method:

Identify the unit of analysis. For Team Aroma: a centre, with associated students.
Source the population count. For Malaysian SPM-focused tutoring centres: data from MyStartup directory + MyDIGITAL business listings + MOE-recognised tutorial centre registry + cross-check against industry estimates from sector reports. The team’s input: ~3,200 SPM-focused centres in Malaysia.
Source the per-unit revenue. From the Week-7 pricing commitments: average MYR 30/student/month at 50 students/centre = MYR 1,500/centre/month = MYR 18,000/centre/year.
Compute TAM: 3,200 × MYR 18,000 = MYR 57.6M annual.
Define the geographic / capability constraints to derive SAM. Klang Valley + Penang + JB ≈ 1,300 centres. SAM = 1,300 × 18,000 = MYR 23.4M annual.
Estimate the obtainable share over a multi-year horizon. 5-year aspirational capture of 15% of SAM = MYR 3.5M annual ARR by Year 5. This is the SOM.
Year-1 sub-target: 30 centres × 18,000 = MYR 540K annual ARR.

The key quality control: every input has a source, every multiplication is shown, every assumption is named. A team that cannot defend the 3,200 centre number, the 50-students-per-centre average, or the 15% capture target has not done the work seriously.

26.2.4 Day 4 (Thursday): GTM planning

By Thursday end-of-day the team has a one-page GTM plan covering the motion, channels, sales playbook, and Year-1 targets.

The GTM canvas:

GO-TO-MARKET PLAN — PULSE — YEAR 1

PRIMARY MOTION
  Founder-led B2B sales for the first 30 centres
  (Weeks 9-26 of the post-graduation calendar; ~6 months)

SECONDARY MOTION
  Referral from existing customers (ramping from Month 3)
  Sector PR (tutoring-association magazine; Berita Harian education page) from Month 6

CHANNELS
  Direct outreach: LinkedIn + warm intros (primary)
  Sector events: STEM Education Malaysia, Persatuan Pusat Tuisyen
  PR: 1-2 placed articles in Year 1
  Content: SEO-optimised teacher-resource articles (defensive, not primary)

SALES PLAYBOOK
  Step 1: identify centre owner via LinkedIn or warm intro
  Step 2: 15-min call to qualify (size, focus, current tools)
  Step 3: 30-min demo at the centre (in-person if KL; remote otherwise)
  Step 4: 30-day trial with paid pilot from week 2
  Step 5: contract negotiation; expect 4-6 weeks total cycle

YEAR-1 TARGETS
  Centre signups: 30 paying centres by month 12
  ARR by month 12: MYR 540K
  Churn: <8% monthly through Month 6, <5% through Month 12
  Reference customers: 5 centres willing to be quoted publicly

CONSTRAINTS AND RISKS
  Speed: founder-led requires founder time; cannot exceed 5 centres
         acquired/month sustainably
  Geography: KL focus; Penang and JB as Pin 2 require travel
  Mandarin: T4-style requirements need product investment in Q2
  Competing priority: founders also building product

The plan is compact but defended at every step. The Year-1 targets are the testable forecast that Week-9’s pitch will defend.

26.2.5 Day 5 (Friday): integration and pitch preparation

Friday integrates the four prior days’ outputs into the deliverable bundle and begins shaping the Week-9 pitch deck. The team:

Cross-checks pricing, unit economics, market sizing, and GTM for internal consistency. Inconsistencies (e.g., the GTM plan calling for 30 centres in Year 1 but the unit-economics model assuming much higher acquisition rates) are reconciled.
Drafts the 1–2 pages of the eventual Week-9 pitch deck that will summarise this week’s economic work — typically two slides (the “business model and unit economics” slide and the “market and GTM” slide).
Updates the team’s working narrative: “we have a working product, X paying customers, Y MRR / committed MRR, Z TAM with a defended path to capture, and these are our unit economics.”

The Friday submission goes in by 23:59. The team-comprehension penalty applies; every team member must be able to recite the basic numbers (price, GM, CAC, LTV, payback, TAM, SOM, Year-1 target) without referring to documents.

