TMC vs. self-booking: which model for a mid-sized company in 2026?

French mid-sized companies manage an average of 500 to 2,000 business trips per year. Between the personalized support of a TMC and the autonomy of a self-booking platform, choosing between a corporate TMC or betting on automation has become a strategic trade-off. The traditional model—a dedicated agent for every booking—no longer meets the expectations of many travelers. However, total autonomy exposes companies to compliance risks, cost overruns, and a loss of control. In 2026, the right answer is no longer binary.

The structural dilemma for mid-sized companies

Mid-sized companies do not operate like large corporations. They often lack a dedicated travel department, a formal policy embedded in a tool, or consolidated spending data. Administrative managers or CFOs oversee travel alongside their other duties. As a result, travel management oscillates between two extremes: either everything goes through a TMC agent (maximum control, but slow and expensive), or everyone books directly on sites like Booking or Trainline (fast, but with no visibility or negotiation power).

This siloed approach creates three major points of friction. Travelers waste time waiting for approvals or bookings. Budget control relies on Excel files reconstructed after the fact. And the company fails to leverage data to negotiate preferential rates with suppliers.

Traditional TMCs: real services and hidden costs

A traditional TMC provides a dedicated agent, reachable by phone or email, who manages the entire cycle: research, booking, modifications, and incident support. This model offers three tangible benefits. First, compliance: the agent enforces the travel policy and blocks out-of-policy bookings. Second, time savings for frequent travelers: delegating the logistics of complex trips (multi-destination, trains + flights + hotels) frees up time for core business priorities. Finally, access to negotiated rates: TMCs have agreements with airlines and hotel chains that are inaccessible when booking directly.

But this model comes at a cost. Service fees range from €15 to €40 per transaction, depending on the level of support. For 1,000 annual trips, this represents between €15,000 and €40,000 in pure fees, even before the cost of tickets. The second friction is latency. A Paris-Lyon round trip booked via email can take 24 to 48 hours, whereas a self-booking platform allows for confirmation in 3 minutes. The third limitation is the lack of real-time visibility. Spending data arrives in monthly reports, which is too late to adjust behavior during the month.

The TMC remains relevant for three profiles: frequent international travelers (due to route complexity), sensitive trips (security or representation issues), and companies without internal digital maturity (lacking the resources to manage a tool).

Self-booking platforms: autonomy and structural limits

A self-booking platform (e.g., TravelPerk, Navan, SAP Concur) reverses the model: the traveler books directly via a web or mobile interface, in compliance with rules set by the company. There are three immediate advantages. Speed: booking in a few clicks with automatic approval if it complies with the policy. Visibility: real-time dashboards of expenses by department, traveler, or supplier. And scalability: adding 50 new users does not change the marginal cost.

Modern platforms also integrate compliance features. They automatically block out-of-budget bookings, alert users if they select a higher class, and integrate with expense management tools (like Expensya or Jenji) to automate the reconciliation of expenses and receipts.

However, self-booking exposes companies to three risks. First, the learning curve: some employees (especially the less tech-savvy) take time to adopt the tool and continue to book directly. Second, handling exceptions: a flight canceled at 10 PM, a missed train, or a medical emergency abroad requires responsive human support. Not all platforms offer this level of service (or they charge extra for it). Finally, the quality of rates: without significant volume, a mid-sized company does not obtain the same agreements as it would through a TMC with consolidated bargaining power.

Self-booking is suitable for companies with a population of autonomous travelers (executives, sales staff, consultants), simple travel needs (domestic flights, recurring routes), and an IT department capable of managing integration with existing systems.

Hybrid model: segmenting by complexity and profile

Most high-performing mid-sized companies no longer choose between a TMC and self-booking. They segment. Simple trips (domestic round trips, advance bookings, standard routes) go through the platform. Complex trips (multi-destination, short notice, representation issues) are handled by a TMC agent.

This approach requires three prerequisites. First, a clear policy that defines what falls under self-booking (domestic travel, budget < X, lead time > 7 days) and what justifies TMC support. Second, traveler training: they must understand when to use which channel and why. Finally, data-driven management: a monthly dashboard that measures the self-booking adoption rate, policy compliance, and savings achieved compared to the previous year.

In practical terms, a 300-person company with 1,200 annual trips can structure its model as follows: 70% of trips (domestic flights, Paris-Lyon-Marseille trains) go through TravelPerk with automatic approval if compliant. 20% (international travel, groups, events) are managed by a TMC agent for support. 10% (emergencies, last-minute changes) switch to the 24/7 assistance of the platform or the TMC, depending on the context.

This hybrid model reduces TMC fees by 40 to 60% while retaining human expertise for high-value situations.

Decision matrix: 4 criteria for making the choice

Volume and frequency. For fewer than 500 trips per year, a traditional TMC is often easier to manage (no integration projects, no user training required). Between 500 and 2,000 trips, a hybrid model becomes cost-effective. Beyond 2,000, pure self-booking (with premium support) is the way to go.

Traveler profiles. A workforce of autonomous executives who are comfortable with digital tools will easily adopt a platform. Field teams (technicians, senior sales staff with low digital proficiency) will prefer to delegate to an agent. Measure the actual digital maturity of your travelers before making a decision.

Trip complexity. If 80% of your trips are domestic round-trips booked 10 days in advance, self-booking is sufficient. If you regularly manage multi-country travel, groups, or events, TMC expertise remains essential.

Travel policy maturity. A self-booking platform only works if you have formalized your rules (budgets per trip, authorized classes, booking lead times, approval processes). Without a clear policy, you lose the main benefit of self-booking: automated compliance. If your policy does not yet exist, start by formalizing it with your current TMC before making the switch.

Next step: test before scaling

Rather than a sudden transition, launch a pilot program. Select 20 to 30 representative travelers (a mix of profiles: digital-savvy executives, field sales staff, assistants). Test a self-booking platform for 3 months alongside your current TMC. Measure three KPIs: adoption rate (% of bookings made through the platform), policy compliance (% of compliant bookings), and savings achieved (fees + negotiated rates).

This pilot will provide the real-world data you need to make an informed decision. If the adoption rate exceeds 70% and the savings are there, scale it up. If travelers struggle with or bypass the tool, revert to a hybrid model with more human support. The goal is not to choose the most modern model, but the one that fits your operational reality.