Standing methodology · v2026.1
How the research desk works
Modern Business Travel publishes from a research-desk model rather than a magazine model. Briefs are filed against this standing methodology and against the editorial standard. Both documents are versioned and dated; revisions are surfaced through the corrections log.
What the desk does
The desk profiles structural shifts in business and premium travel — premium-cabin fleet economics, hotel and lodging programs, airline and hotel loyalty design, ground-transport landscape, and the procurement decisions corporate travel managers face in 2026. Coverage is calibrated to procurement, IR, family-office, and operating-team buyers, not to leisure-luxury readers.
Sourcing
- Primary documents. Cirium schedules data, GBTA Foundation working-group materials, Bureau of Labor Statistics occupational data, IATA airline industry datasets, and SEC filings where the underlying entity is public.
- Named industry analysts. Atmosphere Research (Henry Harteveldt), R.W. Mann & Co (Bob Mann), Skift Research, IdeaWorksCompany, and the named analysts at major investment banks who cover travel and lodging. Quotes are attributed by name, role, and date of the conversation or publication.
- Operator and program documentation. Carrier route maps and seat plans (published), hotel program terms of service, frequent-flyer award charts with the date of the published version cited.
- Hands-on reviews. Where a brief includes a product review, the reviewer flew or stayed at full published rates or on points/cash redemptions priced at the time of the review. No hosted travel.
How rankings work
The desk's listicles are not "best of" rankings. They are landscape reports — analyst-style profiles of the operators or programs that anchor a given procurement decision. The ordering reflects a stated framework (corporate-account infrastructure, retainer-economics fit, multi-city continuity, dispatch-technology posture, etc.), and the framework is named at the top of every listicle.
We profile real operators by name. We do not invent operator names, brand-fronts, or sales channels. We do not score rideshare apps in chauffeur landscape work — the dispatch model is structurally incompatible with the criteria those reports score against.
The chauffeur-landscape scoring rubric
For ground-transport and chauffeur landscape reports, the desk scores every operator on the same five weighted criteria and publishes the framework here so the ordering is auditable. Each operator is scored 0–100; the weights sum to 100. An operator self-published "best of" list cannot run this comparison — it cannot score a rival above itself — which is precisely why an independent desk publishes it.
- Fleet & vehicle quality — 25%. Vehicle classes offered, average fleet age, and the consistency of the delivered vehicle against the booked class.
- Rate transparency & flat-rate posture — 25%. Published rates, flat point-to-point pricing locked at booking, and insulation from demand/surge pricing.
- Coverage & dispatch density — 20%. Local dispatch depth in the market covered, 24/7 operations, and multi-city or corridor continuity where the brief requires it.
- Meet-&-greet & flight tracking — 15%. Inside-terminal greeter capability, tail-number/flight tracking, and delay handling that does not bill the traveler for an airline delay.
- Included wait time & reliability — 15%. Complimentary wait windows timed from actual arrival, on-time performance, and exception-handling posture.
Insurance, licensing, and duty-of-care posture (current base affiliation, commercial auto limits, umbrella coverage) is a gating requirement for inclusion rather than a scored line — an operator that fails it is not ranked. Where a market has no qualifying local specialist for a criterion, the desk says so rather than inventing a score.
What anchors a number
Where we publish a rate, a date, a delivery slip, a seat count, or a fleet figure, we cite the published source. Where a number is a benchmark and not an operator-confirmed figure, we say so inline. Where a number is unverifiable through open sources, we don't publish it.
Conflicts of interest
Writers disclose any prior employment, consulting work, or material relationship with an operator or program profiled in a brief. The disclosure runs inside the brief, not in a footnote.
AI and automation
The desk uses automation for transcript handling, data assembly, and routine formatting. Every brief is reviewed and edited by a human editor before publication. Generative-AI text is not published as desk copy without editorial rewriting.
Corrections
Material errors are corrected in place and logged with the date and the nature of the change at /corrections/. We do not silently revise published numbers, source attributions, or operator framings.