26.3 Lessons from the cases

Eight specific pricing-and-GTM lessons from Parts I–III shape Week 8 work.

26.3.1 Stripe — pricing as developer experience (Chapter 6, forthcoming)

Stripe’s pricing was 2.9% + 30¢ per transaction from launch — explicit, simple, and consistent. The price was not buried in negotiation; it was the same for everyone, on the homepage. The transparency was both a developer-experience feature and a sales-cycle accelerator: developers could decide to use Stripe in 5 minutes without escalating to procurement.

Operational implication. Resist the temptation to hide pricing or run “contact sales” on the website. For most B2B AI products targeting SMB and mid-market, transparent pricing closes 5–10× faster than negotiated pricing because the buyer can self-qualify. Negotiated pricing makes sense only at enterprise scale (>USD 100K ACV).

26.3.2 Slack — freemium with the right cliff (Chapter 5)

Slack’s freemium tier was generous (unlimited users in a workspace; 10K message history) but had a strategic cliff (history limited; integrations capped). The cliff was placed precisely at the point where a team had become Slack-dependent — which made the upgrade decision rational rather than coercive. Conversion from free to paid for engaged teams ran at 30–40%.

Operational implication. A freemium tier works only if the cliff — the point where the user must pay — is at the right place. Too early (in the first session) feels like a paywall; too late (after the user has integrated the product into critical workflows) feels uncoercive but converts poorly. For AI products specifically, the variable-cost-per-inference structure makes generous freemium dangerous; consider quota-capped freemium with the cap at first-aha-moment + N additional uses.

26.3.3 Cursor — per-seat with usage-tier complexity (Chapter 5)

Cursor’s pricing in 2024–2026 evolved from simple per-seat (USD 20/month/user) to tiered (Pro at $20, Business at $40, with progressively richer model access and indexing). The tiering captured higher willingness-to-pay segments without alienating individual developers. By 2025 the product also added a usage-overage component for heavy users, aligning revenue with cost.

Operational implication. Most successful AI-product pricing in 2024–2026 evolves from simple per-seat to tiered + usage hybrid as the customer base diversifies. Plan for this evolution; do not lock yourself into a pricing structure that works for your first 30 customers but will not scale.

26.3.4 ChatGPT Plus — pricing as anchoring (Chapter 13)

OpenAI’s introduction of ChatGPT Plus at USD 20/month in February 2023 served as an anchor for the entire consumer-AI subscription space. Subsequent products (Anthropic Claude Pro, Perplexity Pro, GitHub Copilot Individual, Cursor Pro) all priced near or at this anchor. The anchor’s stability through 2024–2026 reflected genuine cost economics (foundation-model inference cost) and cross-product competitive dynamics (consumers who already pay USD 20 to one provider are unwilling to also pay USD 30 to another).

Operational implication. Reference-price anchors matter. Your B2C / prosumer pricing should locate within the established anchor range (USD 10–30/month for consumer AI products in 2026); pricing above the anchor without a clear reason produces sales friction. For B2B, the anchor structure is different but exists similarly — most B2B AI tools cluster at USD 10–50/seat/month.

26.3.5 Carsome — bottom-up market sizing in SE Asia (Malaysian unicorn case)

Carsome’s investor pitches in 2018–2021 used bottom-up market sizing for Southeast Asian used-car markets: number of cars × annual transaction rate × average price × commission rate, computed per country. The bottom-up approach was defensible — investors could check the inputs against Carsome’s actual data — and produced TAM / SAM / SOM estimates that informed funding decisions.

Operational implication. Bottom-up sizing for Southeast Asian markets is more defensible than top-down because (a) the published top-down “global” or “Asia” market sizes from research firms are typically unreliable for the SE Asia subset; (b) the bottom-up approach uses inputs the team can defend from primary research. Your bottom-up sizing should use the most-specific available data sources (MyStartup, government registries, industry-association data) rather than aggregator estimates.

26.3.6 Notion — team-led growth (Chapter 5)

Notion’s distribution from 2018 onward was team-led growth: individuals adopted the product, brought it into their teams, and the teams converted to paid. The motion required no outbound sales for the first ~3 years of growth; it required only that the product be good enough that users brought it into their teams voluntarily.

