Market size is the first question asked and among the least useful, because a large market with the wrong structure is worse than a small one with the right one.
The sizing ritual
Total addressable market, serviceable available market, serviceable obtainable market.
Which is a reasonable framework and is frequently used to produce a large number by defining the total broadly.
The figure that matters is what a realistic share of the reachable segment is worth, and that requires assumptions about share that are generally the weakest part of the analysis.
Bottom-up sizing — number of potential customers multiplied by plausible spend — is more defensible than top-down division of an industry figure.
The questions that predict better
Is the problem acute enough that people are already spending money on it?
Which is the most informative single question, since a market where nobody currently spends anything requires creating a budget, and that is far harder than capturing one.
What are people currently doing instead? The alternative is generally a manual process or a general-purpose tool, and understanding it tells you what you must beat.
Who decides and who pays? These are frequently different people with different concerns.
Structure
Concentration on either side determines bargaining power.
A market with a few large buyers means those buyers dictate terms.
A market with a few large suppliers of a critical input means margins are constrained.
Which is the standard framework and it remains useful precisely because it is about structure rather than size.
Barriers
Barriers to entry protect incumbents, which is bad on entry and good afterwards.
Which means a market with no barriers is easy to enter and offers no protection once entered, so competition erodes returns.
Regulatory requirements, capital intensity, network effects, switching costs and accumulated data are the common ones.
Identifying which barrier you would eventually build is part of assessing whether entry is worthwhile.
Switching costs
What it costs a customer to change from their current solution.
Which determines how much better you must be to win, and it is frequently underestimated.
Data migration, retraining, process change, integration work and contractual commitments all count, and together they can exceed several years of price advantage.
Growth and timing
A growing market allows entry without taking share from incumbents, which is far easier.
Which is why timing matters so much — entering as a market forms is easier than entering an established one.
The corresponding risk is entering before the market exists, which is the failure mode of being early.
Identifying what has changed to make the market viable now is the discipline that distinguishes the two.
Adjacency
Entering a market related to what you already do carries advantages — existing customers, capabilities, brand, channels.
Which is why adjacent expansion succeeds more often than unrelated diversification, a finding that is well supported.
The trap is defining adjacency loosely, so that unrelated moves are rationalised as adjacent.
The test to actually run
Attempt to sell before building.
Which produces evidence rather than analysis, and it answers the question sizing cannot — whether these specific people will pay this specific amount for this specific thing.
A market analysis that has never involved talking to a potential customer is an exercise rather than an assessment.
Distribution
How you will actually reach customers, which is frequently the binding constraint rather than the product.
Markets with concentrated distribution channels controlled by incumbents are hard to enter regardless of product quality.
Which is why identifying a viable route to customers should precede product development rather than following it.
Regulatory entry
Licensing, certification and compliance requirements determine time and cost to enter.
Which is a barrier and therefore a protection once cleared, and the time required is frequently underestimated by an order of magnitude.
Regulated markets reward patience and punish businesses that assume approval will be quick.
Testing before committing
Landing pages measuring interest, pre-orders, concierge delivery where the service is provided manually before being built.
Each produces evidence at low cost, and each is more informative than any amount of desk research.
The consistent finding from failure analysis is that this step was skipped.
Competitive response
Incumbents respond to entry, and the response should be anticipated rather than discovered.
Which depends on how much the incumbent has to lose and how easily they can match.
Entering a segment an incumbent regards as core invites a full response. Entering one they neglect frequently does not.
Exit
Worth considering before entering.
Markets requiring specific assets, long contracts or regulatory commitments are expensive to leave.
Which raises the stakes of the entry decision, and it is why testing before committing capital matters most in exactly those markets.
Sequencing
Entering a narrow segment first, establishing a position, then expanding is more reliable than entering broadly.
Which concentrates limited resources where they can matter and produces reference customers that support the next segment.