02.01 Demand Insight: TAM_SAM_SOM Market Sizing and Must-Have Validation
Governance metadata | Owner: Product Lead | Last reviewed: 2026-10-22 | Next review: 2027-01-31
Core principle: industry field experience shows that most hardware projects die of "pseudo-demand" and "a market space that only flatters itself". This chapter provides an industry-proven framework that cross-validates in both directions — Top-Down and Bottom-Up — and adds category crawler market pools, Google Trends search-weight calibration, and Mom Test quality checks to keep the project kickoff data solid and self-consistent.
1. Three-Layer Market Sizing Model (AstroNext Primary Case Worked Recalculation)
Market sizing must never rely on the macro report of a single consulting firm; it must be micro-verified in both directions — Top-Down and Bottom-Up.
【TAM Global Telescope and Astronomical Observation Market】➔ approx. USD 400 million (all-category pool)
│ Exclude traditional pure-optical telescopes; focus only on the "smart motorized control + integrated imaging" niche
▼
【SAM Serviceable Addressable Market】➔ approx. USD 72 million (18.0% of TAM)
│ Combined with the team's year-1 supply chain first-phase capacity limits and the core e-commerce channels of the US/Europe focus markets
▼
【SOM Serviceable Obtainable Market】➔ approx. USD 10.8 million (15.0% of SAM)
│ Converted at the suggested retail price (MSRP) USD 399.00
▼
【Year-1 Target Sales Volume】➔ approx. 27,000 units/year (equivalent to approx. 2,250 units/month)1. Bottom-Up Category Crawler and Micro-Calculation Model
The bottom-up micro-calculation is based on real sell-through crawler data from Amazon (Amazon BSR), the DTC site, and vertical channels (via Helium 10 / Keepa / JS aggregation), combined with a search-trend weighted calculation:
SOM monthly target per channel = channel category monthly sales capacity (M_k) × price-band slice share (P_k) × search heat factor (S_k) × our niche penetration (Share_k)
SOM year-1 total target units = (∑ expected total year-1 sales per channel) × capacity constraint factor (C_capacity)- M_k (channel category base capacity): the average monthly total sales of the Top 100 ASINs in the core category of the target channel (e.g. Amazon Telescope & Monocular);
- P_k (price-band slice penetration): the share of that category's total sales that falls in the target price range (USD 350~450);
- S_k (search-trend weighting factor): the adjustment factor derived from the annual compound growth rate of search interest for core keywords ("Smart Telescope", "AI Astrophotography") on Google Trends and on the platform's own search;
- Share_k (our share of the price band): the market share a new product is expected to take within the target price band, based on product competitiveness benchmarking (battlecard), launch marketing momentum, and pricing advantage (e.g. 30% on North American Amazon, 25% in Europe);
- C_capacity (capacity and fulfillment constraint factor): the ceiling on maximum deliverable supply that year-1 tooling capacity, ramp yield, and working capital can support (set to 1.0 in this case).
AstroNext bottom-up sizing derivation breakdown:
| Channel segment () | Category monthly sales base () | Target price band share () | Search heat factor () | Expected monthly sales in channel year 1 | Total channel year-1 sales (units) |
|---|---|---|---|---|---|
| North American Amazon main site | 18,500 units/month | 12.0% | 1.15 | 2,553 units/month × our 30% penetration = 766 units | 9,192 units |
| Amazon Europe, 5 countries | 12,000 units/month | 10.0% | 1.10 | 1,320 units/month × our 25% penetration = 330 units | 3,960 units |
| Kickstarter crowdfunding launch | One-off campaign | 100.0% | 1.25 | Concentrated burst over the first 45 days of the campaign | 4,500 units |
| Brand DTC site | Ongoing conversion | 100.0% | 1.10 | Approx. 380 units/month | 4,560 units |
| Overseas offline retail KA | Shelf stocking (Best Buy/B&H) | 8.0% | 1.05 | 400 stores × 1.0 unit/store/month | 4,800 units |
| Total unconstrained theoretical value | - | - | - | - | 27,012 units |
| Capacity-constrained conversion () | Combined with tooling capacity (2,500 units/month) | - | - | Final converged value | 27,000 units |
The 27,000 units derived bottom-up and the 27,000 units derived top-down (USD 10.8 million / USD 399) form a perfectly interlocking closed loop, giving the project kickoff logic end-to-end commercial validation.