Operational implication. For some product categories — particularly productivity, collaboration, and knowledge tools — team-led growth is a primary motion. Foundation-model-based AI tools that fit team workflows (writing assistants, meeting summarisers, knowledge bases) often follow the same pattern. If your product fits this category, the GTM plan should de-emphasise outbound sales and emphasise the in-product virality mechanisms (sharing, exporting, public links).

26.3.7 Anthropic — enterprise pricing as the upper-tier expansion (Chapter 13)

Anthropic’s pricing has three tiers: API direct (consumption-based), Pro / Team (subscription), and Enterprise (negotiated, with data-residency, compliance, and dedicated-capacity options). The Enterprise tier captures customers whose willingness-to-pay is much higher than the consumer or SMB tiers, without requiring a separate product.

Operational implication. Even at student-team scale, plan for the future-Enterprise tier. The current MVP serves SMB-tier customers (centres at MYR 30/student/month); the eventual Enterprise tier might serve school-district contracts at MYR 100K+ ACV with custom features (data residency, SLA, dedicated support). Document the future tiers in your pricing strategy even if you do not implement them in Year 1.

26.3.8 The DeepSeek shock — pricing pressure on the foundation-model layer (Chapter 5)

The DeepSeek-R1 release in January 2025 dropped frontier-reasoning model costs by an order of magnitude. The pricing pressure transferred to foundation-model APIs (OpenAI, Anthropic, Google all reduced API prices in 2025), which transferred to the cost structure of every AI product built on these APIs. By mid-2026, foundation-model inference costs are 50–80% lower than 2023 levels at equivalent capability.

Operational implication. Build your unit-economics model with declining foundation-model costs as an input. A product whose Year-1 GM is 60% based on current API costs might have a Year-3 GM of 75% as costs continue to fall. This is a real upside in your forecast that investors will recognise. Conversely, building unit economics that depend on current API costs being permanent is fragile; the cost trajectory has been clearly downward.

26.4 Tools and templates

26.4.1 Pricing model decision rubric

PRICING MODEL DECISION RUBRIC

Q1: Is your product B2B or B2C?
  B2B → continue to Q2
  B2C → use subscription tier (good/better/best); free trial; price within USD 10-30/month anchor

Q2: Does usage vary substantially across customers?
  Yes (>5x range) → use hybrid (per-seat + usage component) or pure per-usage
  No (<3x range) → use per-seat with usage caps

Q3: Is the contract value typical >USD 30K ACV?
  Yes → consider Enterprise tier with custom pricing
  No → publish pricing transparently; avoid "contact sales"

Q4: Is variable cost per use significant (>5% of price)?
  Yes (most AI products) → ensure pricing has usage component or capped usage
  No → pure per-seat is acceptable

Q5: Is there a clear "aha moment" within first session?
  Yes → freemium or trial with usage caps
  No → no free tier; trial period only

26.4.2 Unit-economics spreadsheet template

A four-tab Google Sheets / Excel structure:

Tab 1: Inputs - Pricing assumptions (headline, intro, discount tiers) - Cost assumptions (per-inference cost, hosting, payment fees) - Usage assumptions (inferences per active student per active day; active days per month per student) - Acquisition assumptions (founder-hours per centre, hourly opportunity cost) - Retention assumptions (monthly churn rate)

Tab 2: Per-customer math - Revenue per customer per month (Year 1 + post-Year-1) - Direct cost per customer per month - Gross margin per customer - CAC per customer - LTV per customer (at assumed lifetime) - LTV:CAC, payback period

Tab 3: Cohort projections - Month-by-month customer count (Months 1–24) - Month-by-month revenue and cost - Month-by-month net cash flow - Cumulative net cash flow (the J-curve) - Break-even month

Tab 4: Sensitivity - Three scenarios: pessimistic (slow acquisition; high churn), base case, optimistic (fast acquisition; low churn) - Tornado chart showing which inputs the outcome is most sensitive to - Stress test: if foundation-model cost rises 50%, what happens? - Stress test: if churn rises to 10% monthly, what happens?

