Platform Rescue at Goldwise: From Stalled Build to Live Trading
- Editor
- Jul 29
- 3 min read
Updated: Aug 11
Most fintech delivery failures are not technology failures. They are ownership failures: unclear architecture, drifting priorities, fragile releases, and a product that cannot yet survive real customers.
Goldwise is a UK precious-metals trading platform that lets customers buy, manage and sell physical gold, silver, platinum and palladium, including fractional ownership, with institutional pricing, market and conditional orders, safeguarded funds, and insured global vaulting. Getting a product like that to a trustworthy live state is not a feature factory problem. It is a platform ownership problem.

We were brought in to take technical ownership when delivery had stalled. The brief was not to sprinkle features on a fragile estate. It was to restore control, harden the operating path, and make the product ready for real UK customers.
The product reality we had to respect
A metals trading product only earns trust when the boring layers work: pricing that feels institutional, orders that complete cleanly, portfolio state that stays coherent, vault and custody narratives that match operational truth, and mobile releases that do not become a monthly emergency.
A 24/7 trading experience for physical precious metals, not paper proxies
Fractional bullion ownership with clear product and vault presentation
Institutional pricing against aggregated metals-market liquidity
Market and conditional order flows customers can actually use
Safeguarded funds and a security posture customers can scrutinise
iOS and Android distribution with Apple verification and controlled release
What platform rescue meant for us
Rescue work starts with truth: what is broken, what is risky, and what must be true before launch. At Goldwise we re-established engineering direction, hardened the backend and infrastructure paths, and made release quality non-negotiable.
Stabilised the product and recovered delivery from fragile patterns
Re-established architecture, ownership boundaries and delivery cadence
Drove production readiness across backend, frontend and infrastructure
Acted as technical owners for reliability, scalability and security
Worked with founders on product, architecture and go-to-market readiness
Mentored a multi-region engineering team to raise execution quality
Reached Apple verification so release became a controlled one-click path
Where the platform is now
The work is no longer a recovery story. The platform is live for UK customers, the company closed a funding round on the back of a working product, and partner integration has moved from roadmap to production.
Live on the App Store for UK customers, who buy, hold and sell physical metals
A funding round closed with a live product rather than a prototype behind it
Integration with partner platform Aspora delivered and running in production
Releases ship through a controlled, repeatable path instead of an event
The partner side of the platform is Goldwise Connect: pricing, execution, allocation, settlement and reporting behind one integration, so another platform can offer physical metals without building any of it. That is the surface the Aspora integration runs on, and it is the part of the estate least forgiving of loose engineering — a partner's own customers are on the other side of every call.

That sequence matters. Funding and partnerships arrived after the engineering was trustworthy, not before. Investors and integration partners both did their own diligence on a system that had to hold up.
Why this pattern keeps repeating
Across enterprise data platforms, regulated systems and high-availability fintech, we keep seeing the same failure modes: manual releases, weak observability, and architecture decided by ticket pressure instead of operating reality.
From FactSet-scale financial data to fintech platforms, healthcare infrastructure, ecommerce modernisation and public-sector programmes, we have learned the same lesson as a team: control beats novelty. Infrastructure as Code, deterministic CI/CD and explicit ownership are what make ambitious products shippable.
What we would ask if we were buying
If you are evaluating a platform partner, ask for operating evidence, not slideware. Can they take ownership of a messy estate? Can they make releases boring? Can they put reliability, security and product truth under the same control plane?
That is the standard we work to, whether the engagement is a fintech turnaround like Goldwise, a public-sector migration, or a cloud operating-model rebuild.


Comments