Distribution Channel Economics of a Mid-Size Travel Agency: OTA vs B2B vs Direct Margins in Booking Data

Distribution Channel Economics of a Mid-Size Travel Agency is a quantitative case study of where the money actually is in a travel agency that sells through four channels at once: direct walk-in and phone bookings, direct digital orders from repeat clients, listings on the two major Indian online travel agencies (referred to as OTA-A and OTA-B in the report), and wholesale business from B2B consolidators. The site is Skyline Holidays, a fictionalised but realistic twelve-staff, IATA-accredited agency in Jaipur that mixes retail air ticketing, hotel bookings, outbound packages and visa-and-insurance ancillaries.

The study extracts a full year of the agency's booking register — roughly 3,800 bookings with revenue, net cost of supply, channel fees, cancellations and staff handling time — anonymises it, and builds a contribution-margin model per channel in Excel: for every booking, revenue minus net cost, minus the channel's explicit costs (OTA commission, payment-gateway MDR, BSP and GDS transaction fees), minus an estimate of staff handling time. On top of the model sit a cost–volume–profit break-even for the agency as a whole, a Pareto analysis of revenue concentration, and a channel-mix trend across the Indian travel season. Five semi-structured interviews with the owner, the retail manager and the OTA desk staff explain why the numbers look the way they do.

The result is a viva-defensible answer to a question most agencies settle with folklore: which channel earns the agency its living, and what a rational channel strategy looks like for a tier-two Indian firm.

Mid-size Indian travel agencies sell through four distribution channels at once — direct retail, direct digital, OTA marketplaces and B2B consolidators — each with a different commission structure, fee load, cancellation behaviour and handling effort. Yet most of them steer by top-line sales and habit: the owner of my study site, like many of his peers, could not say which channel actually contributed profit per booking once OTA commissions, gateway MDR, BSP fees and staff time were counted, or how many bookings a month the agency needed to break even under its current mix. Decisions about dropping or courting channels were being made on folklore. There is a need for a booking-level, data-based analysis of channel economics — margins, costs, cancellations and concentration — that a small agency can actually afford to run, understand and act on.

Abstract

Indian travel agencies have lived through a structural shock: base airline commission has been zero for over a decade and a half, OTAs dominate air search, and consolidators now wholesale hotel inventory to agencies and to each other. This study quantifies what that means for one mid-size agency in Jaipur by analysing a full year of booking-level data across four distribution channels, computing contribution margins and break-even per channel, and triangulating the numbers with interviews.

Introduction

A mid-size Indian agency is simultaneously a retailer, an OTA supplier and a B2B buyer. It issues tickets in the IATA BSP settlement system, marks up hotel net rates, packages outbound tours and sells ancillaries such as visa assistance and insurance. Each of these activities earns money differently — and costs differently, once OTA commissions, payment-gateway MDR and staff handling time are counted. During my training at Skyline Holidays I noticed the owner judged channels by top-line sales, not by what each booking actually contributed. That gap between what is counted and what matters is the subject of this project.

Existing understanding (the folklore) vs this study (the data)

• Folklore: OTAs are killing agencies; direct clients are everything; packages are where the profit is.
• This study tests all three claims on 12 months of real transaction data: contribution per booking by channel and product, cancellations and refund-processing costs by channel, share of bookings and share of profit by channel, and monthly seasonality of the mix.
• What the data shows (headline findings): OTA air bookings are high-volume but near-zero-margin after fees; hotel packages sourced from B2B consolidators carry the best gross margins but demand working capital and carry cancellation risk; repeat direct clients generate the largest share of profit on the smallest share of bookings; and ancillaries — visa assistance, insurance, cab add-ons — quietly stabilise the margin on every channel.

Proposed system of analysis

• A booking-level contribution-margin model: contribution = revenue − net cost of supply − channel fees − estimated handling cost.
• A CVP break-even for the agency: monthly fixed costs (rent, salaries, BSP bank-guarantee charges, software) against average contribution per booking, by channel mix.
• A Pareto analysis of client and channel concentration, and a channel-mix trend across the October–March peak and the summer outbound season.
• Five interviews with the owner and managers to explain the patterns and to pressure-test the recommendations.