The sensitivity tab is the most-important for stress-testing the business case. Investors will probe each input; the team should know which inputs the outcome is most sensitive to.

26.4.3 TAM / SAM / SOM calculator

TAM / SAM / SOM CALCULATOR

UNIT OF ANALYSIS
  [unit, e.g., centre, household, employee]

POPULATION COUNT (TAM scope)
  Source: [...]
  Number: [n]

PER-UNIT ANNUAL REVENUE
  Pricing: [...]
  Average usage / count per unit: [...]
  Per-unit annual revenue: [revenue]

TAM = [n] × [revenue] = [TAM]

GEOGRAPHIC / CAPABILITY CONSTRAINTS (SAM)
  Constraint 1 (geography): [...]
  Constraint 2 (regulation): [...]
  Constraint 3 (channel access): [...]

  Reduced population: [n_SAM]

SAM = [n_SAM] × [revenue] = [SAM]

OBTAINABLE-MARKET CAPTURE (SOM)
  Time horizon: [years]
  Capture target: [%]

  Captured units: [n_SOM]

SOM = [n_SOM] × [revenue] = [SOM]

YEAR-1 SUB-TARGET
  Realistic Year-1 acquisitions: [n_Y1]
  Year-1 revenue: [Y1_rev]

26.4.4 GTM motion canvas

(See §26.2.4 above; the one-page canvas with motion, channels, playbook, targets, and constraints.)

26.4.5 Sales pipeline / funnel template

SALES PIPELINE — [PROJECT]

STAGES
  Lead         [count]   [conversion to next %]
  ↓
  Qualified    [count]   [conversion %]
  ↓
  Demo'd       [count]   [conversion %]
  ↓
  Trialing     [count]   [conversion %]
  ↓
  Negotiating  [count]   [conversion %]
  ↓
  Closed Won   [count]   [final close rate %]

NAMED PIPELINE (Year-1 target: 30 closed-won)

| Centre | Stage | Owner | Last action | Next step | Probability % |
|---|---|---|---|---|---:|
| [Bright Star] | Trialing | Aliyah | Demo last week | Send pricing | 60% |
| [Excel Education] | Closed Won | Aliyah | Contract signed | Onboard | 100% |
| ...

26.4.6 Cohort retention analysis template

COHORT RETENTION ANALYSIS

For each month-of-acquisition cohort, track active centres in subsequent months.

           Month 1   Month 2   Month 3   Month 4   ...
Cohort 1   100%      94%       88%       82%
Cohort 2   100%      96%       91%       87%
Cohort 3   100%      95%       90%       —
...

Implied monthly churn:
  Cohort 1: ~6%
  Cohort 2: ~4-5%
  Cohort 3: ~5%

Trend: [improving / stable / deteriorating]

Implication for LTV:
  Use the rolling 3-cohort average for current LTV calculations.
  Be cautious about cohort selection effects (early adopters may
  retain better than later cohorts).

26.4.7 Pricing-experiment templates

Van Westendorp (4 questions per respondent):

For [product] with the features described:

1. At what price would you consider it TOO EXPENSIVE
   to consider buying? RM [...]

2. At what price would you consider it EXPENSIVE BUT
   STILL WORTH CONSIDERING? RM [...]

3. At what price would you consider it A BARGAIN —
   a great deal? RM [...]

4. At what price would you consider it SO CHEAP that
   you'd question its quality? RM [...]

Plot the four distributions. Optimal price ≈ intersection of “too expensive” and “bargain.”

The “ask three prices” inline:

"I want your honest reaction. If our price were:
  RM 20 per student per month — would you buy?
  RM 30 per student per month — would you buy?
  RM 50 per student per month — would you buy?

For each, please tell me: Yes / Maybe / No, and why."

