Segment Your Market and Choose Where to Compete

You are a market strategist for an Australian small business that currently sells to 'whoever comes' — and pays for it in scattered marketing and mushy pricing. Your job: cut its market into real segments, score them with evidence, and force a choice. Segmentation without a choice at the end is just a diagram.

Details:
- Business and offer: [BUSINESS — e.g. 'Fairweather Joinery — custom cabinetry, Coffs Harbour']
- Everyone served in the last year or two, roughly grouped: [CURRENT MIX — e.g. 'renovating homeowners, a couple of builders, café fit-outs, one-off repair jobs']
- Job value and pleasantness by group, honestly: [ECONOMICS — e.g. 'builder jobs steady but thin margin; reno kitchens $20k+, best margin; repairs barely worth the ute trip']
- Capacity and constraints: [CAPACITY — e.g. 'one workshop, 2 tradies; ~3 big jobs a month max']
- Where current work comes from: [CHANNELS — e.g. 'word of mouth, Google, one builder relationship']
- The owner's honest preference: [PREFERENCE — e.g. 'loves the design-heavy reno work, tolerates builders']

Before segmenting, set the rule: a segment is real only if the group shares (a) a similar trigger for buying, (b) similar things they weigh up, and (c) a way to reach them as a group. Groupings that fail (c) are demographics cosplay — note any in [CURRENT MIX] that fail.

Then:
1. Draw 3-5 segments from [CURRENT MIX], each defined as: name, the buying trigger, what they weigh up, deal size and margin from [ECONOMICS], and how they're reachable (from [CHANNELS] or an obvious extension). Segments must be mutually distinguishable in one sentence.
2. Score each segment 1-5 on: margin attractiveness ([ECONOMICS]), fit with [CAPACITY] (do these jobs fit the workshop and the 3-a-month reality?), reachability, defensibility (would this segment prefer us over rivals for a reason we control?), and owner energy ([PREFERENCE] counts — burnout is a business risk). Show the grid with one-line justifications; no invented market sizes — where attractiveness depends on an unknown (how many café fit-outs a year exist locally), mark [VERIFY: how] rather than guessing.
3. Make the call: primary segment (the business builds around it), secondary (served opportunistically, not marketed to), and explicitly DE-prioritised (the polite-decline or reprice list — draft the 2-sentence reprice/decline script for the worst of [ECONOMICS]).
4. Cascade the consequences of the choice — this is where segmentation earns its keep: the positioning line for the primary segment; the 3 marketing changes ([CHANNELS] adjusted to where the primary segment looks); the pricing consequence (premium posture for the chosen work, honest reprice for the de-prioritised); and what in [CAPACITY] gets protected for primary-segment work.
5. Set the review: the 2 numbers that test the choice over 2 quarters (share of jobs from the primary segment, average margin), and the tripwire that would reopen the decision.

Format: 'What counts as a segment' → 'The segments' → 'Scoring grid' → 'The call' → 'Consequences' → 'Review'. Under 1,100 words, Australian spelling, decisive tone — the deliverable is a choice, not a matrix.

Rules: segments and scores build only from the supplied details — no invented market statistics, competitor counts or demand figures; every unknown that matters becomes [VERIFY: …] with a cheap way to check. If [ECONOMICS] is missing, stop and ask for rough per-group numbers first — scoring without economics is astrology. Repricing scripts stay honest and courteous; no advice about refusing customers in ways that could breach consumer guarantees on work already agreed (flag to check obligations on existing commitments with their adviser if relevant).

Copy the block above straight into Any AI tool — anything in [BRACKETS] is yours to fill in.