The Four Eras of Service Delivery — and Why Era 3 Is the Most Dangerous
Service economics frameworks were built for a world where delivery was 100% human. That world no longer exists. Work is now delivered through a blend of human expertise, AI tools, and software — and almost nobody is tracking the economics of that blend. This is the founding narrative of Service Economics: four eras, each with its own economics, and each with its own failure modes.
Intuition-Led Services
Decisions made on relationships, experience, and gut feel. No systematic tracking.
Day-to-Day
Partners price based on what feels right. Margins are discovered after invoicing — or when a client complains. Knowledge lives entirely in people's heads. New hires take months to become productive because nothing is documented.
What Breaks
- Margin surprises on projects that looked profitable
- No ability to scale without adding expensive senior hires
- Key person dependency — one departure can destabilize client relationships
The Signal You Recognise
The moment you realize a project that "felt fine" actually lost money — and you only found out 60 days later.
The Transition
The first systematic tracking. Usually triggered by a margin surprise painful enough to force change. Someone builds a spreadsheet. Then another. Then a dashboard.
Data-Driven Services
Metrics, dashboards, KPIs. Firms track utilisation, revenue per head, project status.
Day-to-Day
Tools like Harvest, Teamwork, or Kantata enter the picture. Weekly status reports exist. Monthly finance reviews happen. There are dashboards — perhaps too many. But the data lives in silos. Finance doesn't talk to delivery. Utilisation numbers don't connect to margin.
What Breaks
- Data exists but isn't connected — you have numbers without insight
- Finance reports arrive 30–45 days after the period they describe
- Dashboards become a reporting obligation rather than a decision tool
The Signal You Recognise
The meeting where three dashboards show different numbers for the same engagement, and nobody knows which is right.
The Transition
AI tools arrive. Someone starts using GPT for drafts. Claude for analysis. A developer integrates an API. Nobody tracks what it costs.
AI-Augmented Services
Teams use AI tools daily. Delivery is faster. But the economics are invisible.
Day-to-Day
Everyone has AI subscriptions. Clients benefit from faster turnaround. But nobody knows what AI actually costs per engagement. Pricing still assumes human-only delivery. AI costs are buried in 'software' or 'overhead.' Governance is absent where it's needed most.
What Breaks
- New cost structures with no tracking mechanism
- Margin erosion from unattributed AI costs across engagements
- Client transparency gaps — can you answer 'how much of this was AI?'
- Governance absent where it's needed most
The Signal You Recognise
The client who asks 'how much of this was done by AI?' — and you don't have an answer.
The Transition
A governance failure or margin crisis forces a rethink. The firm realizes it needs to treat AI as a tracked, attributed delivery input — not just a productivity hack.
Intelligent Services
Proactive intelligence. The operation surfaces problems before they become crises.
Day-to-Day
Delivery composition is understood and priced correctly. Governance is automatic. AI is a tracked, attributed input. Engagement health is monitored in real time. Problems surface as signals, not as surprises. Pricing reflects the true cost of blended delivery.
The Signal You Recognise
The team gets an alert that an engagement's margin is trending 8 points below target — three weeks before the monthly review would have caught it.
Which Era Are You In?
Take the 10-question Maturity Diagnostic to find out where your firm stands — and what it takes to move forward.
Take the Diagnostic