Market Sizing & TAM Estimator
Structure a transparent market-sizing estimate with the math shown.
What is the Market Sizing & TAM Estimator?
The Market Sizing & TAM Estimator is a free AI skill that structures a total-addressable-market estimate for a food or beverage opportunity, showing the math and assumptions transparently rather than handing over a single unexplained number. You give it the product or category, the geography, and whatever market data points you already have; it returns a top-down estimate using category and population data, a bottom-up estimate built from a unit-economics path (target consumers, purchase frequency, price), a reconciliation of the two when they disagree, and every assumption labeled clearly enough that someone else could challenge or update it later. It is built for insights managers and analysts who need to defend a market-size number in front of finance or leadership, where an opaque figure gets torn apart in the first question. Because every input is shown as a labeled assumption or a cited data point, the estimate survives scrutiny even when the underlying data is imperfect. Pairing the estimate with live food and beverage demand data sharpens the category-growth assumptions the top-down math depends on.
Who it's for
- Insights managers building the market-size case for a new opportunity
- Innovation teams sizing a benefit platform or new category entry
- Strategy and finance partners stress-testing a business case
- Founders estimating TAM for an investor or leadership pitch
What you get
- A top-down estimate built from category and population data
- A bottom-up estimate built from a unit-economics path
- A reconciliation explaining any gap between the two methods
- Every assumption labeled and stated separately from cited data
- A sensitivity view showing how the estimate moves with key inputs
- A one-page summary defensible in front of finance or leadership
How to use it
- 1Paste the prompt into ChatGPT, Claude or Gemini.
- 2Fill in the brackets (product, geography, data points you have).
- 3Answer its clarifying questions about what data you actually have versus need.
- 4Ask it to build a sensitivity table for your two or three riskiest assumptions.
See it in action
Example input
Product: fictional 'Hazelbourne' oat-based coffee creamer, planning a US launch. Geography: United States. Data I have: US households roughly 130 million, and roughly 40% of US households buy coffee creamer of any kind. I don't have the exact oat milk category size handy. I need a defensible TAM estimate for a leadership deck next week.
Example output
Market sizing — Hazelbourne oat-based coffee creamer, US TAM Important flag before the numbers: you noted you don't have the exact oat milk category figure. Every number below is either a data point you gave me, a placeholder you must fill with a sourced figure before this goes in front of leadership, or an explicitly labeled assumption. Do not present this deck with placeholder figures still in it; replace every [INSERT FIGURE] before leadership sees it. Top-down estimate Start from the coffee creamer category as the relevant frame, since Hazelbourne competes for a creamer occasion, not a drink-oat-milk occasion. - US coffee creamer category size: [INSERT FIGURE — pull from Circana/Nielsen or a market report]. - Share of that category realistically addressable by a plant-based/oat entrant: estimate a range, e.g. 8-15% based on current plant-based creamer category share where you have it, or flag as [INSERT FIGURE — confirm plant-based creamer's current % of total creamer category]. - Top-down TAM = category size x addressable share. Until the two bracketed figures are filled in, this stays a formula, not a number; do not estimate the category size itself, since that must come from a sourced number. Bottom-up estimate Built from your own data points: - US households: 130 million (your figure). - Households buying any coffee creamer: 40% of 130M = 52 million households (your figure, multiplied). - Assumption to validate: what share of creamer-buying households would consider an oat-based option; this is the single most important number in the whole model and you don't have it yet. Propose a testable range: conservative 5%, base case 12%, optimistic 20%, based on typical plant-based adoption rates within an existing category (label this range as a benchmark assumption pending your own concept-test data, not a cited fact). - Applying the base case: 52M x 12% = 6.24 million households as the addressable bottom-up base. - Annual spend per adopting household: [INSERT FIGURE — estimate from typical creamer purchase frequency (roughly weekly repurchase is common in the category) x average price per unit; you'll need a price point decision first]. - Bottom-up TAM = 6.24M households x annual spend per household. Cannot finalize without the price and frequency assumptions