An operator starting out who gets their first twenty jobs from a platform has a very specific feeling: this works. And it does work. The problem shows up two years later, when 80% of revenue still arrives through the same place and that place decides to change its commission, its allocation algorithm or its cancellation policy.
This isn't an article against marketplaces. It's the sum worth having done before somebody else does it for you.
What a marketplace actually gives you
Three things, and none of them small:
- Demand from day one, without spending on marketing or waiting to rank.
- Volume to fill gaps, which is what turns a car sitting idle at eleven in the morning into a job.
- No commercial work: no negotiating, no chasing payment, no explaining rates.
To get started, or to fill the troughs, it's hard to beat. The same holds for taking jobs from other operators, which is the peer-to-peer version of the same idea, covered in how a dispatch pool works for the receiving operator.
What it costs, in euros
Do this sum with your numbers, not the example's:
A hundred jobs a month averaging €60 is €6,000 of revenue. At 20% commission, the platform takes €1,200 a month, €14,400 a year. At 25%, €18,000.
That's the figure to put next to what a channel of your own costs, rather than the vague sense of "they charge a bit much". A site with a booking engine, your own domain and maintenance costs, per the Spanish market, between €300 and €3,000 once. If your own channel brings you a single job a week, at €60 that's €3,120 a year paying no commission: the website has paid for itself before the first year is out.
What you don't get back: the customer
Commission can be calculated. The other thing, less so.
When a job comes through a platform, the passenger is the platform's customer. You don't have their email, you can't write to them in December to remind them you exist, and if they need a transfer next year they'll look where they found one the first time. You did the work, put up the car, carried the risk, and handed over the relationship.
With a customer of your own, the second job has no acquisition cost. With a platform one, that passenger's tenth job pays commission exactly like the first.
That's also the conversation to have with hotels and agencies, where who keeps the customer record genuinely is negotiable: it's set out in working with hotels and travel agencies without losing the customer.
The risk that doesn't show up until it does
Depending on a single channel isn't just expensive, it's fragile. Three things that genuinely happen:
- The commission goes up. There's no negotiating when 80% of your revenue comes through there.
- The allocation changes. A tweak to the algorithm, a new rate from another operator, a shift in priority, and your volume drops without you doing anything differently.
- Ratings rule. Two incidents in one bad week can cost you months of visibility, on criteria you don't control.
None of the three gives you notice.
The sensible answer: both, in this order
Almost no profitable operator lives off just one of the two sides. The split that works is this:
- Your own channel is the base. The customers who look for you by name, the repeat ones, the corporate accounts and the hotels. Full margin and the relationship is yours.
- The marketplace is the filler. The gaps, the seasonal troughs, the dead hours. Lower margin, but on capacity that would otherwise sit idle.
And the order matters: building your own channel once you're 80% dependent on a platform is far harder than building it while that dependency is still 40%, because by then you have neither the time nor the urgency.
What to do next
Look at the last three months of revenue and work out what share came through a channel you don't control. If it's over 60%, that's the number to bring down this year, and you don't have to break with anyone to do it: it's enough for your own side to grow faster.
And for your own side to grow, first of all somebody has to be able to book you without talking to you, at any hour, with a price in front of them. That's what separates a website from a contact form answered tomorrow.
