liveBook · Manning

Return on investment

As explained in Architecting Autonomous Agentic Systems

Return on investment is a financial measure comparing the gain or loss from an initiative with the investment required to produce it.

Return on investment (ROI) compares an initiative’s gain or loss with the investment required to produce it. Traditional ROI works best for deterministic automation: estimate the work volume, calculate labor savings, and project a payback period. section 3.0

Agentic systems weaken those assumptions because behavior is uncertain, governance continues after implementation, and value may emerge along a nonlinear path. A single ROI estimate can therefore make an autonomy initiative look attractive while overlooking persistent governance costs, probabilistic failure, and delayed benefits. section 3.0 section 3.1

Use ROI cautiously when evaluating delegated autonomy:

  • Estimate direct investment and expected benefits, but include the continuing cost of governance. section 3.0
  • Treat failure as probabilistic and allow risk to reshape both cost and benefit over time. section 3.1
  • Read early negative results as evidence about the initiative’s trajectory rather than as an automatic cancellation signal. section 3.1

Claims-triage pilot

A traditional ROI lens records the pilot’s deeply negative result at month six and may recommend cancellation. A trajectory-based assessment asks whether the initiative is on a credible path to profitability, recognizing that learning and governance maturation can precede later value. section 3.1

Figure 3.2 Cost, benefit, and risk do not combine additively in agentic systems. Each input shapes Net Agentic Value, and risk continuously reshapes both cost and benefit over time. NAV therefore emerges as a trajectory, not a fixed output. Figure 3.2 traces this trajectory through the economic lifecycle.

Shows how cost, benefit, and risk combine into a time-dependent economic trajectory rather than a fixed ROI output. section 3.1

How Architecting Autonomous Agentic Systems covers it

How deep it goes
Defines traditional ROI and contrasts its point estimate with time-aware analysis, but defers detailed calculation to the broader economic framework.
Examples
Uses a claims-triage pilot whose deeply negative month-six result illustrates how traditional ROI can misread autonomous initiatives.
What it assumes
Assumes familiarity with investment, labor savings, payback periods, and deterministic automation while introducing autonomy-specific economic concerns.
What you can do after
Evaluate autonomous initiatives by distinguishing immediate ROI from a credible path toward time-aware profitability.
How it connects
Connects ROI with deterministic automation, governance, risk adjustment, Agentic Economics Canvas, and Net Agentic Value.

Also explained in

Compare how 14 books explain return on investment →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Architecting Autonomous Agentic Systems. The section links point to the parts of the book it draws on.