12–14. The three assumptions the model rests on, other verified items, and the verdict

12. The three assumptions the whole model rests on

Ranked by how much of the outcome they control, with what each costs to verify for real.

1. That liability insurance is obtainable at all, at a price near $24,000/yr

Why it dominates. It is the largest single fixed cost, it is 100% unverified, and its
plausible range ($11,440 → $54,120) is wider than the entire mature EBIT. At the low end
the business reaches EBITDA break-even in Year 3; at the high end it never does at any
volume one van can serve. This is not a line item, it is a go/no-go gate, and it should
be resolved before a dollar of capex.

The specific risk is declinature, not price — three or four markets refusing
"unaccompanied riders drinking alcohol," leaving one Lloyd's option at a take-it-or-leave-it
number.

Cost to verify: $0 and 2–4 weeks. Three broker submissions:
ASR Underwriting (the only
market found naming winery tours, cycling and bicycle hire in published appetite),
Affinity Insurance Brokers / Aon on 1300 130 535 (Beazley $20m adventure binder), and a
Lloyd's coverholder via a specialist broker. Ask Yarra Valley X who placed theirs — they
hold $20m at Lloyd's for the guided version of this exact product.

Go in with the mitigation package already written (§5.7 in the insurance chapter). The
difference between a quote and a declinature is likely to be the documented
refusal-of-service policy, not the risk itself.

2. That one van can serve 14+ riders/day, including on a bad day

Why it matters. Break-even is 14.3 riders/day against a practical ceiling of ~18 — only
21% of headroom. If the true ceiling is 10, the price must rise to $229+ or the business
needs a second van at 12 riders/day, adding $53,674/yr. And the binding constraint is not
the average day, it is the wet Saturday when six groups want extraction in the same hour
(§8.3 in the capacity chapter).

Barossa makes this harder than Clare, because there the van is an access vehicle consuming
half its productive day on scheduled shuttle legs before any rescue (§8.4).

Cost to verify: ~$400 and one weekend. The ride-and-drive test in the pilot plan,
with three amendments: (a) do it on the Barossa access variant too — timed shuttle
legs to Seppeltsfield, Yalumba and Rockford, not just circulating; (b) simulate the
surge, not the average — six simultaneous pickup requests at 3 pm, and measure the
last group's wait, not the median; (c) log how long a bike-plus-rider load actually
takes, because that number drives everything in the capacity chapter.

Pass: last-group wait under 45 min in the surge test. Fail: over 60 min, or the loop
cannot be driven continuously.

3. That ~944 riders/year is sellable at $189 or more

Why it matters. 944 riders/year is break-even, and the model does not reach it until
after Year 3. Against Barossa's ~31,000 booked weekend short-let nights (AirDNA: 375
listings, 46% occupancy — VERIFIED) that is ~1% penetration, which is plausible — but
plausible is not proven, and the Year 1–3 ramp is where the $240,000 gets spent.

Cost to verify: $2,500–4,000 and 6 weeks. A landing page with real dates, real prices
and a $50 deposit, plus $500–1,000 of ad spend. Ten paid deposits before a single bike
is bought. Deposits, not survey answers — the concept already has this right.

But run it as a price test, not a demand test. Split traffic $189 vs $229 and
measure conversion on both. Price is the dominant variable by 1.6× over volume,
and the market evidence says $189 is at the bottom of the unbundled range. If $229
converts at even 70% of the $189 rate, price it at $229 and the payback period halves.
This single test is worth more than every other number in this document.

13. Other verified items worth recording

Tax and structure

• GST threshold $75,000; tours are taxable supplies; input tax credits available on
  van, bikes, fuel — VERIFIED
• ASIC: company registration $636; annual review $342; business name $47/yr
  or $108/3yr — VERIFIED (ASIC fees)
• Company tax, base rate entity: 25% (VERIFIED for 2025-26; ATO publishes no 2026-27
  table yet — 2026-27 ASSUMED unchanged)
• ⚠️ INSTANT ASSET WRITE-OFF HAS LAPSED. The $20,000 IAWO applied to assets first used
  1 July 2023 – 30 June 2026. No limit is published for 2026-27; absent new
  legislation it reverts to $1,000 — VERIFIED
  (ATO).
  A $52,500 bike fleet therefore goes into the small business pool at 15% in year 1, 30%
  thereafter — a $7,875 deduction, not $52,500. If a $20k IAWO is re-legislated and
  each bike is individually under the limit, the whole fleet expenses immediately: a
  ~$10,600 year-1 cash swing. Model both. (Academic while loss-making — losses carry
  forward — but it matters from the first profitable year.)

Licensing and permits

Market sizing — VERIFIED

⚠️ Do not treat 100,000 as an addressable pool. The Committee's own report notes these
are sensor passes, not unique users, and that a counter near Clare records ~4× the
traffic of one further north — i.e. it is substantially local exercise traffic.

The site trade-off in one line: Barossa has the market (4.6× the short-let supply,
2.5× the day trips, 1 hour from Adelaide); Clare has the product (car-free 33 km trail,
12–15 cellar doors within 1 km, verified usage, growing visitation) — and per the
capacity chapter, Clare's trail does for free what Barossa's van must be paid to do.

14. Verdict

The draft's core insight survives. Its numbers do not.

The shared roving van across independent groups is a real and defensible idea, and nothing
in this research found anyone running it. The relocation to South Australia is correct on
every dimension checked: route safety, regulatory burden (free driver accreditation, no
metro levy, a 12-seat ceiling), liability geography (SA has statutory intoxication
presumptions; Victoria has none), and free ATDW distribution.

But at $189 and 18 riders/day the business returns ~$26,500 of EBIT, not $103,000 of
contribution. It needs ~$240,000 of funding, does not break even inside three years,
and pays back beyond Year 9. It is an owner-operator job that requires a quarter of a
million dollars of capital to create.

Three things change that verdict, and they are cheap to test:

1. Price at $229, not $189. Tiered analogue/e-bike at $189/$229 with a 53% e-bike
   attach lifts mature EBIT to $49,439 — 1.9× — and cuts break-even to 10.7 riders/day.
   Every market anchor found says there is room: the unbundled components cost $180–290,
   and TrailHopper already charges up to $189.95 for the van alone.
2. Do not buy bikes in Year 1. Partner. It saves ~$63,000 of Year-1 cash and removes
   the hardest line from the insurance submission in the year it is hardest to place.
3. Get the insurance indication before anything else. It is free, it takes a month,
   and it is the only line in this model that can be a zero.

If insurance comes back above ~$45,000/year, the business does not work at any price or
volume one van can serve, and that should be the end of it.