Fewer outages. No surprise bills.
Stabilise production and control cost
Sound familiar?
- "150,000 Retail and Wealth Management customers were impacted by a severe 2 hour 45 minute digital channel degrade." A Scotiabank director's words, August 2018. In market risk, a bad night costs real money.
- The same incident comes back every quarter with a new ticket number.
- Hosting and vendor bills arrive that nobody forecast, and the vendors can't say what you got for them.
- Seventeen vendors, four tool stacks, and every one of them has its own contract.
What I do
Problem management, not incident management: root cause, then the change to how the team operates that stops the incident returning. I run the service as a product for the people who depend on it, with Risk or the business owning the roadmap alongside Technology. On cost, FinOps: spend visible by service and by capability, vendors explaining their value, contracts consolidated.
What it has produced
RBC Capital Markets, enterprise market risk technology: production incidents down 83% in six months, with a "No Broken Windows" health check left running. → RBC market risk
Scotiabank.com, 2022 to 2024: the FinOps programme that restored cloud capacity and cost control with the hosting vendors. → Scotiabank, reuse before build
Deutsche Bank: 17 vendors' tooling brought under one service and one set of contracts, vendor headcount down 50%. → Deutsche Bank developer tooling
How we'd start
Six months, as at RBC: the incident record analysed, the top causes fixed in order, the operating changes made and measured, and the health check handed over.