"What's the TAM?" is the first question every investor asks and the first number most ground-transportation decks get wrong. Market sizing in this sector fails in predictable ways — scope creep, double counting, and forecasts borrowed from documents that were themselves borrowed. This guide walks through how to size a ground-transport market properly: TAM, SAM, SOM, and the data sources that survive due diligence.
TAM, SAM, SOM — in transfer-market terms
- TAM (Total Addressable Market): all spending on the service category in your geography — e.g., all pre-booked passenger ground transport in Europe. Useful for narrative, dangerous for planning.
- SAM (Serviceable Addressable Market): the slice your model can actually serve — say, airport transfers in the eight countries where you hold licences, in the vehicle classes you run.
- SOM (Serviceable Obtainable Market): what you can realistically win given competition, capacity and channel access — the only number that belongs in a financial plan.
The discipline is refusing to let TAM logic (big number, everything counted) leak into SOM claims (your revenue plan).
The double-counting trap
Ground transportation is a layered value chain: platforms take bookings, operators run vehicles, aggregators resell both. Add "platform market" revenue to "operator market" revenue and you have counted the same trip twice. Decide the layer you are sizing — gross booking value, operator net revenue, or commission pool — and hold it constant across every source you cite. Half the inflated market sizes we review are layer confusion, not bad data.
Bottom-up beats top-down in niche transport
Top-down ("global mobility is $X trillion; our segment is surely 1%") collapses under the first serious question. The defensible route is bottom-up from observable units:
- Demand base: airport passenger volumes (official statistics), hotel nights, or corporate trip counts — pick the unit closest to your service trigger.
- Capture behaviour: what share of that base pre-books ground transport — from airport access surveys, platform data, operator interviews.
- Value per use: route-level pricing data by vehicle class, collected consistently (see our pricing index methodology).
Each layer has a named source and an error band; the product of the three is a market size you can defend line by line.
Data sources that survive diligence
| Layer | Defensible sources | Weak substitutes |
|---|---|---|
| Demand base | Airport/aviation authorities, Eurostat, national statistics | Press releases, blog statistics |
| Mode share | Airport access studies, operator data, dedicated surveys | "Industry consensus" without citation |
| Pricing | Systematic route-basket collection, booking-platform data | A handful of screenshot quotes |
| Competitive structure | Licensing registries, interview programmes, country reports | Company marketing pages |
Forecasting without fiction
A ground-transport forecast is three drivers in a trench coat: passenger-volume growth (take official aviation forecasts, not your optimism), mode-shift trends (pre-booking penetration, app substitution — slow-moving, evidence-based), and price evolution (inflation pass-through, supply caps from licensing). State each driver separately with its source; let readers swap assumptions. Single-line CAGRs with no decomposition are how decks die in diligence.
When to buy the sizing instead of building it
If the market you are sizing is core to a funding round or expansion decision, an independent report does two jobs your spreadsheet cannot: it provides a citable third-party reference, and it carries interview-layer context you have no access to. Our country and regional reports publish the full bottom-up construction — demand base, capture, pricing — so you can lift the framework and defend it as your own diligence.
Frequently asked questions
What error band is acceptable in a market size?
±20–30% is honest for niche transport segments; tighter claims usually signal false precision. What matters in diligence is that the band is stated and its drivers are named.
Should ride-hailing be inside an airport-transfer TAM?
Only if your product competes for those trips. Best practice: size pre-booked transfers as core, show app-based airport rides as an adjacent layer with explicit overlap assumptions.
How current must the demand base be?
Use the latest full-year airport statistics and the current official forecast vintage; mixing 2023 volumes with 2026 prices is a classic silent error.
