Skip to main content
Aynitech Factory
EN ES
background-header
Work / A Multi-Tenant Analytics Platform for a Health-Benefits Card Program

A Multi-Tenant Analytics Platform for a Health-Benefits Card Program

Reporting Database + Self-Serve BI · Competitive-Privacy Access Across Four Stakeholder Types

Client
Challenge

Speak Benefits administers a restricted-spend payment-card program on behalf of US health insurance providers — cards that can be limited to approved purchases such as healthy foods and over-the-counter medicine, and that also carry coupons and discounts. The program touches four very different stakeholders, each needing its own view of the data: the health plans that issue the benefit, the retailers where the cards are spent, the CPG manufacturers whose products and coupons are involved, and Speak itself. The underlying data was spread across the card processor, a coupon-processing partner, the issuing bank, the health plans and Speak’s own core system (Speak HUB) — and any reporting had to keep every party walled off from its competitors.

Factories

Business Analytics + Data Engineering

Engagement
Project

The Approach

AYNITECH designed the data architecture and built the analytics platform. A reporting database is fed by ingestion and transformation from Speak’s many sources — direct connections and daily or real-time feeds from the credit-card processor, coupon partner and issuing bank, plus pushes from Speak HUB — bringing together member and card data, benefit configuration, monetary and behavioral purchase data (what, where, when, how often), health-related activity and health-plan bank data. On top sits a web-based BI tool with per-user configurable dashboards, both self-serve and Speak-built reports, automated exports and emailed reporting, and operations/admin tooling. Access is governed by competitive-privacy filters, and all user activity is tracked. Data refreshes as close to real time as possible, with daily updates guaranteed.

Built for Four Stakeholders

  • Health Plans — member acquisition and retention, STARS/HEDIS performance, member-health and medical-cost impact, plus operational and financial metrics — across each plan’s multiple programs (Medicaid, Medicare, PPO), which carry different funding rules. Each plan sees only its own members, never a competitor’s.
  • Retailers — customer acquisition, loyalty, foot traffic and transactions driven by the program — limited to their own stores.
  • CPG Manufacturers — product purchase share, which products members buy, and coupon take-up and redemption — limited to their own products, across all plans.
  • Speak — program and benefit design, ROI for clients and partners, and accurate financial management — card funding and the accounting between all parties.

The Data Foundation

Sources — health plans, the credit-card processor, the coupon-processing partner, the issuing bank, retailers (indirectly, via the card and coupon partners), Speak’s qualitative learnings, and the Speak HUB core system.

Data sets — member info; card info (balance, expiration, issuance); benefit configuration; monetary purchases; behavioral purchases (item, quantity, frequency, day, time, location, store type); health-related activity; and health-plan bank data (deposits, balances, withdrawals).

On the Roadmap

The platform is designed to grow toward AI-driven decisioning:

  • Guidance back to health plans on how to run more effective programs — what works best

  • Dynamic, personalized offer pricing — giving the right discount to the right member so coupon spend isn’t wasted

  • Real-time A/B testing feeding intelligent decision-making

  • Additional data sources, and a freemium model where some reports are free and others unlock for paid subscribers

At a Glance

  • Reporting database fed by ingestion + transformation from the card processor, coupon partner, issuing bank, health plans and Speak HUB

  • Member, card, benefit-configuration, monetary and behavioral purchase, health-activity and bank data unified in one place

  • Web-based, self-serve BI with per-user configurable dashboards, automated exports and emailed reports

  • Competitive-privacy access control — every party sees only its own slice — with full user-activity tracking

  • Near-real-time refresh, daily guaranteed; foundation laid for AI decisioning

Why It Matters

Speak sits at the intersection of healthcare, payments and benefits — among the most regulated, privacy-sensitive data there is — and serves four competing stakeholder types from a single platform. Building multi-tenant analytics where data isolation between competitors is mission-critical, fed by multiple external financial feeds and refreshed near-real-time, is direct evidence of AYNITECH delivering secure, enterprise-grade analytics in a demanding regulatory setting, and laying the data foundation for the AI-driven decisioning Speak plans next.