PREP Before You RACE: Why AI Readiness Determines AI Success

The PREP framework for building process, data, and operating foundations before accelerating with AI

AI on a Broken Process is a Faster Broken Process

Why process excellence, clean data, and shared operations are the non-negotiable foundation before AI can deliver on its promise. (PREP before you RACE)

The Athlete Who Skipped the Warm-Up

Picture an elite sprinter stepping onto the track for the 100-metre final. She is gifted. Her coach has mapped her stride, analyzed her reaction time, and invested in the most advanced gear money can buy.

But this morning, she skipped her warm-up routine. No dynamic stretches. No activation drills. No mental preparation. She simply laced up and walked to the blocks.

The gun fires. She explodes out, faster than ever in the first 20 meters. Then, at 40 meters, a hamstring pulls. She crumples.

The very speed that was her advantage becomes the force that tears her apart.

This is precisely what happens when organizations deploy AI on processes that are not ready for it.

AI does not transform processes. It amplifies them.

Feed it a clean, well-structured, standardized workflow, and it returns compounding efficiency gains that feel like a superpower.
Feed it a fragmented, inconsistent, data-poor process, and it returns compounding errors at machine speed, and bills you for the privilege.

Miss any of these, and you are asking your AI to sprint on a torn hamstring. It will move fast, and fail just as fast. The failure will also be harder to diagnose precisely because it happened so quickly.

Why Process Inefficiencies Exist, and Why They Hide So Well

Before you can fix a process, you have to understand why it broke in the first place.

Across finance and accounting, recruitment, real estate, senior living, and workforce services, inefficiencies follow consistent patterns, and almost none of them were born from negligence. Most were rational responses to circumstances that have since changed.

The Seven Origins of Process Waste

1. Accumulation — Organic growth without governance
   The process worked perfectly for ten people in one office. Nobody redesigned it when the organization grew to 2,000 people across eight countries. Each regional team added their own workaround. Nobody mapped the aggregate result.
2. Legacy debt — Technology layer-cake
   ERP systems from one era sit beneath middleware from another, beneath point solutions added this year. Data flows through manual re-entry between systems that were never designed to speak to each other. The process is held together with spreadsheets and institutional memory.
3. Compliance drift — Regulatory accretion
   Every new regulation added a step. Nobody removed the step when regulations changed or when better controls emerged. The process carries ghost compliance that serves no current purpose.
4. Ownership gaps — Functional silos
   Finance does its version. Operations does theirs. No single owner sees the end-to-end journey. Handoffs become the graveyard of efficiency, queued tasks, mismatched formats, and downstream errors.
5. Undocumented decisions — The tribal knowledge trap
   The process works because Pamela in Manchester knows when to override the system, and why. When Pamela leaves, the process fails or gets incorrectly documented.
6. Surge responses — Volume-driven improvisation
   During peak periods, shortcuts are introduced. Over time, they become the process, without ever being validated.
7. No baseline — Measurement blindness
   No measurement of cycle time, error rates, or cost per transaction. Without a baseline, there is no pressure to improve.

Discovering the Hidden Waste: Discover, Analyze, Eliminate

Process improvement is a disciplined methodology, not a workshop exercise or opinion survey.

Phase 1 — Discover: Make the Invisible Visible

Most organizations believe they know their processes. They have flowcharts and SOPs.

But these describe the intended process, not the actual one.

Discovery closes that gap.

Effective techniques include:

• Process walkthroughs with practitioners (not managers)
• Time-and-motion observation across transactions
• System log extraction from ERP, ATS, and workflow tools
• Voice of the customer interviews
• Volume and exception analysis

Phase 2 — Analyze: Quantify and Prioritize

Discovery shows what is. Analysis shows what it costs, and what it should be.

Without quantification, everything feels important, and nothing gets fixed.

Frameworks include:

• Value stream mapping
• Root cause analysis (5-Why, fishbone)
• Failure mode and effects analysis (FMEA)
• Benchmarking against industry standards

Phase 3 — Eliminate: Redesign Before You Automate

This is where organizations go wrong, automating before redesigning.

The principle is simple:
Automating a bad process does not fix it, it makes the badness faster.

For every step, ask:

• Can it be removed entirely?
• Can it be combined?
• Can it be simplified?

Only after this is a process truly ready for AI.

Process Mining and AI: Your Discovery Superpower

Historically, process discovery was slow, expensive, and subjective.

Today, process mining, combined with AI, has transformed this into a continuous, data-driven capability.

What Process Mining Does

It extracts event logs from ERP, ATS, CRM, and workflow systems, and reconstructs the actual process.

Not a tidy flowchart, but reality.

Where AI Enhances Process Mining

• Anomaly detection
• Predictive monitoring
• Natural language querying
• Intelligent benchmarking
• Automated root cause analysis

The Compound Dividend

Organizations that combine process improvement with continuous mining create a virtuous cycle, the system improves as it runs.

Shared Ops as the Structural Multiplier

The same process often runs in multiple ways across business units, creating variation, inconsistency, and hidden inefficiencies. Shared operations bring standardization, control, and scale, making processes consistent and AI-ready.

PREP before you RACE

You are aware of the RACE framework, but before you RACE, you need to PREP.

PREP is not a gate to slow AI adoption.
It is what ensures AI adoption succeeds.

THE PREP FRAMEWORK

• P — Process Readiness Assessment
  Map, measure, and score processes. Identify waste, duplication, and data gaps before AI is introduced.
• R — Rationalize & Standardize
  Eliminate redundant variants. Standardize inputs, outputs, naming conventions, and data formats.
• E — Enable Clean Data Foundations
  Audit data quality, completeness, and lineage. Establish governance and integration hygiene.
• P — Position for Continuous Improvement
  Embed process mining, feedback loops, and performance dashboards for ongoing optimization.

Running PREP in Practice

PREP is not a one-time audit.

It is a readiness state, revisited whenever:

• New processes are considered for AI
• Mergers or regulatory changes occur
• Systems are upgraded or replaced

A Note for Leadership Teams

PREP is not a technology programme.

It requires active sponsorship from:

• Finance
• Operations
• HR

The CTO can enable, but process owners must lead.

AI transformation owned only by IT is another route to a faster broken process.

The Race Awaits, But So Does the Warm-Up

The excitement around AI is justified.

But the organizations pulling ahead share one trait: they prepared.

They:

• Mapped before automating
• Cleaned data before training
• Standardized before scaling
• Built shared operational foundations before deploying AI

The sprinter who wins is not the one who trained hardest in the final week, but the one who prepared consistently.

Your AI investment deserves the same discipline.

PREP is your warm-up. Run it without compromise, and then go win the RACE.

If you’re exploring AI within your operations, start with readiness.

Connect with our Business Transformation Services team to assess where PREP can unlock measurable value in your organization, before you accelerate with AI.

Share this post with your network.

Authors

Mandar Vanarse

Chief Technology Officer, QX Global Group

Know More chevron_forward

Amit Simon

Sr. VP, Business Excellence

Know More chevron_forward