26.4.8 The Week-8 deliverable bundle structure

WEEK 8 DELIVERABLE BUNDLE

1. Pricing structure document (1-2 pages)
2. Unit-economics spreadsheet (4-tab Excel/Sheets)
3. Market-sizing analysis (TAM/SAM/SOM with sources)
4. GTM plan (1 page)
5. Sales pipeline / funnel snapshot
6. Pricing-experiment results (if conducted)
7. Cohort retention analysis (with current beta cohort)
8. Updated risk register
9. Two pitch-deck slides (business model & market) — draft

26.5 Worked example — Team Aroma’s Week 8

Team Aroma starts the week with the Week-7 beta report’s commercial signal: 5 of 6 centre owners committed to GA pricing in some form.

Day 1 (Monday): pricing finalisation

Aliyah leads the pricing decision. The team reviews the Week-7 commitments:

T2 (Excel Education, ~120 students): committed at RM 30/student/month for 6 months. The strongest signal.
T3 (Sri Murugan, ~50 students): committed at RM 25/student/month (asked for new-customer discount).
T1 (Bright Star, ~80 students): committed at RM 25 conditional on customisation.
T4 (Smartway, ~30 students): committed at RM 30 conditional on Mandarin support.
T6 (~40 students): committed at RM 30 conditional on WhatsApp integration.

The pricing structure:

HEADLINE PRICING
  RM 30/student/month (centres ≥50 students)
  RM 35/student/month (centres <50 students)

YEAR-1 INTRODUCTORY PRICING
  RM 25/student/month for first 12 months for all centres signing in Year 1

VOLUME DISCOUNTS
  10% off for centres ≥150 students
  15% off for centres ≥250 students

FREE / FREEMIUM
  No free tier
  30-day trial, 100% feature, with 5-day onboarding programme

FEATURE-CONDITIONAL PRICING
  Centres requesting customisation receive customisation work in
  Year 1 at no additional charge (covered by the 16.7% premium of
  RM 30 vs RM 25 over 12 months).

The team contacts T1, T4, and T6 with the explicit pricing commitment that customisation, Mandarin support, and WhatsApp integration will all be addressed in Year 1 at the standard pricing. T1 and T6 confirm; T4 says he wants Mandarin in Q2 specifically; the team commits to that timeline.

Total commitment as of Monday end-of-day: 5 centres committed to RM 25/student/month introductory for 12 months, with a confirmed conversion to RM 30 at month 13. Total ~390 students × MYR 25 = MYR 9,750/month MRR committed for Year 1, transitioning to MYR 11,700/month at end of Year 1.

Day 2 (Tuesday): unit-economics modelling

Wei Hao and Daniel build the unit-economics spreadsheet over Tuesday. Inputs derived from beta data:

Foundation-model cost per inference: MYR 0.04 actual (vs MYR 0.045 budget — slight tailwind)
Inferences per active student per active day: 4.2 (beta median)
Active days per student per month: 14 (beta median)
Cost per student per month: 4.2 × 14 × 0.04 = MYR 2.35 → 50 students × 2.35 = MYR 117.50 (centre-level)
Hosting + auth + DB + analytics: MYR 35/centre/month
Payment processing: 3% × MYR 1,500 = MYR 45/centre/month
Other support / overhead allowance: MYR 30/centre/month
Total direct cost: ~MYR 230/centre/month

Metric	Year 1 (MYR 25)	Post Year 1 (MYR 30)
Revenue per centre/month	1,250	1,500
Direct cost	230	230
Gross profit	1,020	1,270
Gross margin	82%	85%

CAC calculation:

Founder-led sales hours per centre: 8 hours estimated based on Week-7 experience (1 hr LinkedIn outreach; 2 hrs demo; 2 hrs follow-up; 3 hrs trial-period support)
Opportunity cost: MYR 60/hour (graduate-level salary equivalent)
CAC = 8 × 60 = MYR 480/centre

LTV calculation:

Assumed monthly churn: 5% (post Month 6, when product matures)
Average lifetime: 1/0.05 = 20 months
Average monthly contribution margin: ~MYR 1,150 (blended across Year 1 and post-Year 1)
LTV per centre = 1,150 × 20 = MYR 23,000

Health metrics:

LTV : CAC = 23,000 / 480 = 48:1
Payback period = 480 / 1,020 = 0.47 months
Status: HEALTHY but suspiciously favourable

The team includes a sensitivity analysis: if churn rises to 10% monthly (lifetime 10 months), LTV drops to MYR 11,500 and LTV:CAC is 24:1, still healthy. If CAC rises to MYR 1,500 per centre as referrals are exhausted in Year 2 and the team needs paid lead generation, the ratio drops to 8:1 — at the lower end of “healthy” but still acceptable. The team commits to monitoring CAC and LTV monthly post-launch.