above. Reconciliation Until both bracketed figures are filled, the two methods can't be reconciled numerically; but structurally, expect the bottom-up number to come in lower than top-down, since bottom-up is built from a narrower, more conservative adoption assumption while top-down assumes the full addressable share converts. A wide gap between the two (more than roughly 2-3x) usually means one of the two share assumptions (8-15% top-down, or 5-20% bottom-up) needs tightening with real data before the estimate is presentation-ready. Sensitivity — the two riskiest assumptions 1. Oat-adoption share of creamer buyers (5/12/20% range): this single variable swings the bottom-up estimate roughly 4x between conservative and optimistic cases; this is the number most worth spending research budget to nail down before the leadership deck, ideally via a real concept or usage-intent survey rather than a benchmark guess. 2. Category size itself (top-down): without a sourced figure, the entire top-down line is unusable as presented; this is a research-budget-zero fix, just pull the number from Circana, Nielsen, or a paid market report. One-page summary for leadership Hazelbourne's US TAM sizing uses two methods that should converge once two data gaps are closed: the actual US coffee creamer category size (a quick data pull) and a real adoption-share estimate for oat-based creamer among existing buyers (best obtained via a concept/usage-intent survey, not a benchmark guess). The bottom-up base case, using your household and creamer-buyer figures with a benchmark 12% adoption assumption, suggests roughly 6.2 million addressable households; treat this as directional until the adoption assumption is tested. Want me to build the sensitivity table across all three adoption scenarios once you have a price point, or draft the specific concept-test question that would pin down the adoption-share number?
The prompt
Here's the start of the prompt. Download the free bundle for the full, ready-to-paste version — plus the installable Claude Skill and Custom-GPT instructions.
# Role You are a senior market-sizing analyst for food & beverage opportunities. You never hand over a single unexplained TAM number; every estimate you produce shows its math, cites what's real data versus a labeled assumption, and survives a skeptical question from finance. # Context I'll provide - Product or opportunity: [PRODUCT / OPPORTUNITY] - Geography: [GEOGRAPHY] - Data points I already have: [DATA — category size, population, purchase rates, whatever you know] - Data I don't have yet (optional): [GAPS] - Audience for the estimate: [AUDIENCE e.g. leadership deck, investor pitch, business case] # Your task 1. If the product, geography, or existing data points are missing or vague, ask up to 3 clarifying questions BEFORE writing anything.
Frequently asked questions
- What is a TAM (total addressable market) estimate?
- A TAM, or total addressable market, estimate is the total revenue opportunity available for a product or category if every reachable customer bought it, typically built two ways: top-down from category and population data, and bottom-up from a unit-economics path, then reconciled. This skill structures both methods with every input labeled, so the resulting number can be defended rather than just asserted.
- How is this different from the Whitespace & Trend Scout skill?
- The Whitespace & Trend Scout maps a whole category to find open positioning territory: where to play, not how big the prize is. This skill answers a different question, once you already know the opportunity: how many dollars is it actually worth, built from transparent top-down and bottom-up math with every assumption labeled. Use the scout to find the opportunity, then this skill to size it before it goes into a business case.
- Can I use this with any AI model?
- Yes. ChatGPT, Claude, and Google Gemini all run it; the prompt is model-agnostic. Many teams save it as a Custom GPT or reusable skill so every new opportunity gets sized with the same transparent top-down and bottom-up method.
- Will it just make up the market size if I don't have data?
- No. Every unsourced number gets marked '[INSERT FIGURE — source needed]' rather than filled with a plausible-sounding guess, and every assumption is labeled separately from cited data. The output is a transparent framework you complete with real figures, not a finished number built on invented inputs.
Related skills
Assortment Gap Finder
Find the SKUs your shelf is missing before a buyer does.
Get it freeBrand Health Report Translator
Turn brand tracker scores into a plain-English story and a next-actions list.
Get it freeBrand Positioning Statement Builder
Nail the positioning your whole brand can build on.
Get it freeWant the live data behind sharper outputs?
These skills get better with real-time F&B intelligence. See what Tastewise can do for your team.