The Model Context Protocol (MCP) lets you connect an AI assistant such as ChatGPT, Claude, Copilot or Gemini to external tools. No need to switch tools: you can keep using the AI assistant you're already familiar with.
With Upply's MCP, your assistant queries freight market data directly: it answers with up-to-date prices and trends, rather than a rough estimate.
1. General-purpose AI tools are not freight rate specialists
Ask ChatGPT or Copilot "How much does a full truckload from Lyon to Barcelona cost?". You will get a price range, but it is hard to check its accuracy.
Three reasons:
-
Knowledge frozen in time: an AI model is trained up to a specific date, while freight prices evolve every week.
-
No access to reliable market data: actual spot and contract rates are not publicly available on the web.
-
No explicit assumptions: vehicle type, spot or contract market, fuel surcharge… the AI does not say what its figure is based on.
For a purchasing decision or a negotiation, that is not enough. AI needs reliable data to rely on.
2. What is the Model Context Protocol (MCP)?
The Model Context Protocol (MCP) is an open standard that allows AI assistants to connect to external tools and data sources. Once connected, the AI can fetch information at the source instead of guessing it.
To put it simply, you keep your usual AI assistant: the MCP just adds access to Upply's data.
In practice:
-
You ask your question in plain language, in the AI assistant you use every day.
-
The assistant calls the connector, in this case Upply.
-
It retrieves the data and answers as text, a table or a chart.
No API key, no code needed.
3. What it changes for shippers, carriers and consultants
Upply is a freight transport market data platform. Shippers, carriers, 4PLs and consulting firms use it to compare their prices with the market, track market trends and support their decisions.
Upply's MCP gives access to price benchmarks (road, ocean, air, spot and contract), over 200 market indices (Upply Freight Indexes, fuel, tolls, inflation, rail…) and price forecasts, based on 2 billion real freight prices. Your AI can then turn this data into charts, spreadsheets, slides or written analyses.
Benchmark a list of lanes
Before a negotiation, paste your lanes and current rates into the conversation and ask your AI to benchmark each one with Upply. It returns a spreadsheet with the market low, median and high next to your current rate, sorted by gap, and flags the lanes priced above or below the market.
Prepare a market update for management (shipper)
Ask your AI to summarise how road freight prices in Europe have changed over the past 12 months, using the Upply Freight Indexes. It produces a few slides with charts and key takeaways, ready for your next management meeting.
Justify a rate increase (carrier)
Votre contrat client date de janvier 2025 et vous souhaitez le réviser. L'IA récupère l'évolution du gasoil, des péages et de l'indice routier sur la période, puis rédige un paragraphe factuel à envoyer à votre client.
Prepare a client recommendation (4PL, consultant)
You manage or advise on transport for several shippers. Ask your AI to compare the prices paid on their main lanes with market benchmarks: it produces a summary table and the gaps to explain, ready to present to your client.
👉 Find more use cases in our prompt library.
4. How it works with Upply
Compatible with leading AI assistants
The Upply MCP works with ChatGPT, Claude, Microsoft Copilot, Gemini, Perplexity, Mistral Vibe, Notion, Dust, GitHub Copilot et any other MCP-compatible tool.
Secure by default
- Read-only: the AI consults the data and cannot change anything.
- Sign in with your Upply credentials, via standard authentication (OAuth 2.0).
- Same rights as in the app: your AI sees exactly what you see in Upply, according to your subscription.
Connexion en quelques minutes
-
Add the Upply server to your assistant using the URL https://mcp.upply.com.
-
Sign in with your Upply credentials.
-
Ask your question and mention Upply, so the AI uses its data rather than its own knowledge.
Detailed steps for each assistant are available in our developer documentation.
Not an Upply customer yet? The MCP is included in Upply subscriptions. 👉 Book a demo to try it on your own lanes.
Key takeaways
AI is only as reliable as the data behind it. With the MCP, you keep the assistant you already use and connect it to Upply's freight market data, for answers that are quantified, sourced and up to date.
Go further:
-
Discover the MCP Upply and its use cases.
-
Webinar on November 24 at 11:00 CET: Thomas Larrieu, CEO, and Charles Joannon, CPO of Upply, show you real use cases live. 👉 Register
- Want to see the MCP in action on your data? 👉 Book a demo with our experts
