Partners
EY accelerates AI discovery and business impact with Skan AI
Skan AI shows how work actually gets done across the teams and applications in scope, including the workarounds nobody wrote down. EY brings the industry knowledge, transformation and delivery experience to act on it. Together, we take clients from evidence to a funded plan in 6–8 weeks and then help deliver it.
Know where to invest in enterprise AI in 6-8 weeks
Teams work around systems, toggle between them, and manually work around exceptions that appear in no procedure document. Up to 40% of effort goes toward administrative work that isn’t written down anywhere.
Even when AI adoption is high, strategic value is often lagging because deployments are layered onto existing inefficiencies, limiting adoption, value, and scale. To realize the full benefits of agentic automation, organizations need visibility into how work is performed and what opportunities will deliver the most value.
Four challenges come up in almost every operation we look at:
- Complex, fragmented processes across teams, systems and channels, causing delays, duplication and inconsistent outcomes
- High manual effort, rework and process variation, driving up cost, slowing service and eating into staff capacity
- Little visibility of how work is actually done, so root causes are hard to find and decisions rest on assumptions
- Unfocused automation and AI spend, so value is hard to realize, risk goes up and it’s hard to prioritize and scale
Each of these can be fixed. The first step is seeing the work as it actually happens, step by step, before deciding where AI and automation should go.
"With Skan AI, EY can show clients exactly where their siloed workflows are bridged with time-consuming manual effort and realize the transformation they need in weeks.
Build proprietary agentic intelligence designed for enterprise scale
Skan creates a visual representation of actual operations, rather than perceived workflows, providing visibility and key insights into how work is performed. EY uses those evidence-backed insights to shape the future-state design of clients’ operations and build the roadmap and the business case behind it.
What EY and Skan AI do together
Skan AI brings the observation layer: a desktop plug-in that captures how work is done, with no system integration and no SOPs needed upfront. EY brings sector knowledge, operating model and transformation experience, AI readiness, stakeholder alignment, the business case and delivery. Skan provides the evidence. EY turns it into decisions, a roadmap and delivered change. Technology alone doesn’t change an operation. Lasting transformation comes from combining operational evidence with human judgment, deep sector knowledge and the experience to turn insight into action.
Start with a short, sharp proof of concept (PoC). A PoC is a small, time-boxed trial on one real process that shows what the data reveals before any larger commitment. We agree one business area and a user group of around five people, then capture around 10 days of data. EY runs an operational review alongside it, covering the current operating model, productivity and profitability data, stakeholder interviews and a hypothesis workshop. Cost and client effort stay low, and the route to value is short.
Weeks 1–3: Discover transformation opportunities
Skan AI Blueprint observes how work is actually performed across the teams and applications in scope, with no SOPs to write and no interview program to schedule. EY reviews the operating model and the productivity and profitability data in parallel, so the observed data lands against a clear view of where cost sits.
Weeks 3–6: Diagnose workflows for redesign
Skan AI Intelligence uncovers process inefficiencies, rework and automation opportunities, each quantified by effort, frequency, and dollar impact to identify priority actions to add transformational value. EY uses internally developed domain models to translate the findings into a target-state design, incorporating regulatory, control and operational guardrails so AI and automation can scale with confidence.
Weeks 6–8: Prioritize transformation efforts
Using Skan AI’s evidence-based operational insights, EY converts the findings into a ranked opportunity backlog, a future-state design and a business case an executive committee can approve. Value delivered: a prioritized backlog in 6–8 weeks; a quantified, prioritized opportunity backlog, ready to fund; high-value hypotheses on where cost-to-serve and service can improve most; a rollout plan for scaling beyond the PoC, sequenced by EY; a board-ready summary of findings and next steps.
Ongoing: Deliver solutions and track outcomes
EY delivers measurable business outcomes through transformation execution, supported by Skan AI’s continuous intelligence layer, with ongoing visibility into how work is performed across teams and AI agents. EY mobilizes delivery, embeds the change with the teams doing the work and tracks benefits against the baseline measured in discovery.
Skan AI's Context Graph of Work
EY uses Skan AI’s Blueprint and Intelligence to create a visual representation of your operations. That means your transformation efforts are based on how work actually happens, rather than what makes it into documentation.
The result is end-to-end operational visibility, less rework, manual effort and cycle time, and AI and automation spend focused where the benefit can be measured. Change scales because ownership is clear and benefits are tracked from the start.
Frequently asked questions
The EY and Skan AI offering gives banking and insurance leaders a clear, evidence-based view of how their operations actually run, so AI budgets can be allocated on evidence rather than assumption.
EY applies internally developed domain models and deep sector knowledge to translate operational evidence into target-state processes, embedding the appropriate regulatory, control and operational guardrails, and shaping an actionable transformation roadmap and business case. Skan AI provides the observation layer, capturing how work is actually performed across people, processes, and applications. Most discovery programs start with procedure documents or workshops. This starts with the work itself.
The value is in the combination. Skan AI produces the evidence. EY reads it with sector knowledge, turns it into a plan the business can fund and then delivers the change. Discovery that usually takes a quarter of interviews is done in 6–8 weeks, and the plan that comes out of it is one EY can deliver.
Process mining reads system logs, so it can only show you what your systems recorded. It cannot see the spreadsheet a handler opens between two applications, the rekeying when a system times out, or the workaround your team runs every Friday afternoon.
That is usually where the actual cost sits. Observation captures all of it, including every step that appears in no log and no SOP. The difference matters most when you are sizing an AI business case or designing an agent workflow, because gaps in evidence become gaps in performance after go-live.
No. Documentation is not a prerequisite. It is one of the things the engagement produces.
You do not need existing SOPs, a mapping exercise, or a process library to begin. The engagement observes how work is actually performed and delivers the evidence baseline as an output, not an input. If you have documentation, it will be read. If you do not, the work still starts on time.
The engagement produces a ranked, costed view of where AI and automation will return value, built on a measured baseline of how your operations actually run.
Deliverables include:
- A quantified opportunity backlog, each item sized by effort, frequency, and cost, and tagged automate, optimize, eliminate, or agent
- A future-state design covering process, technology, data, and controls
- A prioritized roadmap sequenced by value and feasibility
- A business case with a measured baseline behind every number
- A live intelligence layer that keeps measuring after the program starts, so benefit is tracked rather than asserted
Skan AI captures and analyzes work at the process level, not the individual level. Your team fully controls what is captured, what is excluded, reporting aggregation, employee notification, and data residency.
Discovery runs in weeks, not months. The analysis needed to size an AI investment opportunity and build a defensible business case is available fast enough to inform a committee decision in the same quarter.
Traditional analyst-led discovery can take a quarter or more to cover a single workstream, using interviews and SOP reviews that capture how work is believed to happen. Observation captures how it actually happens and scales across the operation without an interview program to schedule or documents to write first.
The views reflected in this page are the views of its authors and do not necessarily reflect the views of the global EY organization or its member firms.