Van Economics — what shared support vehicles cost and sell for

Agent scan, 2026-08-12, ~45 operator pages across MTB uplift shuttles, gravel-race
SAG, charity-ride support and pro neutral service. The closest real-world
analogue to the roving van. Destined for FreeIdeaStore as kind=evidence.

1. Nobody anywhere promises a response time

Across ~45 operator pages, not one response-time commitment exists. The
closest statements found:

• Michigan Coast to Coast, on its $100 rescue tier: "you may have to wait
  quite a while to be picked up." An anti-promise, published.
• 3 Feet Cycling: "we will do our best to track all riders."
• RAGBRAI: manual hailing — "turn your bike upside down ('wheels up') on the
  left side of the road" plus a thumbs-down.

This cuts both ways and both ways matter. A guaranteed response window is
genuine white space — no incumbent to benchmark against, and it would be the
sharpest differentiator the product could carry. It is also unpromised by
everyone who has ever tried to run one of these, which is not a coincidence.
Treat "we'll be there in 20 minutes" as the single riskiest sentence the
business could print.

2. Shared shuttle operators sell frequency, not seats

Capacity appears in exactly four places across all 45 pages. Everyone else
markets dispatch rate instead: "up to 5 vans running up the hill"
(Up Down Around, Derby), "continuous
uplift 10–4" (Dyfi), "every 30
minutes" (Yuba Expeditions, event week).

The product being sold is wait time, not a seat. That is a direct
product-design instruction: price and market the van on how often it comes past,
not on how many people it holds.

3. Vehicle size is the biggest lever in the model

Mountain Bike Rotorua publishes its
own sharing curve — the best single dataset found anywhere:

Going from 5 to 24 seats cuts cost per rider by 62% while total vehicle cost
rises only 83%. The draft model assumed a 1-tonne van. If the van's real
job is access — hopping riders to off-trail cellar doors, which is a Barossa
hypothesis and not an established finding — a larger vehicle may be both
operationally necessary and dramatically cheaper per head. Model at three sizes.
Note the finance model's counter-finding: on the Barossa an access van burns
half its productive day on shuttle legs, which is why Clare's car-free trail
wins on economics.

Corroborating: Adrenalin Uplift prices
private hire of a 16-seat bus at £690 against a £46 public day pass —
£43.13/rider at full load, implying a ~94% assumed load factor and effectively
zero private-hire premium.

4. Shared rescue is cheap wholesale and sells dear

IRRC
publishes an organiser-side fee schedule: $1,750 for events up to 450 riders
(= $3.89/rider), then $4.00/rider to 1,000, $3.75 to 1,500, $3.50
above. That buys incident command, support crew, rapid-response vehicles,
comms and agency coordination.

Retail for the same category of service:

• UNBOUND Gravel "Crew For Hire" —
  $50 (100-mile) / $85 (200-mile). The crew will "encourage, support,
  and should it come to it, pick up a rider if they need to abandon the course."
  The cleanest unbundled shared-SAG retail price found.
• Michigan Coast to Coast —
  unbundled exactly three ways: full service $200, shuttle only $100, drop
  bag/rescue only $100. Three vehicles on course including a sweeper, room for
  55 riders.
• Train With Nathan — private crew at $190 / $95, ~2.2× the shared price.

A 13–22× spread between wholesale cost and retail price. Riders will pay
$50–100 for shared support that costs $4/head to provide at scale. That is the
margin the business is reaching for, and it is real — but note every one of
these is an event with hundreds of riders on a single course on a single day.
The spread comes from density this product does not yet have.

Supported tours converge globally at $180–265/rider/day

Independent operators on three continents land in the same band. The draft
$189pp sits inside it — reassuring for the price point, though these include
accommodation and meals that this product deliberately excludes.

Australian shuttle rates — the nearest domestic comparables

Derby Bike Shop: $17.50
per uplift, 10-pass $135 ($13.50/run), unlimited day pass $90, 9am–4pm.
Up Down Around: Black Stump all day
$99, 10-uplift card $135, 2-day Super Pass $145.
MTB Express:
2 days $375pp incl. return airport transfers, and notably
"No minimum Numbers Required" / "Solo Travellers OK".
Maydena: 1-day summit uplift $95,
single uplift $35, season $1,450.

