Skip to content
Ashish Kumar AgnihotriOperating Partner

The discipline, explained

Operational Excellence Consultant: What the Best Ones Actually Do

Operational excellence has been diluted into a poster word, which makes the consultants who sell it hard to judge. Underneath the slogan sits a real discipline — flow, variation, standard work, governance — and a real difference between practitioners who move numbers and presenters who laminate frameworks. This guide explains the discipline, the actual work, and how to tell the two apart.

Operational excellence is not a mindset, a culture programme or a wall of values. It is an operating condition: work flows without heroics, variation is understood and controlled, the standard way is the easiest way, and a governance rhythm catches drift before customers do. Toyota’s influence gave the discipline its vocabulary and most of its tools, but the underlying logic is older and simpler — study the work, stabilise it, standardise it, improve it, and hold the gains with governance. Any organisation that delivers repeatedly — a plant, a newsroom, a claims team, an agency network — is running some version of that loop, well or badly.

The consulting market around this discipline is noisy. Operational excellence consultant covers everyone from veterans who have stabilised billion-dollar operations to trainers reselling belt courses, and the language — Lean, Six Sigma, kaizen, agile operations — has been marketed far past its meaning. The vetting problem is real because the certificates look identical while the capabilities differ by an order of magnitude. What separates them is never the framework; it is whether the consultant can walk into your specific operation, find the constraint, put a number on it, and leave behind a system that holds after they stop being paid. This guide is built around that single test.

I spent nineteen years inside this discipline, most recently as Senior Director, Business Excellence at Publicis Groupe — running quality and operational standards across 500+ clients, 2,000+ teams and more than USD 750 million in annual media spend, for brands including Disney, Samsung, Adobe and P&G. Lean Six Sigma Green Belt; more usefully, measured results: a quality score lifted from 95% to 99% across 2,000+ campaigns, a billing cycle cut from roughly two months to fifteen days across 75 entities. I now do this work as an embedded fractional COO. What follows is the practitioner’s view of what the job is — and how to buy it well.

In depth

What you need to know.

Operational excellence, defined without the poster

Strip the branding and operational excellence is four conditions holding at once. Work flows: orders, projects or cases move through the operation without queues silently forming and heroics silently compensating. Variation is controlled: the operation knows its normal range, so it reacts to signals rather than noise. Standard work exists and is followed: the best-known way of doing each critical task is defined, and reality matches the definition — the deepest test, because most companies own process documents that describe a fiction. And governance holds the gains: a review rhythm that spots drift early and closes decisions fast. When all four hold, quality stops depending on individual brilliance, and the operation delivers on a bad week roughly what it delivers on a good one. That — not the banners — is the product being bought.

Flow: the first discipline, and the least practised

Flow asks one question: how long does a unit of work take to cross the operation, and where does it wait? In most service businesses the answer is embarrassing — total elapsed time is dominated by queues, approvals and handoffs, not by work. The excellence consultant’s first real act is making that visible: walking a real order from sale to delivery to cash, timestamping every stop, and separating touch time from wait time. The findings are usually undramatic and expensive — a two-day task inside a three-week elapsed cycle, approvals that add days and no judgement, rework loops nobody logged. Improving flow rarely requires new software; it requires removing steps, moving decisions closer to the work and capping how much unfinished work the system carries at once. Companies feel this as speed. Finance feels it as cash.

Variation and standard work: making quality a system property

Variation is the enemy of promised outcomes: the same process producing a three-day turnaround on Monday and a nine-day one on Thursday is not a process, it is a lottery with an average. The discipline is to measure the spread, find its causes — inputs, skills, workload, definitions — and narrow it until the operation’s promise is honest. Standard work is the instrument: the current best way of doing a critical task, written by the people who do it, visible where the work happens, and improved on evidence rather than opinion. This is where my own quality work at global scale lived — moving a delivery-quality score from 95% to 99% across 2,000+ campaigns was not a slogan campaign; it was standards, checks at the point of work, and variance chased case by case until the causes ran out.

Governance: the discipline that keeps the other three alive

Every operation that improved and slid back is missing the fourth discipline. Governance is the operating rhythm that holds gains: a weekly review where the few numbers that matter are examined against thresholds, deviations get owners and dates, and decisions close instead of rolling forward. It sounds bureaucratic and is the opposite — a disciplined hour that replaces the standing meetings where everyone reports and nothing moves. The consultant’s job is to install it: the scorecard with agreed definitions, the thresholds that trigger action, the escalation path, the decision log. Then to hand over the chairing of it. An operational excellence engagement that ends without a governance rhythm running in the client’s own hands has produced a project, not a capability — and projects decay. This is the difference between an improvement and an operating model.

Lean, Six Sigma and the toolkits — used, not worshipped

The toolkits earn their keep when they serve a specific problem. Value-stream mapping for flow. Statistical thinking and root-cause discipline for variation — the durable core of Six Sigma. 5S and visual management where the workplace itself hides problems. Kaizen where frontline improvement needs a legitimate channel. What corrupts them is method religion: deploying the full liturgy because the manual says so, measuring belts trained instead of numbers moved, translating a manufacturing vocabulary word-for-word onto service work it fits badly. My Green Belt taught me grammar; scale taught me editing. The best operational excellence consultants treat the toolbox the way a good editor treats style guides — mastered thoroughly, applied selectively, never mistaken for the point. If a proposal leads with the methodology rather than your constraint, the methodology is the product being sold.

Certificates versus measured results

Belts and certifications certify training, not capability. A Green Belt or Black Belt means someone studied the method and ran a supervised project; it says nothing about whether they can find the constraint in your operation under real conditions and move it. So weigh evidence in this order: numbers moved, with baselines and definitions — a cycle time cut, a quality score raised, a cost of rework reduced; scale context — where they did it and what was at stake; references who will take a call; then certificates, as confirmation of vocabulary. Be suspicious of biographies that list credentials but no baselines. The question that sorts this market in one sentence: show me two numbers you moved, and tell me how you would find the third one here. Anyone worth hiring enjoys that question.