Feasibility

• Technical: everything runs on Excel — pivot tables, contribution formulas and charts; no special software is required, which makes the study exactly reproducible in any agency.
• Economic: the study costs nothing; the agency already keeps the data in its register and accounting software.
• Operational and ethical: the owner's written permission was obtained; client names, phone numbers and passport details were replaced with codes before any analysis left the office; staff interviewees signed a simple consent note. The agency received the full report and the channel-strategy recommendations before submission.

In scope

• One mid-size, IATA-accredited agency in Jaipur, studied across one full financial year of bookings (≈3,800 records).
• Four channels: direct walk-in/phone, direct digital, OTA marketplace listings, B2B consolidator wholesale.
• Four products: air tickets, hotel bookings, tour packages and ancillaries (visa, insurance, transfers).
• Contribution margin, CVP break-even, Pareto concentration, seasonality of the mix.
• Five semi-structured interviews with the owner and two managers, plus OTA desk staff.

Out of scope

• The agency's corporate-credit accounts above the retail register (confidential contract rates), included only as an aggregate line.
• Competitive benchmarking against other Jaipur agencies — no access to their books.
• Marketing-effectiveness measurement (ad-spend attribution) beyond what the interviews reveal.
• Tax computation: GST and TDS are treated as known context, not analysed line by line.

Research design: a single-case, mixed-method study — quantitative analysis of one financial year of booking-level data, explained by qualitative semi-structured interviews. The unit of analysis is the distribution channel; the unit of observation is the booking.

Instruments: the booking-extraction template, the channel-cost sheet (OTA commission rates, gateway MDR, BSP/GDS fees), a two-week handling-time observation log, and the semi-structured interview guide (eight open questions on pricing, channel choice and cancellations).

Analysis plan: contribution per booking is computed channel-wise and product-wise; medians are reported alongside means because air-ticket values are skewed; break-even bookings per month = fixed costs ÷ average contribution per booking under the current mix; Pareto charts quantify the concentration of revenue in clients and channels; the channel-mix trend is plotted month by month across the peak and lean seasons. Interview themes are coded against the four claims the data tests, and any disagreement between numbers and narratives is treated as a finding, not an embarrassment.

This is a research study, so the architecture below is the study's data-and-analysis pipeline: how raw bookings become a defensible channel strategy.

1. Data layer — the anonymised booking register (one row per booking: channel, product, revenue, net cost, fees, cancellations) plus the fixed-cost sheet for CVP.
2. Model layer — the Excel contribution-margin model and break-even calculation, with every rate (OTA commission, MDR, BSP fee) stored on a separate assumptions sheet so it can be updated without touching formulas.
3. Evidence layer — Pareto and channel-mix charts, cancellation-cost tables, and coded interview notes kept side by side so each numeric finding has its explanation next to it.
4. Decision layer — the channel-strategy recommendations, each traced to at least one table and one interview theme.

flowchart TD
  A[Permission and anonymisation codebook] --> B["Extract 12 months of booking-register data"]
  B --> C["Clean and code: channel, product, revenue, net cost, fees"]
  C --> D["Contribution-margin model per channel in Excel"]
  D --> E[CVP break-even for the agency]
  E --> F[Pareto and channel-mix analysis]
  B --> G["Five semi-structured interviews with owner and managers"]
  G --> H[Thematic coding of pricing and channel decisions]
  F --> I[Triangulate numbers with interview themes]
  H --> I
  I --> J[Channel strategy recommendations and final report]

The pipeline is deliberately boring and reproducible: any student with permission, the register and Excel can rerun it, and every recommendation in the final chapter points back to one table and at least one quotation.

• Repeat the model next financial year to measure whether the recommendations moved contribution per booking.
• Add client-lifetime-value analysis for the direct repeat segment, the agency's clearest long-term asset.
• Extend to a multi-agency comparison across three Jaipur agencies with different channel mixes.
• Model package-pricing scenarios with demand elasticity for the outbound season.
• Build the workbook into a small monthly dashboard the owner can run without a student.
• Study the same question for a destination-management company, where B2B dynamics dominate.