2. Value-Chain Profit Pool: Which Layer Captures the Money
Market sizing answers "how big is the pie". The profit pool answers "which layer captures the money in this pie". Both questions must be asked together — a large market whose profit is concentrated in someone else's hands is the classic trap.
1. Five-Layer Profit Pool Breakdown
| Layer | Typical role | Primary profit source | Our position |
|---|---|---|---|
| L1 Upstream components | Chip/sensor/motor makers | Technology barriers and scale | Usually not entered |
| L2 Design & solutions | ID houses / solution providers / ODM | Design premium | Optional |
| L3 Manufacturing & assembly | Contract manufacturers | Scale and efficiency | Usually not entered |
| L4 Brand & channel | Brands / distributors / platforms | Brand premium and channel spread | Primary battleground |
| L5 Service & accessories | After-sales / consumables / subscriptions | Repeat purchase and stickiness | Can be layered on |
Core judgement: we enter L4 (brand and channel) and can layer L5 (service and accessories) on top. L1–L3 usually exist only as cost lines.
2. Profit Pool Map (template)
⚠️ The percentage column must cite a verifiable industry report or be explicitly labelled "assumption"; never fill it from impression. Follow the 00.04 Source Ledger.
| Layer | Industry stage | Typical gross margin | Data source | Our accessibility |
|---|---|---|---|---|
| L1 | Upstream components | Not entered | ||
| L2 | Design & solutions | Optional | ||
| L3 | Manufacturing & assembly | Not entered | ||
| L4 | Brand & channel | Primary battleground | ||
| L5 | Service & accessories | Can be layered on |
3. Three Business-Model Choices
The profit distribution directly determines the business model, and the business model determines the valuation logic of 02.03 Hardware + SaaS:
| Model | Which layer we earn | Precondition | Risk |
|---|---|---|---|
| Build a brand | L4 + L5 | Differentiated selling points and channel capability | Long brand-building cycle, heavy upfront investment |
| Contract manufacturing / ODM | L3 + L2 | Manufacturing capability and cost advantage | Thin margin, customer concentration |
| Distribution / channel | Channel spread within L4 | Local network and capital | Inventory and payment-term risk |
4. Decision Rules
- If L4 gross margin is materially below L3 and we have no brand differentiation → building a brand is a negative-expectation move; evaluate ODM or distribution instead;
- If L5 (accessories/service/subscriptions) is a large share with strong repeat purchase → the L4 + L5 combination holds; move to 02.03;
- If L1 is monopolised by a single supplier with strong bargaining power → the profit ceiling of that category is set upstream; never estimate your own profit from downstream scale.
Relationship to 02.08: 02.08 Competitor Teardown reveals the real cost structure of L1–L3, which is what gives this profit pool map its evidence. Together they answer "is this business worth doing".
3. Three-Dimensional Demand Strength Scale and the Mom Test Interview Defense Line
1. Three-Dimensional Pain-Point Strength Scale (Painkiller vs Vitamin)
Before entering product definition, a product must pass the following 3 hard filters; disguising an "itch point/vitamin" as a "painkiller" is strictly forbidden:
- Pain-point strength (is it a painkiller?): are users currently solving the problem in an extremely clumsy, expensive, or time-consuming way (e.g. traditional telescopes require tedious polar-axis calibration, 30 minutes of manual star-finding, and freezing fingers);
- Willingness to pay (WTP) validation: reach a 3%~5% real payment conversion rate through crowdfunding deposits (USD 1 or a USD 10 VIP reservation voucher), proving a purchase commitment made with real money;
- Technical deliverability: whether core optical components and image sensors have mature supply chain support, and whether the whole-unit hardware BOM cost can be strictly held within the 30.0% of MSRP red line.