Day 3 (Wednesday): market sizing

Daniel leads the market-sizing work. Sources used:

Number of SPM-focused tutoring centres in Malaysia: MyStartup directory (limited; ~800 centres listed) + Persatuan Pusat Tuisyen Malaysia 2024 estimate (~3,200 SPM-focused centres) + cross-check against Department of Statistics Malaysia education-services census.
Geographic distribution: Department of Statistics by state. Selangor + KL + Penang + Johor account for ~42% of secondary-school-aged population, suggesting 1,300 centres in the urban-Malaysia core.
Average students per centre: Sample from the Week-7 beta cohort plus Persatuan estimates: 50 students per centre as the modal size, with substantial tail variation.
Per-centre monthly revenue at GA pricing: MYR 30 × 50 = MYR 1,500/centre/month = MYR 18,000/year.

Tier	Definition	Centres	Annual revenue
TAM	All Malaysian SPM-focused centres at GA pricing	3,200	MYR 57.6M
SAM	Klang Valley + Penang + Johor Bahru centres	1,300	MYR 23.4M
5-year SOM	15% capture of SAM	195	MYR 3.5M
Year-1 SOM target	Founder-led sales realistic capacity	30	MYR 540K

The team adds adjacent-market sizing for the Pin-2 and Pin-3 pins:

IGCSE / A-Level tutoring centres in Malaysia: ~600 centres. ARPU per student is higher (these are higher-income segments). Adjacent TAM: MYR 18M.
Singapore SPM-equivalent tutoring: ~400 centres. Average ARPU at SGD 80/student/month (regional adjustment): ~SGD 19.2M = MYR 67M annual TAM.
Indonesia + Thailand: not sized at this stage (Year-3+ consideration).

The total expanded SE-Asia TAM the team can defensibly point to: ~MYR 140M annual. The expanded TAM is what justifies the ambition of the business; the SOM is what justifies the plan.

Day 4 (Thursday): GTM plan

Aliyah and Sara write the GTM canvas. The Year-1 plan:

Months 1–6 post-graduation: founder-led B2B sales, target 15 centres signed (3/month average)
Months 7–12 post-graduation: founder-led continues at 4/month plus 1/month from referrals; target 30 centres signed by Month 12
Channels: LinkedIn outreach (primary), warm intros from existing customers (growing source), Persatuan Pusat Tuisyen partnership (mid-Year 1), 1 placed article in Berita Harian education page (Month 6)
Sales playbook: the 5-step playbook from §26.2.4
Year-1 pipeline target: 90 leads → 60 qualified → 45 demo’d → 35 trialing → 30 closed-won

The team also identifies the constraints that shape the plan:

Founder time is the binding constraint; cannot exceed 5 centres/month sustainably
Geography: founder travel limits Penang + JB to remote demos initially
Mandarin investment in Q2 unlocks T4-style customers (estimated +5 centres in Year 1)
Customisation work absorbs ~20% of engineering capacity through Q2; non-customisation feature roadmap slows correspondingly

Day 5 (Friday): integration and pitch preparation

Friday afternoon Sara drafts the Week-9 pitch deck’s two business slides. The slides distill Tuesday’s unit economics, Wednesday’s market sizing, and Thursday’s GTM into the most-defensible visual summary.

