Skip to main content

    Interactive assessment

    Human-AI Workflow Design Canvas

    A structured assessment of how well your organisation has designed its human-AI workflows across five dimensions: workflow selection, role allocation, decision rights, handovers, and measurement.

    Most organisations have individual AI use but lack structured human-AI workflows (McKinsey, 2025), and only 3% report extensive organisational AI adoption (Watermark, 2025). This diagnostic helps you assess how deliberately your organisation has designed its human-AI workflows, from role allocation and decision rights to handovers and measurement, so you can move from ad hoc individual use to scalable organisational capability.

    Questions or items
    20 questions or items
    Dimensions
    Workflow Selection & Purpose, Role Allocation, Decision Rights & Accountability, Handover Integrity, Outcome Measurement
    What you receive
    Personalised results with configured interpretation bands, section scores and development guidance.

    What it covers

    • Workflow Selection & Purpose
    • Role Allocation
    • Decision Rights & Accountability
    • Handover Integrity
    • Outcome Measurement

    Choose one critical workflow in your organisation and rate each statement from 1 (Strongly Disagree) to 5 (Strongly Agree). Be honest, the diagnostic is most valuable when it reflects reality, not aspiration.

    Results: Personalised results with configured interpretation bands, section scores and development guidance.

    How this interactive diagnostic works

    This page provides the documented overview before the interactive experience is loaded. It lets you read the purpose, the material covered and the way the result is presented without entering answers or sending a request. When you choose to continue, the interactive version uses the same material shown here. It records your selections while you work and only calculates its result when its required items are complete.

    The interactive experience remains responsible for its own input, score and lead-capture step. Where the route requests a first name and work email, its existing form handles validation and submission. This overview does not replace those steps or submit a lead.

    Statements in this diagnostic

    The interactive version groups the statements below by section. Read them in relation to the situation named in the instructions, then use the response scale shown by the diagnostic. The section descriptions are part of the existing configuration and explain the aspect each group is intended to examine.

    Workflow Selection & Purpose

    Effective human-AI design starts by targeting a small number of high-value workflows with a clear purpose, rather than scattering individual experiments (McKinsey, 2025).

    1. We have clearly identified which specific workflows would benefit most from AI augmentation
    2. We understand the current pain points and inefficiencies in the workflows we are targeting
    3. We have defined the specific outcome we expect AI augmentation to deliver for each workflow
    4. Our AI efforts focus on a few high-value workflows rather than scattered individual experiments

    Role Allocation

    Mature workflows make explicit, for every step, whether AI augments, humans lead, or the two collaborate (Raisch and Krakowski, 2021).

    1. For each step in our priority workflows, we have defined whether AI augments, humans lead, or both collaborate
    2. Our people understand which parts of their work AI is expected to support and which remain human-led
    3. We have removed ambiguity about who or what owns each step of the workflow
    4. We design AI into the workflow deliberately rather than letting individuals adopt tools ad hoc

    Decision Rights & Accountability

    Clear decision rights reduce paralysis and risk when humans and AI work together (Raisch and Krakowski, 2021).

    1. For each critical decision, we have defined whether a human decides, AI recommends, or the two operate as a hybrid
    2. A named person is accountable for the outcome of each AI-supported decision
    3. People know when they can override an AI recommendation and how to do so
    4. We have a clear escalation process for when AI outputs are questioned or appear wrong

    Handover Integrity

    Handover points, where work transitions between humans and AI, are where errors and miscommunication most commonly occur (Deloitte, 2024).

    1. We have identified the points where work transitions between humans and AI
    2. Each human-to-AI and AI-to-human handover has a defined quality check
    3. We actively monitor handover points for errors and miscommunication
    4. Our teams know what good and poor quality looks like at each handover

    Outcome Measurement

    Organisations that succeed with AI measure by performance outcomes and behaviour change, not by adoption alone (McKinsey, 2025).

    1. We measure our human-AI workflows by performance outcomes, not just by how many people use the tools
    2. We have baselined metrics such as decision speed and decision quality before scaling
    3. We track the impact of AI augmentation on team engagement and learning, not only efficiency
    4. We pilot with a small team, learn and iterate before scaling our human-AI workflows

    How the result is read

    The interactive diagnostic totals the selected item scores by section and across the whole assessment. It then matches those totals to the configured interpretation bands and shows the section-level results alongside the overall result.

    • Ad Hoc, 20 to 40: AI use is individual and unstructured. Deliberate workflow design is needed before AI can scale.
    • Emerging, 41 to 60: Some workflow design is in place, but decision rights, handovers, or measurement remain unclear.
    • Structured, 61 to 80: Well-designed human-AI workflows with specific areas to strengthen.
    • Orchestrated, 81 to 100: Mature, deliberately designed human-AI workflows ready to scale across the organisation.

    Sources named in this diagnostic

    • Deloitte (2024) State of AI in the Enterprise, 6th edn. Deloitte AI Institute.
    • McKinsey & Company (2025) The State of AI: How organisations are rewiring to capture value. McKinsey Global Institute.
    • Raisch, S. and Krakowski, S. (2021) 'Artificial intelligence and management: The automation-augmentation paradox', Academy of Management Review, 46(1), pp. 192-210.
    • Watermark Search International (2025) AI in Senior Leadership. London: Watermark Search.

    Where to take the next step

    Explore development for designing human-AI workflows with clear roles, decision rights, handovers and measurement.

    Explore AI Leadership Readiness

    Frequently asked questions

    What does the Human-AI Workflow Design Canvas measure?

    Most organisations have individual AI use but lack structured human-AI workflows (McKinsey, 2025), and only 3% report extensive organisational AI adoption (Watermark, 2025). This diagnostic helps you assess how deliberately your organisation has designed its human-AI workflows, from role allocation and decision rights to handovers and measurement, so you can move from ad hoc individual use to scalable organisational capability.

    How many questions are in the Human-AI Workflow Design Canvas?

    It contains 20 questions grouped into 5 dimensions: Workflow Selection & Purpose, Role Allocation, Decision Rights & Accountability, Handover Integrity, Outcome Measurement.

    How is the Human-AI Workflow Design Canvas scored?

    Responses are scored within each section and totalled across the diagnostic. The total is compared with the configured interpretation bands.

    What happens after I complete the Human-AI Workflow Design Canvas?

    After the required items are answered, the existing first-name and email gate is shown before the results are displayed. The result includes an overall interpretation, section scores and development guidance.

    More to explore