Sixty Partner Microsites on One Sovereign Platform — a New Distribution Partner Live in Three Days
Replacing agency-built, one-off partner websites with a managed multi-tenant platform — compliance-approved content blocks, AI-assisted localisation, and lead capture wired straight into the insurer's CRM.
The challenge
The insurer — a private life insurer with a large bancassurance and broker network across South Asia — sells most of its new business through partners: banks, non-bank lenders, brokers and affinity programmes. Every partner wanted a co-branded digital journey, and over six years the insurer had said yes sixty times. Each site had been built by whichever agency the partner or the regional team was using at the time. The estate ran on four content-management systems, two static-site generators and a handful of hand-built React applications, hosted in seven different places.
The cost of that sprawl was not the hosting bill. It was the compliance team. Insurance advertising rules require product wording, benefit illustrations and disclaimers to match what the regulator has seen; every site was reviewed by hand, and when a product was refiled or a disclaimer changed, the compliance team had to chase sixty agencies to update sixty pages. A single wording change took three to five weeks to propagate, and nobody could say with confidence that it had reached every site. Leads arrived as spreadsheet exports, contact-form emails and, in two cases, nothing at all.
Launching a new partner took six weeks or more, most of it waiting: for an agency quote, for a design round, for compliance sign-off on copy that had already been signed off elsewhere. The distribution team was losing partners to competitors who could get a co-branded journey live in a fortnight.
The approach
We started with an inventory rather than a redesign: every page on every site, mapped to the product and the disclosure it depended on. That map became the content model for a single multi-tenant platform. Product cards, benefit illustrations, eligibility tables, disclaimers and regulatory notices became versioned content blocks that the compliance team approves once. Partner sites are assembled from those blocks plus partner-specific configuration — logo, colours, domain, the products the partner is licensed to distribute, and the lead-routing rules for that partner.
AI does three specific jobs on the platform, none of them customer-facing without a human check. Summarization Wizard drafts localised versions of approved English blocks into the six other languages the partner network sells in, and a reviewer in each language signs them off before publication. Guardrail System checks every draft page against the approved wording set and flags any sentence that departs from it, so compliance reviews exceptions rather than whole sites. ChatNext runs a partner-site assistant that answers product questions only from the approved blocks, with a hard hand-off to a human adviser for anything that resembles advice.
The insurer did not want a project that ended with a handover. MindMap runs the platform as a managed service: monitoring, patching, the quarterly content refresh, the partner onboarding kit, and a small squad that turns a signed partner agreement into a live site. Migration of the existing sixty sites ran partner by partner over fourteen weeks, with each old site retired only after its replacement had matched or improved its search rankings.
The pre-built building blocks
Rather than commission a ground-up build, the engagement leaned on MindMap's pre-built accelerator library — production-tested components that compress what would otherwise be a six-to-nine-month build into weeks.
ChatNext
Partner-site assistant answering only from approved content blocks
Summarization Wizard
Localisation drafts across seven languages, human-reviewed
Guardrail System
Approved-wording checks on every draft page
Workflow Automator
Draft → review → compliance → partner approval → publish
The architecture
The platform is a single Next.js application deployed on the insurer's private cloud in-country, behind the insurer's existing web application firewall and content delivery network. A headless CMS holds the approved content blocks and their version history; tenant configuration lives alongside it, so adding a partner is a configuration change and a domain mapping rather than a deployment. Every published page records which block versions it was built from, which is the evidence the compliance team shows the regulator.
Lead capture runs through a consent manager aligned to India's Digital Personal Data Protection Act and the insurer's own privacy notice. Consented leads are posted to the insurer's lead-management system through its API gateway, tagged with the partner, the product and the campaign, so the right sales team sees them within minutes. Workflow Automator drives the publishing workflow: draft, language review, compliance exception review, partner approval, publish — with each step logged.
The AI components run inside the same perimeter. Summarization Wizard and Guardrail System use the insurer's Sovereign LLM Platform instance, so approved copy and draft copy never leave the insurer's network. ChatNext's retrieval index is rebuilt from the approved blocks on every publish; it cannot answer from anything else.
The numbers behind the story
All sixty legacy sites were migrated and retired in fourteen weeks. A new distribution partner now goes live in three days from a signed agreement, most of which is the partner's own logo and domain paperwork. Twelve partners have been added since go-live on that timeline.
A wording change to a product block now reaches every site that uses it within hours, with a record of when and where it was published. The compliance team reviews flagged exceptions rather than whole sites, and their review time per partner launch has fallen from days to under an hour.
Leads from every partner site now land in the insurer's lead-management system with consent recorded, routed to the partner's assigned team. The distribution team has partner-level analytics for the first time: which products convert on which partner's site, and where journeys drop.
The estate went from seven hosting arrangements and four content-management systems to one platform, one codebase and one security patching cycle.
“We used to measure a partner launch in weeks and a wording change in follow-up emails. Now a partner is live in three days, compliance signs off exceptions instead of whole websites, and every lead lands where it should. It runs itself — which is the point.”— Head of Partnerships & Distribution· Life Insurer, South Asia
Why MindMap was chosen
The insurer had been offered a redesign of the sixty sites by two agencies and a marketing-cloud licence by a large vendor. Neither solved the problem the compliance team actually had, which was propagation of approved wording, and neither would run inside the insurer's own perimeter.
MindMap proposed a managed platform rather than a project, with the content model built around regulatory disclosures instead of marketing pages, and AI used narrowly where it removed work — localisation drafts, wording checks and partner-site questions — with a human approving every customer-facing word.
The sovereign hosting mattered more than the insurer had expected. Partner sites collect personal data at the top of the funnel, and the insurer's board wanted that data, and the AI that touches it, inside the network the regulator already inspects.
Related deployments
Prior Auth Acceleration
Prior Auth Accelerator + DocGenie automated 70% of payer prior auth submissions end-to-end, reducing turnaround from 3 days to 4 hours.
Motor Claims Automation
Claims Router + DocuMage automated 80% of low-complexity motor claims to same-day settlement, with FCA-grade explainability on every decision.
Commercial Underwriting Automation
DocuMage + Multi-Agent Orchestrator automated the data-capture layer of commercial underwriting — 4x quote throughput at the same underwriter headcount.
Want an outcome like this?
Start with a 2-week AI Readiness Sprint. We deliver a prioritised use-case backlog and business case grounded in what's actually buildable with our accelerator library.