Reducing Vendor Dependency and Moving to In-House Engineering to Improve Velocity and Lower Cost
The Situation
Several critical platforms were heavily dependent on third-party vendors for development, support and change delivery. Release cycles were slow, costs were high, and knowledge lived outside the organisation. When issues occurred, resolution required vendor intervention, causing delays and frustration for product and customer teams. Senior leadership wanted lower spend, faster delivery, more reliability, and internal ownership.
The Task
Transition delivery of key systems away from vendors and build a high-performing in-house engineering capability. The goal was to reduce cost, shorten release cycles, improve stability, and embed real ownership within teams — without interrupting live service or creating unmanageable risk during the transition.
The Action / Approach
-
Conducted a full cost, capability and risk assessment of vendor-owned platforms and prioritised which systems to insource first based on business criticality and ROI.
-
Created a phased insourcing roadmap, ensuring knowledge transfer, documentation, and shadow-support were in place before removing vendor dependencies.
-
Built internal squads with product, engineering, QA, platform and SRE skills; hired key technical roles to close capability gaps while upskilling existing staff.
-
Negotiated new commercial terms, reducing vendor footprint and shifting from “black-box” delivery to transparent, shared ownership with internal teams.
-
Introduced CI/CD pipelines, infrastructure-as-code, automated testing and observability so new in-house teams could deploy safely and frequently.
-
Embedded a culture of end-to-end accountability, with clear SLIs/SLOs, ownership of incidents, and continuous improvement of service performance.
The Result
-
Reduced third-party spend by 25%, unlocking funds that were re-invested into internal capability and platform improvement.
-
Improved delivery velocity by 30%, with internal teams able to release multiple times per week instead of vendor-scheduled release windows.
-
Cut incident resolution time by 40% through in-house troubleshooting, observability and direct platform knowledge.
-
Eliminated vendor bottlenecks, giving the business greater control, transparency and predictability over delivery timelines.
-
Increased internal engagement and retention, building a culture where engineers owned outcomes instead of passing issues to external suppliers.
-
Transition completed with zero customer-impacting outages, proving the business could safely run and evolve its own platforms.