How to vet an operational excellence consultant

Run three tests before price enters the conversation. The definition test: ask them to define, for your specific product or service, what the quality standard should be and how they would measure it this month — vague answers here predict vague engagements. The week-one test: ask what they would do in the first week; the best answers involve walking real work end to end and pulling real data, not scheduling workshops. The transfer test: ask what your team runs without them in six months, and which artefacts — standards, scorecard, governance cadence — they leave behind. Then match altitude: someone who has only optimised single teams will drown in a multi-entity operation, and someone who has only governed may be too far from the floor for a fifty-person company. Specific beats certified, every single time.

What a serious engagement looks like, start to finish

The honest arc runs diagnose, stabilise, standardise, govern, transfer. Diagnosis: two to four weeks walking the flow, measuring the leaks, and returning a written read with baselines — priced as a fixed fee, so you can judge the fit on evidence. Stabilise: stop the bleeding at the constraint — the rework loop, the approval queue, the overloaded step. Standardise: define standard work with the people who do it, and wire checks where errors are born, not where they are caught. Govern: install the scorecard and the weekly rhythm, with thresholds and owners. Transfer: your leaders chairing the cadence, your team owning the standards, the consultant progressively unnecessary. Timescales vary with scope; the sequence should not. Anyone proposing to start with a training rollout or a tool purchase has reversed the order — and priced the reversal.

Questions

Common questions.

They make delivery reliable. Concretely: walk your flow from sale to delivery to cash and measure where work waits; quantify variation so the operation reacts to signals, not noise; define standard work with the people who do it; wire quality checks where errors are born; and install a governance rhythm — scorecard, thresholds, owners — that holds the gains. The deliverable is not a report; it is an operating condition your team can run without them. If a proposal reads as training plus workshops plus a maturity assessment, you are buying awareness, not excellence.

It is when good outcomes are the default rather than the achievement. Work flows without heroics; the operation knows its normal range and spots the abnormal early; the standard way of doing critical tasks is defined, followed and improved; and a steady review rhythm catches drift before customers feel it. A simple test: does your operation deliver roughly the same quality in a bad week as in a good one? If yes, you have it. If quality depends on which person handled the job and how their week went, you have talent — not a system.

Weigh evidence over vocabulary. The best operational excellence consultants show numbers they moved — baselines, definitions, results — at a scale comparable to yours; they can define a measurable quality standard for your product in the first conversation; their week-one plan involves real work and real data rather than workshops; and they can describe exactly what your team will run without them afterwards. Certificates confirm training and belong at the bottom of the list. One more filter: ask what would make them advise stopping the engagement. Practitioners have an answer; vendors do not.

As vocabulary, yes; as proof, no. A belt certifies that someone learned a method and ran a supervised project — useful grammar, shared language, disciplined root-cause habits. I hold a Green Belt and use its thinking weekly. But certification says nothing about judgement under real constraints: which problems to ignore, which tools to skip, when the manufacturing vocabulary fits service work and when it insults it. Hire on baselines moved and scale handled; treat belts the way you treat a driving licence — necessary evidence of training, and no evidence at all of where someone can take you.

Named numbers, moved and held. The pattern from my own operating work: a quality score from 95% to 99% across 2,000+ campaigns and 450 clients; a billing approval cycle from roughly two months to fifteen days across 75 entities; a newsroom’s output quadrupled to roughly 400 stories a day with two added desks. Your equivalents will differ — cycle time, first-pass yield, rework cost, on-time delivery — but the grammar is fixed: baseline, agreed definition, result, and a governance rhythm that keeps it true a year later. Agree those measures before signing anything.

A project improves a process and ends; operational excellence changes how the operation runs and persists. The project inherits the company’s old habits — no flow visibility, no standard work, no governance — so its gains decay the month attention moves elsewhere, which is why so many companies remember three improvement programmes and no improvement. Excellence work installs the habits themselves: visible flow, controlled variation, living standards, a review rhythm with teeth. Projects are sometimes the right buy for bounded problems; just never confuse a completed project with a changed operation.

Honest sequencing beats honest-sounding deadlines. A diagnostic that produces baselines takes two to four weeks. Stabilising a real constraint shows measurable movement within a quarter — if nothing has moved in ninety days, question the engagement. Standardising and wiring governance into leadership habit takes two or three quarters of steady cadence. Deeper change — where improvement runs without any external push — is a year or more, and anyone promising it faster is selling posters. The protection is structure: fixed-fee diagnosis first, measures agreed upfront, and an engagement designed to transfer rather than renew.

Scope and seat. An operational excellence consultant typically improves defined parts of the operation — a flow, a quality system, a governance layer — and may work from outside the accountability line. A fractional COO takes the operating seat itself: owns the whole system, runs the leadership cadence, answers to the board, at partial time. The disciplines overlap heavily; the difference is mandate. If the operation needs its excellence layer rebuilt, hire the discipline. If it needs an owner for the entire operating system, hire the seat — the excellence layer comes installed.

The next step

A short conversation settles most of this — and a fixed-fee diagnostic settles the rest.

Book a consultationThe diagnostic

Keep exploring

Process transformation & Lean Six SigmaThe service where this discipline gets installed.Business excellence & quality auditsThe quality layer, audited and rebuilt.Defining the quality standardThe framework behind the definition test.Operations audit checklistThe eight areas to examine before hiring anyone.Business excellence awards in IndiaCII-EXIM, EFQM and what actually builds capability.The fixed-fee diagnosticEvidence first, engagement second.