Manufacturing operations improvement with our fee tied to the success
The eleven steps of a Sprint, in full.
- 01 — A day spent on your plant floor (Gemba, Grounding). What we do: Gemba means the “actual place”. Day one is spent on your floor and not in the conference room. We walk the plant like it actually runs on a normal day and talk to your team running it. What you get: A grounded picture of the plant as it operates, with different perspectives captured. We review the numbers but challenge at the root. Grounded in: Genchi genbutsu — Toyota Production System.
- 02 — Capture the customer perspective (Customer view, Start). What we do: We begin at the shipping dock and walk against the flow. Whatever your customer is unhappy about is visible there first — controlled shipping, a sort, a containment nobody has managed to exit. What you get: The escalation state written down the way your customer sees it, before anyone internal reframes it. Then we derive together a target state for de-escalation criteria. Grounded in: Rother & Shook — Learning to See.
- 03 — Follow the process backwards (Material flow, Trace). What we do: One continuous pass upstream, station by station, recording cycle time as we go. Departments own stations; nobody owns what happens between them, and that gap is where most of the loss lives. What you get: Cycle time at every station, measured against takt. We identify and document bottlenecks and an initial direction for the highest-value fixes. Grounded in: Value stream mapping.
- 04 — Every number gets a provenance (Data review, Qualify). What we do: We review your plant’s data integrity and grade it on coverage, quality and freshness. The world-class benchmark is real-time data from source systems. What you get: A comprehensive data maturity assessment benchmarked against world class and grounded in best-in-class manufacturing practices and standards. Grounded in: Measurement system discipline.
- 05 — Identify the constraint (Constraint, Bottleneck). What we do: One station caps the throughput of the whole plant. We confirm which one with cycle time against takt. Then we express the cap the way operations sees it: in units per hour, not percentages. What you get: The constraint named with its cycle time, and the improvement list re-ordered around it — what to exploit, what to subordinate, and which projects move nothing until the constraint does. Grounded in: Goldratt — Theory of Constraints.
- 06 — Plan working capital improvement (Cash in WIP, Quantify). What we do: Everything the plant produces faster than the constraint can absorb becomes inventory instead of shipments. We count what is standing in front of each station and convert it: bays into days of flow time, days into working capital. What you get: A priced queue: where the cash is parked, how much of it, and what it releases once the constraint moves. Grounded in: Little’s Law — Hopp & Spearman, Factory Physics.
- 07 — Rank by value at stake (Actions, Prioritise). What we do: Every abnormality observed on the walk goes into one list, ordered by what it is costing — not by department, not by who raised it. Items we cannot honestly price stay marked unpriced. What you get: A ranked backlog translated into a Kanban board: what to fix first, and what it is worth. Each action gets a named owner and a due date. This is where improvement discipline starts. Grounded in: Cost of delay — one-way sensitivity.
- 08 — Establish governance for traction (Cadences, Rhythm). What we do: Review of daily meetings at the station, weekly management meetings, and monthly plant meetings. We focus on reducing coordination friction and establish aligned views on the constraint. What you get: A best-in-class tiered management rhythm with clear accountability, effective coordination and a standing agenda per meeting. Grounded in: Leader standard work.
- 09 — Escalate for effective support (Resolution, Support). What we do: What the team at the station cannot close in a shift escalates for support, not for blame. Kanban actions needing capital, headcount, or another department’s input get elevated. We moderate the process to help your team build confidence. What you get: An effective escalation procedure the floor can apply without asking permission. We shift from blame to joint resolution — the core of effective problem solving and coordination. Grounded in: Tiered daily management.
- 10 — Prove it on the floor (Prove it, Verify). What we do: We re-walk the same route. Every outcome we claim corresponds to a physical object in a changed state — a placard down, an inspection booth deleted, a buffer flattened, a standard-work board filled in. What you get: Critical issues addressed at the root, each with a before and after number attached. A team that confidently owns the process, and standards and systems re-established. Grounded in: Solve the problem. Then automate..
- 11 — Exit on exit criteria attainment (Exit gate, Approved). What we do: The Sprint closes against the de-escalation criteria agreed with your customer at the start: the fix proven at the line, the standard re-established, your team running the cadence. Your customer confirms the exit, we don’t declare it. What you get: Handover on evidence: exit criteria met, cadence running, customer on board. Our fee tied to the success. Grounded in: TWI job instruction — sustainment practice.
EngineeringProcess capability
FoundationsSystems, data, connectivity
ComplianceConformance to standard
RigourCost & cash discipline
ReliabilityEquipment uptime
Supply ChainMaterial availability
TrainingPeople capability
The path from escalation to fixed.
Gemba means the actual place.
Day one, we’re standing on your plant floor.
Has throughput actually moved, is the gain in the P&L, and where’s the next constraint?
Is the constraint still the constraint — and is the countermeasure holding across all three shifts?
Did we hit rate this shift — and if not, what stopped us in the last four hours?
The walk, in eleven beats.
- The actual place. Gemba means the actual place. Day one, we’re standing on your plant floor.
- Start at the dock. Walk it backwards. The customer’s view is the only true north.
- Follow the part. Nobody owns the flow. So follow the part.
- What the board doesn’t say. A dashboard shows what’s measured. A walk shows what isn’t.
- The WIP is pointing. You don’t hunt the bottleneck. The inventory is already pointing at it.
- The constraint. One station sets the pace of the whole plant. Everything else is noise.
- The walk becomes the backlog. You don’t get a report. You get a backlog ranked by what it’s costing.
- Three cadences. Finding it is diagnosis. Solving it every shift is the discipline.
- What rises. Every abnormality gets an owner. Unresolved at the cell, it escalates the same day.
- What changed on the floor. The same route, after turnaround. Five fixes running, each one measured.
- The exit gate. No handover without proof. Criteria met, cadence running, customer aligned.
Manufacturing Operations Improvement when it matters
The customer manages you by exception and every shipment is a conversation.
De-escalated — and the customer says so
Controlled shipping exited against agreed criteria, with the evidence pack behind it.
The date won't move and the rate isn't there.
Run at rate, achieved and bought off
Capacity demonstrated against the customer's buy-off criteria, not an internal estimate.
Nominal capacity says one number, the floor delivers another.
More cycles, and more of them good
Cycle count and first-pass yield moved together — good parts out, counted on the line.
Scrap and sorting are buried inside labour hours.
The extra inspection gate, removed
GP-12 or equivalent exited because the escape was designed out — with the scrap rate to prove it.
The person who held it together has left.
The method owned by the team, not a person
The operation stops depending on one individual — the outcome here is continuity, not a number.
Someone who has run the plant, not just studied it
Sprint is delivered by operators who led manufacturing operations, launched programs and ran turnaround projects at scale. You work with senior manufacturing experts that lived your pain before.
Get in touch
The buffer is gone
Manufacturing is not as predictable as it used to be. Absorbing demand fluctuation used to run on inventory, overtime and schedule slack — and those are spent. A problem on the line now reaches the customer the same week it starts.
+62% since 2019
Downtime costs rose 62% while the number of incidents fell. The same problems cost far more than they used to — for the world's largest manufacturers, 11% of revenue, up from 8%.
Siemens, The True Cost of Downtime 2024
The cost is running now.
Every week the problem stays open is a week it's charged to the plant.
