ONLY VTC Observatory : 2026-T3 · version 2 · archived

Key figures — edition 2026 · quarter 3 · version 2

Quarterly archive · see the current edition

Cut-off : October 6, 2026. Source: ONLY VTC platform. Each indicator specifies its population.

Acceptance is a driver's response to a broadcast ride, not their arrival. Approach time measures their travel to the meeting point and remains suspended. Passenger waiting time for an immediate request runs from the request to the driver's arrival; it is published only at the methodology's thresholds.

What the data says

As of October 6, 2026, a ride broadcast to ONLY VTC drivers in Lyon is accepted in a median of 45 sec and 53 % of broadcast rides are accepted in under 60 seconds (30 rides measured). Rides assigned directly to a driver by dispatch — almost instant assignment — are not counted in this time.

As of October 6, 2026, no recorded cancellation initiated by a driver after acceptance at ONLY VTC (0 out of 156 rides accepted since 4 June 2026). This observation of the past sample does not guarantee that a driver will come for every future request; administrative reassignments are not driver cancellations in this indicator.

As of October 6, 2026, a ride booked in advance at ONLY VTC is ordered a median of 3 days before the scheduled pick-up (132 rides): at ONLY VTC, journeys are often planned several days in advance — airport transfers and business travel first and foremost.

As of October 6, 2026, 2.1 % of ONLY VTC rides exceed 50 km (distance estimated at booking, out of 94 rides). This share describes the ONLY VTC sample, not the Lyon market.

As of October 6, 2026, 32 % of ONLY VTC rides have an airport as their departure or arrival point (163 completed rides). This rate cannot be used to classify the reasons for booking across all private-hire cars in Lyon.

As of October 6, 2026, 54 % of ONLY VTC rides have a scheduled pick-up between 7 p.m. and 7 a.m. (163 completed rides). This breakdown does not prove continuous 24/7 coverage or the availability of a driver at any time.

As of October 6, 2026, 81 % of ONLY VTC rides are booked in advance (scheduled pick-up more than 30 minutes after the request): on the platform, rides are planned, especially for airport transfers and business travel.

On the same date, 90 % of measured rides are business rides (B2B) and 19 % are assistance missions (repatriation, breakdown, insurance cover). These categories are not mutually exclusive: an assistance mission may also be a business ride. Each percentage covers all completed rides in the period; they should not be added together. These figures describe the actual activity of the ONLY VTC platform, not the entire Lyon private-hire market.

Where rides start — pick-up municipalities

Municipalities with at least 15 pick-ups over the period (aggregates only — no individual address is published):

Rides completed per month

How to cite these figures

Observatory data is published under the licence CC BY 4.0 : press, researchers, blogs and AI assistants may freely reuse it with attribution. Wording to copy:

Source: ONLY VTC Observatory — onlyvtc.fr/observatoire/2026-t3-v2 (CC BY 4.0 data, edition of October 6, 2026, 2026-T3 · version 2 · archived, coverage from July 1, 2026 to September 30, 2026).

• Each sentence in the "What the data says" section is dated and stands on its own: it can be reused as is.
• Each figure comes with its sample size — please keep it in your citations.
• For a lasting figure, cite an archived quarterly edition (permanent URL, frozen content) rather than the daily showcase.
• Exact definitions — populations, bounds, publication thresholds — are in the full methodology, which is versioned.
• Machine format: JSON export · CSV export — licence and attribution included in each file.

What we do not publish yet — and why

Three indicators — approach time, journey duration and airport transfer duration — are defined but suspended in this edition. They rely on milestones the driver enters during the ride ("on the way", "departure", "end"), and our data shows these milestones are too often entered all at once at the end of the ride: on measured rides, "arrival" and "departure" are entered a median of 10 seconds apart, and one ride in five has all three milestones recorded in under two minutes. Publishing a duration calculated from them would show a false figure under a correct label.

