An Evidence-Based Diagnosis for Your IT Operations.

No guessing. Only analysis. AI-assisted pattern detection identifies exactly where value is leaking—without touching live servers or disrupting operational teams.

The Process

01. Data Request:

A strict "minimum data needed" rule is applied. Specific operational narratives and service catalogue definitions are requested. Access to production servers is never required

02. PII Scan:

Before analysis begins, a dedicated Privacy Scan identifies and neutralizes Personal Identifiable Information (PII). Any residual names, emails, or phone numbers are redacted or pseudonymized

03. AI Pattern Detection:

AI accelerates the analysis of large volumes of operational text to surface recurring failure patterns and hidden risks not visible through standard metrics.

04. Consultant Analysis:

All findings and scores are human-reviewed by the Principal Consultant to expose the operating reality—identifying exactly where instability, escalation, or cost is being created.

05. Executive Roadmap:

Leadership receives a package based on the tier level, designed to translate findings into immediate, manageable action

06. Project Closure:

The engagement concludes with a 30-day retention window, followed by the permanent, auditable deletion of all raw logs, datasets, and intermediate models.

Assessment Tiers

Choose the Right Depth

Three tiers. Same evidence-led framework. Different depth and decision power.

  • Purpose
    Rapid executive-level visibility into service delivery risk and obvious hotspots — with minimal client effort.

    Best for
    A quick “are we at risk?” view across 1–3 business-critical services, before budgeting or challenging “green” reporting.

    Typical timeline
    10–14 days (remote)

    What you get

    • 1-page executive scorecard (3 scored domains + qualitative notes on 2)

    • Top risk hotspots + recurring themes (with concrete examples)

    • Fix / Watch / Ignore recommendations (high-confidence actions only)

    • Lightweight evidence log (what each finding is based on)

    Minimum data (typical)

    • Incident log (last 6–9 months)

    • Service catalogue + service ownership (current)

    • Optional: change log (3–6 months), service review packs, high-level SLA/target definitions

  • Purpose
    Identify root causes behind recurring escalations, instability, and vendor friction — and produce a decision-ready 90-day improvement roadmap.

    Best for
    When leadership needs root causes + what to do next, without launching a full transformation programme.

    Typical timeline
    ~4 weeks (remote)

    What you get

    • Executive health scorecard with overall Health Score (1–5) + risk band

    • Evidence repository (finding → evidence → “so what”)

    • Quantitative diagnostics (recurrence, reopen patterns, change failure signals, handover “hops”, major incident duration drivers)

    • Root-cause themes + control gaps (documented process vs operating reality)

    • Prioritised 90-day roadmap (0–30 / 31–60 / 61–90) with owners at role/group level

    Minimum data (typical)

    • Incident + Problem + Change logs (12 months+; problems often 12–24 months)

    • Service catalogue + ownership

    • Major incident register / bridge timelines (where applicable)

    • Operating model artefacts (RACI, role-based org chart, process maps)

    • Vendor/SLA targets + last 3–6 months of service review packs

  • Purpose
    Decision-grade assessment for major structural change — including scenario options and readiness analysis.

    Best for
    Outsourcing/insourcing, scaling, tool/process overhaul, mergers, or operating model change — when the wrong move is expensive.

    Typical timeline
    6–10 weeks

    What you get

    • Expanded scorecard + deep diagnostic report

    • Readiness assessment (people / process / vendor / governance / data controls) with gaps + prerequisites

    • Scenario + risk analysis (2–3 realistic options with trade-offs, critical path, and likely failure modes)

    • Executive decision pack (what to approve / defer / stop)

    • 12-month plan + initial 90-day tranche

    • Full evidence repository with traceability from conclusion back to data

    Minimum data (typical)

    • Extended data window (often 12–24 months) across incidents / problems / changes / major incidents

    • Availability + “perceived outage” metrics (where available)

    • Full operating model + governance artefacts

    • Vendor strategy + contract/SLA structures

Compare tiers at a glance

Category Tier 1 — Baseline Tier 2 — Core Most Popular Tier 3 — Deep
Overview
Typical timeline 10–14 days ~4 weeks 6–10 weeks
Best for Fast risk visibility Root causes + 90-day plan Scenarios + readiness
Scope depth Directional hotspots Full health check Decision-grade diagnostic
Stakeholder interviews 1–2 (optional) 3–5 interviews 5–10+ interviews
On-site option 1–2 days 2–4 days 5–8 days
Data
Data window 6–9 months 12 months 12–24 months
Core datasets Incidents + Service Requests + Change + Catalogue All Tier 1 + Problems + MI + Process & Governance Artefacts All Tier 2 + Availability/Outage + CI Governance
Outputs
Primary outputs Scorecard + Fix/Watch/Ignore Evidence repo + 90-day roadmap Decision pack + 12-month plan
Evidence traceability Light Full repository Full traceability
Scored health domains 3 of 5 (qualitative notes on 2) All 5 domains All 5 + readiness dimensions
Advanced
Readiness assessment — — Included
Scenario analysis — — Included

Minimum data required

Most clients can export this in 1–3 hours.

  • Incidents Records (required)

  • Service Requests (required)

  • Service catalogue + ownership (required)

  • Change Records (required)

  • Problem Records (Tier 2/3)

  • Major Incidents/bridge timeline (Tier 2/3)

  • Vendor SLA targets + service review packs (Tier 2/3)

  • Availability / perceived outage (Tier 3)

Privacy & Controls (applies to all tiers)

  • Data minimisation: personal data not required for diagnostic outcomes

  • Free-text sanitisation recommended; residual scan + quarantine for anything suspicious

  • AI is gated: analysis runs only after validation; quarantine pauses processing

  • Evidence traceability: every finding ties back to data

  • Retention: data deleted within 30 days after the final report (unless agreed otherwise)

  • No employee performance assessment — this is about system/control failure modes

FAQ

Straight answers to the questions people actually ask before sharing operational data.

  • No. The assessment is designed around operational signals (volumes, timestamps, categories, outcomes, recurrence). We recommend sanitising free-text before sharing. If anything suspicious appears, it’s quarantined and excluded from AI analysis.

  • Only after data is validated and privacy controls are applied. AI is used to accelerate pattern detection (hotspots, recurring failure modes, anomalies). Final conclusions are consultant-validated and linked back to evidence.

  • Anything that can export structured logs. Common sources include ServiceNow, Jira Service Management, Freshservice, BMC, and similar tools. If you can export incidents/changes/problems and service ownership, you’re good.

  • You get decision-grade outputs, not a vague maturity score:

    • An executive scorecard (overall health + domain scores)

    • The key failure themes and hotspots (with supporting evidence)

    • A prioritised roadmap (Tier 2 includes a 90-day plan; Tier 3 extends to scenarios/readiness and a longer horizon)

  • We scope to what’s available and tell you what confidence you can expect. Tier 1 works with a minimal dataset. Tier 2/3 benefits from problems/changes/MI and governance artefacts, but we can phase the analysis.

  • Low. Most effort is exporting data and confirming service ownership. Tier 2/3 typically includes a small number of interviews/workshops to validate findings and ensure the roadmap is implementable.

  • No. This is about system and operating-model failure modes (process reality, handoffs, governance, vendor dynamics, signal quality). We don’t do individual performance assessment.

  • Data is retained only as long as necessary to complete the assessment and produce the deliverables, then deleted within the agreed retention window (your current model: within 30 days after final report, unless explicitly agreed otherwise).

Ready to scope the right tier?

Get in contact