--- title: Conversational AI for Logistics: Shipper Churn Research | Avinya Labs description: A LangChain agent on OpenAI GPT-4.1 that runs shipper churn research over WhatsApp, from brief to a quote-backed report. Live in five weeks with two engineers. url: /case-studies/conversational-ai-logistics-shipper-research.html --- [Work](/work/index.html) / AI · Logistics # Conversational AI for logistics: automating shipper churn research with LangChain and OpenAI A regional freight and warehousing operator needed to know why commercial shippers were moving volume to other carriers. We built a LangChain agent on GPT-4.1 that asks the follow-up questions and writes the report, live in five weeks. [Book a call](https://calendly.com/abbylester/30-mins-meeting) [AI agents & agentic AI](/service-pages/service-ai-agent-platforms.html) Industry Logistics Type AI delivery engagement What we built [AI agents & agentic AI](/service-pages/service-ai-agent-platforms.html) Built with LangChain, OpenAI GPT-4.1, OpenAI API, WhatsApp Business, Reporting store Results **5 weeks**To production, 2 engineers **4**Conversation design iterations **1**Workflow from brief to report Live in production in 5 weeks with 2 engineers One follow-up per reply, held to the campaign brief Every theme linked to the shipper quotes behind it OpenAI GPT-4.1 through LangChain, no model to host Live in production in 5 weeks with 2 engineers One follow-up per reply, held to the campaign brief Every theme linked to the shipper quotes behind it OpenAI GPT-4.1 through LangChain, no model to host ## The challenge The operator needed a faster way to understand why shippers reduced volume or declined renewals. Collecting comments was easy; reading open replies, grouping the real causes and writing a summary the commercial team could act on was the bottleneck. A fixed survey was not enough: “transit is slow” might mean a missed vessel, a congested warehouse or a pricing dispute, and account managers needed a relevant follow-up before the answer went stale. ## What we built A LangChain workflow that runs each campaign from brief to report. The team writes a brief (target accounts, lanes or warehouses in scope, questions to explore); the agent turns it into a short questionnaire, opens the conversation on WhatsApp and asks one follow-up per reply that builds on what the shipper just said. When collection closes, GPT-4.1 analyses the threads and returns recurring themes, a sentiment split and the quotes behind each theme. ## The result Two engineers shipped the system to production in five weeks, through four conversation-design iterations that tightened how the agent follows a reply and holds the brief. Campaign setup, WhatsApp collection and reporting now sit in one workflow, and every finding links back to the source quotes. ## Example: why a shipper moved volume. Each reply drives the next question. Shipper reply | Follow-up | What gets captured “We moved part of our lanes to another carrier.” | “Which lanes, and what tipped the decision?” | Lane-level switching reasons “Transit times have been unreliable.” | “Which corridors are worst, and how does that hit your delivery commitments?” | Reliability gaps and downstream impact “Your detention charges are hard to predict.” | “Where do those charges show up, and what would make them easier to plan for?” | Cost friction and billing clarity “Warehouse cut-off times don’t match our production schedule.” | “What cut-off would fit your dispatch window?” | Operational constraints “We need clearer milestone updates before we commit more volume.” | “Which events does your team need, and how soon after they happen?” | Visibility requirements ## Built with. The orchestration, model and channel behind the agent. ### LangChain Orchestrates the brief, questionnaire, WhatsApp send, reply collection, analysis and report. ### OpenAI GPT-4.1 Runs the follow-up conversations and analyses the replies. ### OpenAI API Serves the model under enterprise terms: prompts and completions are not used for training. ### WhatsApp Business Delivers the questions and collects the replies. ### Reporting store Holds summaries, themes, sentiment and quotes in the operator’s existing warehouse. ## At a glance. Live in production in 5 weeks with 2 engineers One follow-up per reply, held to the campaign brief Every theme linked to the shipper quotes behind it OpenAI GPT-4.1 through LangChain, no model to host Short on time? ## Ask AI to summarise this case study. Opens your assistant with this page and a ready-made prompt. Nothing is sent from our site. [ChatGPT](https://chatgpt.com/?q=Summarize%20the%20key%20results%2C%20approach%20and%20technology%20in%20this%20Avinya%20Labs%20case%20study%3A%20https%3A%2F%2Favinyalabs.co%2Fwork%2Fconversational-ai-logistics-shipper-research%2F) [Claude](https://claude.ai/new?q=Summarize%20the%20key%20results%2C%20approach%20and%20technology%20in%20this%20Avinya%20Labs%20case%20study%3A%20https%3A%2F%2Favinyalabs.co%2Fwork%2Fconversational-ai-logistics-shipper-research%2F) [Perplexity](https://www.perplexity.ai/search?q=Summarize%20the%20key%20results%2C%20approach%20and%20technology%20in%20this%20Avinya%20Labs%20case%20study%3A%20https%3A%2F%2Favinyalabs.co%2Fwork%2Fconversational-ai-logistics-shipper-research%2F) [Google AI](https://www.google.com/search?udm=50&q=Summarize%20the%20key%20results%2C%20approach%20and%20technology%20in%20this%20Avinya%20Labs%20case%20study%3A%20https%3A%2F%2Favinyalabs.co%2Fwork%2Fconversational-ai-logistics-shipper-research%2F) ## Need to know why customers are leaving? [AI agents & agentic AI](/service-pages/service-ai-agent-platforms.html) [Book a call](https://calendly.com/abbylester/30-mins-meeting) ## More work. Other AI engagements we have delivered. AI · Agent platforms ### AI agent platform Agents retrieve, reason and act across systems, with human approval before high-stakes actions. ~40% less manual effort Link: /case-studies/enterprise-ai-agent-platform.html AI · Workflow automation ### Workflow automation platform AI extraction feeding a deterministic rules engine, with consistency checks built in. ~70% faster document prep Link: /case-studies/ai-workflow-automation.html AI · Travel ### Customer support assistant A support assistant grounded in live inventory data, available around the clock. ~35% fewer repeat contacts Link: /case-studies/ai-travel-chatbot.html ## Ready to turn your vision into reality? Tell us the problem, the data, and where manual effort sits today. We'll map the path from MVP to scale. [Book a call](https://calendly.com/abbylester/30-mins-meeting) [See the work](/work/index.html) Or email [contact@avinyalabs.co](mailto:contact@avinyalabs.co)