Season:day ratio tracks the marginal cost of a seat

Nevis Range gondola 2.9 · Christchurch chairlift 5.8 · Kingdom Trails (no
vehicle) 6.25 · Whistler ~6–7 · Skyline Queenstown 7.9 · Trans Cascadia vans
9.3 · Maydena bus 15.3.

Fixed infrastructure means a near-free marginal rider, so the season pass is
priced as membership. A vehicle seat has a real marginal cost — driver, fuel,
a finite seat — so it prices to need 9–15 days. Directly relevant to any
season-pass or multi-visit product here.

Trans Cascadia states plainly: "our shuttles
don't always go out — we require a minimum of two riders for a shuttle to run",
and four for a guided day. Old Ghost Road's
Hike n Bike Shuttle prices every route
with a headcount minimum (min 2, min 3, min 4).

So quorum dispatch is normal industry practice. But it collides head-on with
GetYourGuide's ≤1.0% supplier cancellation rule (see the
Distribution And Take Rates chapter). Either the product runs
for one couple and skips quorum, or it uses quorum and stays off the major OTAs
— selling direct and through Airbnb Experiences instead. That is a real
either/or the business has to choose deliberately.

A neat resolution exists in the same dataset: Buller Adventures prices per
vehicle — "$250 per trip (includes up to 5 passengers + bikes)" — while
Hike n Bike prices per person with a minimum. Same corridor, $70pp (min 3)
versus $50/head at full load. Per-vehicle pricing eliminates the cancellation
problem entirely by making the small group pay for the empty seats.

Shimano Neutral Service
operates 450 race days in 2026 with 29 cars, 6 motorbikes, 9 material
vehicles, 165 spare bikes, 600 wheels, 52 drivers and 59 mechanics. The article
pointedly never says who pays. Endurance.biz evidences an extended
Shimano–A.S.O. partnership over 560 race days — marketing rights bartered
against service in kind, not a service contract. Mavic exited the WorldTour
~2021, went into receivership, and re-entered in 2026 at a single national
championship.

No cost-per-day figure exists for neutral service and probably never will.
The relevance here is the funding model, not the number: a standing support
fleet can be paid for by someone other than the rider.

CORRECTED 2026-08-12 — this section originally continued: "In this
business the analogous payer is the cellar doors, who benefit from riders
buying wine they don't have to carry." The corpus found [NOT FOUND] on any
winery, anywhere in the world, paying a cycling operator. It is a hypothesis
with no precedent, not an analogy with support. The finance model separately
finds the commission line fails until ~5,000 riders/year. The Wine Australia
29% grant does not rescue it — see the Route Survey chapter
and the Audit And Corrections chapter.

Charity rides make the same point differently — Bike MS SAG vehicles are
volunteer-crewed, so cost per rider to the organiser is roughly zero.
C3 Ride is the only ride publishing an explicit
rider:vehicle structure — "each team on the road has a team vehicle with them
most of the time" — and its support people raise $2,775 each, making the
support vehicle a revenue stream rather than a cost centre.

Whistler season pass · Thredbo MTB day pass · Antur Stiniog standard rates ·
BikePark Wales weekend/annual · Maydena bus seat count · riders-per-van at
every Derby operator, Yuba Expeditions, and every cross-country tour operator ·
any response-time SLA anywhere · organiser cost per day for neutral service.

The riders-per-vehicle gap is notable: it was searched for directly and is
genuinely absent from the entire industry's public marketing, which is itself
consistent with finding #2 — operators don't sell seats, so they don't publish
seat counts.

⚠️ Two claims in this chapter were superseded by later research. It states that "the analogous payer is the cellar doors" — but the wider scan found [NOT FOUND] on any winery anywhere paying a cycling operator commission, referral or funded support. It also states "Given the Barossa finding that the van's real job is access" as established, where the Aggregators And Shared-Cost Support Models chapter downgraded this to "If the van's real job turns out to be access". See the Audit And Corrections chapter.