Sales Automation
Cisco's Customer Experience team had to analyze more than 50 signals across structured and unstructured systems to prepare renewal proposals, but the manual process left too little time to build personalized, data-driven proposals and prepare well for customer engagements.
Reported impact
Cisco expects the AI Renewals Agent to cut renewal-proposal preparation time by up to 20%, raise productivity across its large CX organization, and improve customer satisfaction by giving teams more time to tailor renewal discussions around each account's needs.
Cisco built the AI Renewals Agent with Mistral to automate a renewal-proposal workflow that previously required CX teams to manually interpret more than fifty signals across structured and unstructured systems. The agent sits over on-prem customer data and uses text-to-SQL techniques to pull relevant facts from multiple platforms, then assembles account context so teams can prepare tailored renewal proposals and customer-engagement plans faster. Cisco and Mistral's Applied AI team are fine-tuning the underlying models for Cisco-specific data and knowledge and scaling the deployment across Cisco's GPU cluster. The result is an agentic workflow that keeps sensitive data on premise while turning scattered account signals into draft proposals and preparation material for a large global CX organization.
Structured and unstructured customer signals remain on premise across multiple internal platforms; the AI Renewals Agent uses Mistral LLMs and text-to-SQL techniques to pull and interpret those signals, assemble account context, and return draft renewal proposals plus customer-engagement preparation material to CX teams.
"Cisco's use case involved a lot of TXT to SQL across multiple platforms with customer data that needs to stay on premise."
"The AI Renewals Agent was the first jointly developed AI solution to come out of Cisco and Mistral's strategic partnership"
"Up to 20% time reduction (estimated) in creating renewal proposals and preparing for customer engagements."
"collaborating on various fine tuning techniques to solve this use case while managing to scale the deployment across Cisco's GPU cluster"
"members of Cisco's Customer Experience (CX) team have to analyze more than 50 signals"
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
Google Cloud Sample Modified · Oct 9, 2026 Sansan Sansan needed to unify fragmented customer and sales data so teams could move from static reporting to immediate, data-driven sales actions without relying on manual aggregation or hallucination-prone AI outputs. BigQuery · Gemini · LookML 
MongoDB Sample Sep 30, 2026 MongoDB MongoDB needed to stop sales reps from spending 4 to 6 hours per account or deal cycle manually stitching together customer context from Salesforce, email, meeting transcripts, and documents before they could plan… Holly currently runs on 30 agents. Atlas Agent Engine 
OpenAI Sample Sep 25, 2026 Proaction Proaction needed a way to create highly personalized fleet-management demos and operational follow-ups without consuming scarce engineering time, while also extending automation into day-to-day fleet workflows for… 30 to 45 minutes to build each demo ChatGPT-5.6 Sol · Codex · GPT-6 Astra 
