Test the economics and basic readiness of one workflow using visible assumptions. We do not pretend five sliders can rank your whole business. That requires structured discovery of the work, rules, exceptions, data and constraints.
Annual addressable team capacity
Using only visible cost assumptions
FOUNDATION WORK FIRST
Test one hypothesis with visible inputs.
Map workflow, rules, exceptions and data.
Pilot, build, prepare first or stop.
This is a screening model, not the full diagnostic. Adjust the visible assumptions below. The result can tell us whether this workflow deserves deeper discovery; it cannot credibly rank workflows we have not observed.
Capture enquiries quickly, ask approved qualification questions and route genuine opportunities to the right salesperson.
Saved time is converted using the visible realisability percentage. Revenue uplift is entered as gross-profit opportunity rather than being converted using hidden margin or conversion assumptions.
The central case using the visible assumptions currently entered.
Typical human boundary: Sales owns exceptions, high-value conversations and final commercial judgement.
This is not a scoped recommendation. The full diagnostic determines the actual workflow, control logic and human handoffs.
A valuable automation can still be the wrong next project. Mark only what you genuinely know.
Several critical workflow, data or control conditions are missing. Document or repair those foundations before deciding whether automation is the right intervention.
The screen above tells us about the selected workflow. It deliberately leaves the rest of the business unassessed until we collect evidence. That is where the full diagnostic starts.
A landing-page model cannot credibly rank workflows it has not observed. The full HostMetric diagnostic earns that recommendation by mapping the real work, evidence, rules, exceptions, data and system constraints first.
The screen is only a hypothesis check. The full assessment earns the recommendation through structured discovery, evidence validation and implementation readiness.
Start with the people doing the work. Reconstruct the normal path, decisions, exceptions, handoffs and present-day effort rather than asking what AI they want.
Separate reported assumptions from known facts, identify authoritative systems and expose anything that would force engineering to guess.
Only after the workflow is understood do we test realisable value, implementation effort and risk, then recommend build, pilot, preparation or no-go.
Document, simplify or test the workflow first when its current operating model is inconsistent or unreliable.
Saved time is converted using a visible realisability assumption rather than automatically being presented as a financial saving.
Deterministic workflows remain the backbone. AI is introduced only where language, classification or contextual judgement creates genuine value.
The finished review does more than show ROI. It gives the business and the technical team an evidence-backed view of how the workflow works, what data it depends on, what can be automated and what still needs human judgement.
Workflow, data, economics and implementation readiness
Your screening inputs travel with the enquiry as reported or estimated evidence. They become the starting point for discovery — not facts we silently treat as true.