However, arrival at the pick-up point is now measured automatically by our servers (120 m geofence around the meeting point), without depending on a driver's tap. The v2 series restarts on 4 August 2026: it makes it possible to reactivate punctuality and immediate-request waiting time from 15 rides and only when automatic measurement covers exactly at least 80% of the relevant rides. The coverage rate is shown next to each figure. Journey duration depends on departure and end milestones, which this measurement does not yet make reliable: it remains suspended.

The indicators published above are those that do not depend on journey milestones: acceptance time, estimated distance, activity shares and volumes.

Methodology

Each ride booked via ONLY VTC (app, website or phone) is automatically time-stamped by our servers at every stage: acceptance by the driver, departure towards the customer, arrival at the meeting point, pick-up, end of ride. Published times are calculated directly from these time stamps, with no manual reprocessing. Distances are road route estimates calculated at booking — and presented as such. Airport or station locations are identified by the place type recorded at booking, supplemented by analysis of the address.

Volumes and journey statistics cover rides actually completed, excluding cancellations and no-shows. The driver cancellation indicator uses a different population: all rides accepted since 4 June 2026, classified by acceptance date. Inconsistent time stamps are discarded according to the published bounds; for durations, automatic or forced closures are also excluded. A statistic is published only from 15 measurable rides, with exact coverage of at least 80% for punctuality and immediate waiting time. Suspended indicators remain suspended. We apply a non-selection rule: as soon as its methodological conditions are met, an indicator is published, even if unfavourable. Over the period, 0 chauffeur hire booking(s) and 3 automatic or forced closure(s) are excluded from journey statistics.

Definitions of published indicators: an immediate request is a ride whose pick-up is scheduled within 30 minutes of the request — beyond that, it is booked in advance, and its booking lead time is the time between the request and the scheduled pick-up. A night ride has a scheduled pick-up time between 7 p.m. and 7 a.m. (Paris time). The acceptance time counts only rides broadcast to drivers; the travel distance is the road route estimate calculated at booking, excluding chauffeur hire, and a ride is long distance beyond 50 km. A driver cancellation is a ride accepted then cancelled by the driver who accepted it — a series measured since 4 June 2026 across all accepted rides, classified by acceptance date; a customer no-show recorded by the driver is not included. The pick-up municipalities are extracted from the departure address and published only in aggregate, from 15 rides per municipality. The punctuality and theimmediate-request wait are based on arrival measured automatically at 120 m on the server side, on the v2 series opened on 4 August 2026, from 15 rides and with exact coverage of at least 80%. A delay is counted from the first minute beyond the 5-minute tolerance, and the average extent of recorded delays is published next to the rate. Between 15 rides and the coverage threshold, the figure already measured is shown as provisional in the card's note — the main value remains "—" and exports only expose consolidated figures. The 120 m measurement may also count as "late" a driver already on site but parked beyond the radius (drop-off zone or station and airport car park): these cases are kept in the published figure rather than removed by hand. Indicators that still depend on journey milestones entered by the driver — approach time and duration — remain suspended until their collection is made reliable.

The Observatory is young: ONLY VTC's digital booking platform (app and website) ramped up in 2026, while the company has been operating since 2016 — earlier rides, taken by phone, are not time-stamped and therefore cannot be included. Current samples are modest and we accept that: rather than waiting for flattering figures, we publish as we go and let the volume shape the trend. Measured activity today consists mostly of business rides and assistance missions — the exact share is published above —: these figures describe the ONLY VTC platform, not the entire Lyon market.

These definitions are frozen and versioned (methodology v1.3): each published edition references the version that produced it, and any change of definition increments the number — a published figure remains interpretable under the rules in force when it appeared. The full methodology — definitions down to the time stamp, bounds, thresholds, version log — is published.

The data is published under the licence CC BY 4.0 : cite "ONLY VTC Observatory — onlyvtc.fr/observatoire" and reuse it freely.

Download the figures in machine format: JSON export · CSV export : the same aggregates as this page, frozen at the cut-off date.

Quarterly editions

Each published quarter is permanently archived, with its sample size and the methodology version that produced it.

• 2026 — quarter 3 · version 2 · 163 completed rides · methodology v1.3
• 2026 — quarter 3 · 63 completed rides · methodology v1.0

Frequently asked questions