Integrations & systems
Should we build direct carrier APIs or buy a multi-carrier platform?
Short answer
Buy a multi-carrier platform when you need many carriers quickly, use mostly standard services and don't have engineers to spare. Build direct integrations when a few carriers carry most of your volume, you need special services or data the platform doesn't support, or per-parcel platform fees outweigh the cost of building.
Many companies end up hybrid: direct integrations for their top carriers, a platform for the long tail.
The two options
A multi-carrier platform (shipping software that connects to many carriers through one API or interface) gives you labels, tracking and often returns and customs documents for many carriers at once. You pay a subscription and often a fee per parcel or per label.
Direct integrations connect your systems straight to each carrier's API. You build and maintain each one, but you control the data, services and costs.
The decision in one table
| Factor | Points to buying a platform | Points to building direct |
|---|---|---|
| Number of carriers | Many, changing often | Few, stable |
| Volume concentration | Spread over many carriers | Most volume on one to three carriers |
| Services | Standard home, PUDO, returns | Special services, custom label rules, injection, bulky |
| Customs data | Platform supports your lanes' fields | You need fields or logic the platform lacks (e.g. new EU product identifiers early) |
| Engineering capacity | Little or none | A team that can build and maintain integrations |
| Cost at scale | Low to medium volume | High volume, where per-parcel fees add up |
| Speed to launch | Weeks | Weeks to months per carrier |
| Control and data | Acceptable to depend on a vendor | Tracking data and service logic are core to your product |
Compare total cost, not list price
For a platform: subscription, per-parcel or per-label fees, onboarding, plus the integration between your systems and the platform (it still needs building). For direct integrations: build effort per carrier, testing and certification, monitoring, and yearly maintenance as carriers change their APIs. Put both over three years against your real volume per carrier. The answer often changes once volume concentrates on a few carriers.
The hybrid setup
A common pattern: build direct integrations for the two or three carriers that carry most of your parcels, and use a platform for smaller carriers, new markets and tests. The key is one internal shipment model, so your warehouse and customer service see one way of working whatever the route underneath.
Questions to ask a platform vendor
- Which of my carriers and services are supported today, in which countries? Which are on the roadmap only?
- How are tracking events normalised, and can I still see the carrier's original events?
- Which customs fields are supported per lane, and when will new requirements be added?
- What happens when a carrier's API is down: queueing, retries, fallback labels?
- How is pricing structured at my volume in two years, not only today?
- Can I export all my shipment and tracking data if I leave?
For carriers and consolidators
The same question applies from the other side. A carrier onboarding merchants or partners decides whether to offer its own API, to be available through multi-carrier platforms, or both. Consolidators connecting to many last-mile carriers face the build-vs-buy choice per country. The table above still applies; add the cost of supporting your own customers' integrations.
Who this is for, and who it isn't
A good fit
- Merchants outgrowing a plug-in or platform
- 3PLs and consolidators connecting to many carriers
- Carriers deciding how merchants and partners connect to them
Not a good fit
- A small shop where the platform built into your shop system is enough
- You want a reseller to sell you a specific platform
- Software unrelated to delivery and returns
How I help
I answer the build-or-buy question with your volumes and requirements as part of an Integration Blueprint (from €2,250), then specify or lead the chosen route: with your developers (from €4,750) or with developers I contract (from €18,500). I don't resell any platform, so the recommendation is independent.
I've worked across 18+ carrier and API integrations, including enterprise carrier integrations in international product development at Hermes, and I design and run predictapp, my own production platform with APIs and data pipelines.
Questions
Is a multi-carrier platform always cheaper?
No. It's usually cheaper to start. At high volume on a few carriers, per-parcel fees can exceed the cost of direct integrations within a couple of years.
Can we switch later?
Yes, if you keep your own internal shipment model and don't spread platform-specific logic across your systems.
Do platforms handle customs data?
Many do for common lanes. Check that they support the fields your lanes need, including the EU product identifiers required for low-value imports from 1 November 2026.
Do you recommend specific platforms?
I compare options against your requirements and I don't resell any platform, so the recommendation stays independent.
Free discovery call
Not sure whether to build or buy? Let's run the numbers
30 minutes, no cost. Bring your lanes, volumes and questions; you leave with a clear next step, whether or not we work together.