Multi-Supplier Delivery with Flight Levels

When three suppliers work on one programme, you do not have three delivery problems. You have one coordination problem wearing three badges. The instinct is to fix it by installing a scaling framework on everyone. Resist that instinct. Build a coordination system above the teams instead, and leave the teams their own ways of working. That is what Flight Levels does, and it is the most useful model I have found for multi-supplier delivery.

The Multi-supplier Mess, Briefly

I have spent years in delivery environments where our consultants, the client’s own engineers, and other vendors all touch the same value stream. At a large UK retailer that meant onboarding third-party supplier data into the data lake, validating it with commercial managers, and coordinating platform, Paved Road and neighbouring data teams through to visualisation. Each party arrived with its own method, its own board, its own definition of done.

Every team can be locally healthy while the programme is globally stuck, because work does not fail inside teams. It fails between them: at handoffs, dependencies, and priority calls nobody owns.

Why Framework-first Fails

The standard move is to roll out a scaling framework so everyone works the same way. The record is not kind to this. One Flight Levels Academy case study describes a 3,000-person organisation two years into a SAFe transformation: no cycle time improvement, no quality improvement, and employee satisfaction falling.¹ Scaled Agile’s own 2025 survey offers the other side (68% report improved employee satisfaction)², so read both. But my experience with suppliers is specific: you often cannot impose a framework on people from three companies with three contracts, and when you try, you get compliance theatre. Everyone renames their meetings while the handoffs stay broken.

Flight Levels in Plain Words

Flight Levels, from Klaus Leopold, is a thinking model, not a framework to install. It says work happens at three altitudes:

• Level 1: the teams. Where work gets built. Scrum here, Kanban there. Fine.
• Level 2: end-to-end coordination. Where work flows across teams and suppliers, from idea to live. This level is usually invisible, and it is where multi-supplier delivery dies.
• Level 3: strategy. Where the organisation decides what matters, so that Level 2 has something honest to prioritise against.

The insight that changes behaviour: nobody has to change how their team works. You add a thin layer that makes cross-team flow visible and gives it an owner. The Academy’s recent framing calls it an operating system that integrates whatever methods already coexist, rather than another method competing with them.³

The Field Guide

Five steps, in the order I would run them.

1. Draw the real flow. One wall, physical or virtual: idea to live, across every supplier. Not the org chart, the work. Expect arguments about what the stages even are. Those arguments are the diagnosis.

2. Put every supplier’s in-flight work on it. One board, every party, no exceptions. The first version will be ugly and incomplete. Ship it anyway; visibility creates its own pressure to be accurate.
3. Stand up one coordination cadence. A short, regular meeting at Level 2 with someone empowered from each supplier. One agenda: what is blocked between us, what is ageing, what needs a priority call. It replaces the escalation calls; it does not add to them.
4. Limit work in progress at the system level. The hardest step politically, because every supplier is separately incentivised to look busy. Cap what the system works on and finish before starting. This is the step where cycle times fall.
5. Connect it upward. A Level 2 board with no strategic input becomes a complaint wall. Someone must regularly say what matters most this quarter, so the coordination layer prioritises rather than debates.

Contracts are the quiet enabler. If each supplier is measured only on their own throughput, Level 2 fights the incentives. Get shared flow measures (end-to-end cycle time, blocked-time between parties) into the working agreements early.

Does Flight Levels Work for Multi-Supplier Delivery?

Beyond my own experience: the SAFe-rescue organisation above built its Flight Level 2 coordination system in about five weeks, with no new roles.⁴ At Axel Springer, two publishing units report product development accelerating threefold with Flight Levels.⁵ Weeks, not quarters, because you are adding a layer, not retraining thousands of people. Both case studies are published by the Flight Levels Academy itself, so apply the same discount you would to any vendor’s highlight reel; my own experience is the reason I trust the mechanism.

What it will not fix: a supplier that will not share visibility, a client who will not prioritise, or a contract that rewards local busyness. Flight Levels makes those failures visible fast. The conversations remain yours to have.

Takeaway: stop trying to standardise your suppliers and start connecting them. One board across the flow, one cadence, one WIP cap, one line to strategy. The teams keep their methods. The programme finally gets one.

1. Case Study: Agile at Scale with Flight Levels, Flight Levels Academy, February 2026. ↩︎
2. State of SAFe Report 2025, Scaled Agile Inc., 2025. ↩︎
3. Flight Levels: The Operating System for Your Organization, Flight Levels Academy, April 2025. ↩︎
4. Case Study: Agile at Scale with Flight Levels, Flight Levels Academy, February 2026. ↩︎
5. From Slow To Flow: Axel Springer’s COMPUTERBILD & AUTOBILD, Flight Levels Academy, 2025. ↩︎