2. The Mom Test Anti-Fraud Rules for Real Demand Interviews
When product managers run early user interviews, they are most easily fooled by users' "polite praise" (for example, "this product is so cool, I'd definitely buy it if it existed"). The three iron rules of The Mom Test must be followed:
| Interview landmine (never ask) | Fraud trigger | Scientific question (must ask) | Evidence extraction standard |
|---|---|---|---|
| "Do you think our AI auto star-finding telescope is good?" | Leads users to flatter out of politeness | "When was the last time you went outdoors to look at the stars? What equipment did you bring?" | Recover the real, concrete life context and usage frequency |
| "If there were such a product, would you buy it for USD 399?" | A hypothetical commitment that costs the user nothing | "Over the past year, how much did you spend on equipment, accessories, or software to see Saturn's rings clearly?" | Look for real, already-incurred spending bills |
| "How would you feel about adding a night-vision monitoring feature?" | Encourages casual wish-listing and endless feature inflation | "How do you store the telescope when you are done using it? Which step drives you the most crazy?" | Dig out unmet, repeatedly occurring real frustration |
4. Commercial Project Definition Checklist
[ ] 1. The TAM/SAM/SOM estimate is cross-corroborated by authoritative third-party data sources (such as Statista / IDC / Grand View) and Amazon BSR crawler data.
[ ] 2. The deviation rate between the Top-Down and Bottom-Up calculations is controlled within ±15%.
[ ] 3. The description of the target audience's primary use case (Job-to-be-done) does not exceed 30 words.
[ ] 4. At least 20 in-depth Mom Test interviews with the target audience are completed, with verifiable historical spending and frustration recordings/transcripts.
[ ] 5. The estimated BOM cost as a share of the suggested retail price (MSRP) is strictly held within the ≤30.0% red line.
[ ] 6. Core patent FTO (freedom to operate) search has passed, with no unavoidable infringement landmines (aligned with 07.03 Smart Hardware Patent Strategy and FTO).
5. Hardware Architect Counter-Defense Line: Demand Inflation Circuit Breaker and Core Component EOL Defense
At the very start of hardware product definition, the most common fatal disease of the product manager (PM) and the R&D team is "unchecked feature stacking (Feature Creep)": "the competitor has that sensor, so we must too", "the algorithm needs compute headroom", "industrial design must push to ultra-thin". The result is a sharp increase in board area, the PCB stack being forced from 4 layers to 6, and uncontrolled heat and power draw — and the BOM computed before tooling is cut punches straight through the profitability floor.
1. The "Blind Double-Gate Rigid Circuit Breaker" on BOM Cost and Board Area
- Double-blind hard red lines: before EVT (Engineering Validation Test) kicks off, if the estimated whole-unit hardware BOM cost exceeds 30.0% of the target MSRP, or the single-board PCB footprint exceeds the mechanical stack height/width threshold by 85.0%, the project definition system automatically forces a circuit breaker! Special sign-off by any level of management is strictly forbidden; the project must be returned to the spec definition pool for rework;
- Mandatory de-scoping SOP: when the circuit breaker trips, it is strictly forbidden to mask the problem with "we will raise the future price". A brutal "last-placed feature elimination system" applies: rigorously distinguish the "user's core painkiller (Painkiller)" from the "nice-to-have vitamin (Vitamin)", and unconditionally cut the bottom 20% by contribution — peripheral interfaces, redundant sensors, and non-core structural parts.
2. Pin-to-Pin Dual Sourcing and EOL Lifecycle Defense for the Core MCU
- Pin-to-Pin backup rule for critical parts: for the main SoC, power management IC (PMIC), memory (DDR/Flash), and other core long-lead-time components, the schematic design stage must mandate at least one second-source part number with a fully pin-compatible footprint (Pin-to-Pin Drop-in Replacement). Schematics without a second-source backup must not be allowed to pass the 01.08 GR2 Gate review;
- Chip 3~5 year manufacturer longevity commitment: at chip selection, the procurement team must obtain a life-cycle guarantee letter bearing the manufacturer's official seal from the manufacturer or a tier-1 distributor; selecting niche parts already at end-of-life with a lead time greater than 16 weeks is strictly forbidden, so that the company is never forced six months into mass production to spend heavily on re-spinning boards because a chip was discontinued.