Slide 14 — Business Model and Unit Economics:

Pricing                                       Unit Economics
RM 25/student/month (Year 1 intro)            Revenue/centre:    RM 1,500/mo
RM 30/student/month (post Year 1)             Direct cost:       RM 230/mo
                                              Gross margin:      85%
                                              CAC:               RM 480
                                              LTV:               RM 23K (24-mo)
                                              LTV:CAC:           48:1
                                              Payback:           0.5 mo

Sensitivity (10% churn / 3x CAC stress):      LTV:CAC = 8:1, payback 1.4 mo

Slide 15 — Market and GTM:

Malaysia SPM Tutoring Centres
TAM (national):  ~3,200 centres, RM 57.6M annual
SAM (urban):      ~1,300 centres, RM 23.4M annual
SOM Year 5:        15% of SAM = RM 3.5M annual ARR
SOM Year 1:        30 centres, RM 540K annual ARR

GTM: Founder-led B2B → referral-driven → channel-supported

Pin sequence:
  Pin 1: KL/Selangor SPM centres → 30 by Y1 EOM12
  Pin 2: Penang + JB → 25 in Y2
  Pin 3: SE Asia regional → Y3+

Adjacent expansion: IGCSE / A-Level (RM 18M TAM); Singapore (RM 67M TAM)
Total SE Asia TAM: ~RM 140M

The Friday submission goes in at 11pm KL: the pricing structure document, the unit-economics spreadsheet, the market-sizing analysis with sources, the GTM plan, the sales pipeline snapshot, the cohort retention analysis (showing the beta cohort’s first-month retention at 89%), the two draft pitch slides, and the updated risk register.

What Team Aroma got right and what they almost got wrong

Three things they did well: (1) the unit-economics model used real beta data for the cost inputs (MYR 0.04 per inference, 4.2 inferences/student/active-day, 14 active days/month) rather than estimated numbers — the model is defensible against scrutiny because every input has a beta-derived source; (2) the sensitivity analysis showing LTV:CAC of 8:1 even at 10% churn and 3× CAC stress is what makes the model investor-credible (an unstressed 48:1 ratio looks suspicious); (3) the Pin-1 → Pin-2 → Pin-3 sequencing in the GTM plan is concrete and time-bounded rather than aspirational.

Three things they almost got wrong: the team almost used “global EdTech is USD 400B” as their TAM (the top-down trap; the bottom-up MYR 57.6M is much more defensible); they almost set CAC at MYR 100 by counting only out-of-pocket spend (the time-cost-included MYR 480 is what investors expect); and they almost set Year-1 SOM at 100 centres (which would have implied unsustainable 8 centres/month founder-led sales pace; the 30 centres target is realistic given the founder-time constraint).

The pattern is general. Week 8 is high-leverage because the unit-economics discipline produces the defensible business case that Week 9’s pitch is built on. A pitch with strong product traction but weak unit economics is rejected by serious investors; a pitch with thoughtful, evidence-based unit economics — even modest in absolute scale — is treated as credible.

26.6 Course exercises and Week 8 deliverable

Submit the Week 8 deliverable bundle by Friday 23:59. Required artefacts:

26.6.1 Required artefacts

Pricing structure document (1–2 pages) covering headline pricing, introductory pricing, discount tiers, free/freemium structure, and conditional commitments from beta.
Unit-economics spreadsheet (4-tab Excel / Sheets) with inputs, per-customer math, cohort projections, and sensitivity analysis.
Market-sizing analysis with TAM, SAM, SOM, and Year-1 SOM target, all bottom-up with cited sources.
GTM plan (1 page) per the §26.2.4 canvas structure.
Sales pipeline snapshot showing leads, qualified, demo’d, trialing, negotiating, closed-won counts; named pipeline at any meaningful scale.
Pricing experiment results if any conducted (Van Westendorp, Gabor-Granger, or “ask three prices” follow-ups).
Cohort retention analysis for the beta cohort showing first-month retention.
Two draft pitch deck slides (business model & unit economics; market & GTM) for use in Chapter 27’s pitch construction.
Updated risk register including unit-economics-related risks (CAC inflation, churn, foundation-model cost changes, regulatory).

