Insurance Tech

What I did for the insurance-tech client

Via Studio Simpatico, I worked on two related pieces on the client's existing member portal: an Iterable-backed newsletter subscription manager, and a Google sign-in option alongside the portal's existing OKTA login.

The Iterable subscription manager

Logged-in users needed a way to see and change which newsletter lists they were on. The page reads their current subscriptions from Iterable and shows them as checked options; saving writes the changes straight back, and creates a record for anyone who didn't already have one. Identity came from the portal's existing OKTA login, so users never had to sign in again just to manage their preferences.

The Google sign-in gate

A follow-on piece let the client's own staff get past the login gate with a Google account instead of an OKTA one. The gate accepts a valid session from either, with a "Log in with Google" option beneath the existing prompt, connected to the client's own Google organisation.

The constraint that shaped it: this was a read-only gate for content, not an account system. Visitors were never meant to have any notion of an account on the site, so access was granted purely by checking membership of the client's Google organisation, with nothing stored on our side. For anyone who came in via Google rather than OKTA, the newsletter preferences page was simply hidden, since it belongs to the OKTA side.

The work was tested against a separate sandbox before going live, with the client setting up their own side and me helping where it was needed. The sign-in piece was scoped at roughly one to two days.

How I would describe working together

Via Studio Simpatico, with the agency owning the client relationship and me owning the integration. The edge cases were agreed before any building started, which is what kept two login-sensitive pieces straightforward to deliver.

All case studies

This article was drafted with the help of AI to populate the page. I'm in the process of rewriting it - a principle I adhere to across all projects. AI produces boilerplate, not production-quality output.