Use casesCustomer journeys

The pages your customers use, and the systems behind them, run by one team.

A customer arrives with a practical question and a site that has to answer it, keep answering it as the content changes, and never break the things connected to it. We build the journey, the recommendation and search inside it, the controls editors need to keep it current and the release pipeline that keeps it moving.

What gets in the way

A business owner wants to know which coverage fits the work they do. A policy page does not answer that. A journey does: describe your business, see the options, move forward.

Then the content team has to keep it current without an engineer on call, and the engineering team has to change it without losing track of everything attached to it.

Accessibility gets checked once, by hand. Dependencies slip release by release. Three years later the upgrade is a project of its own.

What we build

One team across the pages customers use and the systems behind them.

  • Recommend. A questionnaire, a progress stepper and a recommendations page, reachable from the homepage, the product pages and the navigation, built as a standalone component.
  • Search. Search by profession or need on every brand, with the CMS templates, the publishing workflow and the scripts to populate, migrate and export the data behind it.
  • Control. Warnings for items that cannot be published, feature flags, rules for which controls an editor can add, image validation and references remapped when a page is copied.
  • Modernize. Components moved to TypeScript, platform upgrades through their regression rounds, and performance work aimed at specific causes rather than a score.
  • Release. Staging and production in Argo CD, with weekly dependency releases for the data services behind the business.

How we know it works

Accessibility is checked the same way every time, so a fix can be compared like for like.

  • For HUB we built an internal scanner and dashboard in 12 days, combining axe-core checks, custom probes and recorded baselines, and used it on 40 combinations of page type, device and brand.
  • Components and fixes built for one brand were ported to its sister brand and adapted, not rebuilt.
  • Two data services received 82 dependency releases in a year, 41 each, so nothing fell years behind.
  • We claim what we measured. Those are delivery and coverage figures, not before-and-after conversion, and we worked alongside HUB’s own teams rather than delivering the whole platform.

Client work

    • HUB International

    40 page type, device and brand combinations in the accessibility assessment

    82 dependency releases across two data services from September 2025 to September 2026

    12 days to build an internal accessibility and agent readiness scanner

    Read the story

Where it fits

  • Insurance

    Insurance, from the question a customer asks to the systems that answer it.

Questions

Is this AI?

Mostly it is engineering, and we say so. Recommendation and search are where models can help. The controls, upgrades and releases are what keep the journey alive. We bring one team to both.

Which platforms?

HUB runs on Sitecore JSS, React and TypeScript, with .NET services and Argo CD. We work on what you have.

How long is an engagement like this?

HUB’s ran three years, as steady work with weekly releases, rather than one project with an end date.

What do editors get?

Warnings before they publish something broken, rules for what goes in a section, validation on images, and content references that survive copying a page.

Bring us the project that matters most.

Tell us what you’re building and what’s in the way. You’ll hear back within 24 hours.