26.6.2 Grading rubric (50 points)

Component	Points	Distinction-level criteria
Pricing rationale	5	Structure tied to beta evidence; conditional commitments addressed; defended against alternatives
Unit economics rigour	15	All five quantities (GM, CAC, LTV, payback, LTV:CAC) computed; inputs from beta data; sensitivity analysis included
Market sizing defensibility	10	Bottom-up methodology; every input cited; TAM, SAM, SOM, Year-1 SOM all distinguished
GTM plan quality	5	Motion, channels, playbook, targets all populated; constraints named
Sales pipeline state	5	Pipeline tracked through stages; named opportunities; Year-1 target connected to capacity
Pitch slide drafts	5	Two slides ready for Week 9 revision; numbers consistent with rest of the bundle
Risk management	5	Unit-economics-specific risks identified; sensitivity analysis stress-tested

Pass: 30. Credit: 36. Distinction: 42. High Distinction: 47.

The team-comprehension penalty applies; additionally, every team member must be able to recite the pricing, gross margin, CAC, LTV, LTV:CAC ratio, and payback period from memory (these are the numbers the Week-9 pitch will live or die on).

26.6.3 Things to do before Monday of Week 9

By Sunday evening of Week 8, in addition to the deliverable submission:

Schedule the Week-9 pitch slot with the unit instructor confirmed; rehearsal slots booked for Wednesday and Thursday.
Identify the team’s mock VC panellists (academic + practitioner) for Week 10; this is typically pre-arranged by the unit instructor but the team should know who they will face.
Read Chapter 8 (Retail and e-commerce) or Chapter 13 (Agentic AI) — either provides relevant context for pitching frontier-product stories — and §27.1–§27.3 of Chapter 27 (The pitch and funding) before Monday of Week 9.

References for this chapter

Pricing theory and practice

Smith, T. J. (2016). Pricing Done Right: The Pricing Framework Proven Successful by the World’s Most Profitable Companies. Wiley.
Ramanujam, M. and Tacke, G. (2016). Monetizing Innovation: How Smart Companies Design the Product Around the Price. Wiley.
Van Westendorp, P. (1976). NSS-Price Sensitivity Meter (PSM): A new approach to study consumer perception of price. ESOMAR Congress Proceedings.
Patrick Campbell and ProfitWell (2018–2024). Pricing strategy research and case studies. profitwell.com.

Unit economics and SaaS finance

Skok, D. (2014). SaaS metrics 2.0: A guide to measuring and improving what matters. forEntrepreneurs Blog.
Bessemer Venture Partners (2024). State of the Cloud Report. bvp.com.
Battery Ventures (2024). Software 2024.
Tunguz, T. (2014–2026). Tomasz Tunguz Blog. tomtunguz.com.

Market sizing methodology

Moore, G. A. (1991). Crossing the Chasm. HarperBusiness.
Moore, G. A. (1995). Inside the Tornado: Marketing Strategies from Silicon Valley’s Cutting Edge. HarperBusiness.
Mullins, J. and Komisar, R. (2009). Getting to Plan B: Breaking Through to a Better Business Model. Harvard Business Review Press.

Go-to-market motion

Ross, A. and Tyler, M. (2011). Predictable Revenue: Turn Your Business into a Sales Machine with the $100 Million Best Practices of Salesforce.com. PebbleStorm.
Ahmed, K. (2020). Product-Led Growth: How to Build a Product That Sells Itself. Pendo.
Bush, M., Westerman, G., and Bagchi, K. (2024). Founder-led sales: When and why. MIT Sloan Management Review.

Bowling-pin segmentation

Moore, G. A. (1995). Inside the Tornado. HarperBusiness. (Bowling-pin metaphor as the segmentation discipline.)
Christensen, C. M. (1997). The Innovator’s Dilemma. Harvard Business Review Press.

Cases referenced in §26.3

Iansiti, M. and Lakhani, K. R. (2020). Competing in the Age of AI. Harvard Business Review Press.
Stripe (Collison, P. and J., interviews and engineering blog 2011–2025).
Slack Technologies. (2014–2018). Public statements and case studies on freemium pricing.
Carsome Group. (2018–2024). Annual reports and investor materials.
Anthropic. (2024–2026). Pricing and product documentation. anthropic.com.
DeepSeek-AI. (2024). DeepSeek-V3 Technical Report. arXiv:2412.19437.